Direction of Innovation & Labor Markets

We develop a parsimonious model of innovation to confront firm‐level evidence. It captures the dynamics of individual heterogeneous firms, describes the behavior of an industry with firm entry and exit, and delivers a general equilibrium model of technological change. While unifying the theoretical analysis of firms, industries, and the aggregate economy, the model yields insights into empirical work on innovating firms. It accounts for the persistence of firms’ R&D investment, the concentration of R&D among incumbents, the link between R&D and patenting, and why R&D as a fraction of revenues is positively correlated with firm productivity but not with firm size or growth.

We build a model of firm-level innovation, productivity growth, and reallocation featuring endogenous entry and exit. A new and central economic force is the selection between high- and low-type firms, which differ in terms of their innovative capacity. We estimate the parameters of the model using US Census microdata on firm-level output, R&D, and patenting. The model provides a good fit to the dynamics of firm entry and exit, output, and R&D. Taxing the continued operation of incumbents can lead to sizable gains (of the order of 1.4 percent improvement in welfare) by encouraging exit of less productive firms and freeing up skilled labor to be used for R&D by high-type incumbents. Subsidies to the R&D of incumbents do not achieve this objective because they encourage the survival and expansion of low-type firms. (JEL D21, D24, H25, L52, O31, O34)

This article studies the determinants of college major choice using an experimentally generated panel of beliefs, obtained by providing students with information on the true population distribution of various major-specific characteristics. Students logically revise their beliefs in response to the information, and their subjective beliefs about future major choice are associated with beliefs about their own earnings and ability. We estimate a rich model of college major choice using the panel of beliefs data. While expected earnings and perceived ability are a significant determinant of major choice, heterogeneous tastes are the dominant factor in the choice of major. Analyses that ignore the correlation in tastes with earnings expectations inflate the role of earnings in college major choices. We conclude by computing the welfare gains from the information experiment and find positive average welfare gains.

Productivity differences across firms are large and persistent, but the evidence for worker reallocation as an important source of aggregate productivity growth is mixed. The purpose of this paper is to estimate the structure of an equilibrium model of growth through innovation designed to identify and quantify the role of resource reallocation in the growth process. The model is a version of the Schumpeterian theory of firm evolution and growth developed by Klette and Kortum (2004) extended to allow for firm heterogeneity. The data set is a panel of Danish firms that includes information on value added, employment, and wages. The model's fit is good. The estimated model implies that more productive firms in each cohort grow faster and consequently crowd out less productive firms in steady state. This selection effect accounts for 53% of aggregate growth in the estimated version of the model.

This paper provides a quantitative theory for the recent rise in residual wage inequality consistent with the empirical observation that a sizable part of this increase has a transitory nature, a feature that eludes standard models based on ex ante heterogeneity in ability. An acceleration in the rate of quality improvement of equipment, like the one observed from the early 1970s, increases the productivity/quality differentials across machines (jobs). In a frictional labor market, this force translates into higher wage dispersion even among ex ante equal workers. With vintage-human capital, the acceleration reduces workers' capacity to transfer skills from old to new machines, generating a rise in the cross-sectional variance of skills, and therefore of wages. Through calibration, the paper shows that this mechanism can account for 30 percent of the surge in residual inequality in the U. S. economy (or for most of its transitory component). Two key implications of the theory—faster within-job wage growth and larger wage losses upon displacement—find empirical support in the data.

Markups vary systematically across firms and are a source of misallocation. This paper develops a tractable model of firm dynamics where firms' market power is endogenous and the distribution of markups emerges as an equilibrium outcome. Monopoly power is the result of a process of forward‐looking, risky accumulation: firms invest in productivity growth to increase markups in their existing products but are stochastically replaced by more efficient competitors. Creative destruction therefore has pro‐competitive effects because faster churn gives firms less time to accumulate market power. In an application to firm‐level data from Indonesia, the model predicts that, relative to the United States, misallocation is more severe and firms are substantially smaller. To explain these patterns, the model suggests an important role for frictions that prevent existing firms from entering new markets. Differences in entry costs for new firms are less important.

We develop a model of repeated product improvements in a continuum of sectors. Each product follows a stochastic progression up a quality ladder.

How do innovation policies affect the direction of research? Is market-based innovation too radical or too incremental? We construct a novel and tractable model of the direction of innovation. Firms pursue inefficient research directions because they race to discover easy yet less valuable projects and because they work on difficult inventions where they can appropriate a larger portion of the social value. Fixing these inefficiencies requires policy to condition on properties of inventions that could have been discovered but were not. Policies which do not do so, like patents and prizes, may fail to encourage firms to research in the efficient direction, even if they obtain the optimal quantity of R&D.

We document that nearly half of the global decline in agricultural employment was driven by new cohorts entering the labor market. A new dataset of policy reforms supports an interpretation of these cohort effects as human capital. Using a model of frictional labor reallocation, we conclude that human capital growth led to a sharp decline in the agricultural labor supply, accounting, at fixed prices, for 40 percent of the decrease in agricultural employment. This aggregate effect is halved in general equilibrium and it reflects the role of human capital as both a mediating factor and an independent driver of labor reallocation. (JEL J22, J24, J43, L16, O13, O14, Q10)

How are resources allocated across different R&D areas (i.e., problems to be solved)? As a result of dynamic congestion externalities, the competitive market allocates excessive resources into those of high return, being those with higher private (and social) payoffs. Good problems are tackled too soon, and as a result the distribution of open research problems in the socially optimal solution stochastically dominates that of the competitive equilibrium. A severe form of rent dissipation occurs in the latter, where the total value of R&D activity equals the value of allocating all resources to the least valuable problem solved. Resulting losses can be substantial.

Science, Technology, Engineering, and Math (STEM) jobs are a key contributor to economic growth and national competitiveness. Yet STEM workers are perceived to be in short supply. This paper shows that the “STEM shortage” phenomenon is explained by technological change, which introduces new job tasks and makes old ones obsolete. We find that the initially high economic return to applied STEM degrees declines by more than 50 percent in the first decade of working life. This coincides with a rapid exit of college graduates from STEM occupations. Using detailed job vacancy data, we show that STEM jobs changed especially quickly over the last decade, leading to flatter age-earnings profiles as the skills of older cohorts became obsolete. Our findings highlight the importance of technology-specific skills in explaining life-cycle returns to education, and show that STEM jobs are the leading edge of technology diffusion in the labor market.

Using administrative data from the United States, we document novel stylized facts regarding technological innovation and the riskiness of labor income. Higher rates of industry innovation are associated with significant increases in labor earnings for top workers. Decomposing this result, we find that own firm innovation is associated with a modest increase in the mean, but also variance, of worker earnings growth. Innovation by competing firms is related to lower, and more negatively skewed, future earnings. We construct a structural model featuring creative destruction and displacement of human capital that replicates these patterns. In the model, higher rates of innovation by competing firms increases the likelihood that both the worker and the incumbent producer are displaced. By contrast, a higher rate of innovation by the worker's own firm increases profits, but is a mixed blessing for workers, as it increases odds that the skilled worker is no longer a good match to the new technology. Estimating the parameters of the model using indirect inference, we find significant welfare losses and hedging demand against innovation shocks. Consistent with our model, we find that these left tail effects are more pronounced for process improvements, novel innovations, and are concentrated in movers rather than continuing workers.

Abstract This paper examines how college students in the United States altered their college majors during the energy boom and bust of the 1970s and 1980s. We focus on petroleum engineering and geology, two majors closely related to the energy industry. We find strong evidence that the energy boom increased the prevalence of these two energy-related majors and the energy bust lowered the prevalence of these majors. Effects are particularly strong for young people born in energy intensive states. Thus, college major decisions responded to industry fluctuations with important location-specific effects consistent with frictions to migration and information flows.

Do economies adjust slowly to certain technological innovations and more rapidly to others? We argue that the adjustment is slower when innovations mainly benefit production activities requiring skills that are more different from those used in the rest of the economy. When such skill specificity is stronger, the adjustment of labor markets is driven less by the fast reallocation of older incumbent workers and more by the gradual entry of younger generations. We first document that the US labor market adjusted differently to early twentieth-century manufacturing innovations than to recent information and communication technologies (ICTs). We then build an overlapping-generations model of technological transitions and characterize how skill specificity affects equilibrium dynamics. Skill specificity helps explain why the ICT transition was slower, driven entirely by the entry of younger generations.

Abstract We present a model where firms conduct R&D in both a safe and a risky direction. As patentability standards rise, an innovation in the risky direction is less likely to receive a patent, which decreases the static incentive for new entrants to conduct risky R&D but can increase their dynamic incentive. These, together with a strategic substitution and a market structure effect, result in an inverted‐U shape in the risky direction but a U shape in the safe direction for the relationship between R&D intensity and patentability standards. R&D is biased toward (against) the risky direction under lower (higher) standards.

We study the cross-sector allocation of R&D resources in a multisector growth model with an innovation network, where one sector's past innovations may benefit other sectors' future innovations.Theoretically, we solve for the optimal allocation of R&D resources.We show a planner valuing long-term growth should allocate more R&D toward central sectors in the innovation network, but the incentive is muted in open economies that benefit more from foreign knowledge spillovers.We derive sufficient statistics for evaluating the welfare gains from improving R&D allocation.Empirically, we build the global innovation network based on patent citations and establish its empirical importance for knowledge spillovers.We evaluate R&D allocative efficiency across countries using model-based sufficient statistics.Japan has the highest allocative efficiency among the advanced economies.For the U.S., improving R&D allocative efficiency to Japan's level could generate more than 19.6% welfare gains.

To study the drivers of the employment reallocation across sectors and occupations between 1960 and 2010 in the US we propose a model where technology evolves at the sector-occupation cell level. This framework allows us to quantify the bias of technology across sectors and across occupations. We implement a novel method to extract changes in sector-occupation cell productivities from the data. Using a factor model we find that occupation and sector factors jointly explain 74-87 percent of cell productivity changes, with the occupation component being by far the most important. While in our general equilibrium model both factors imply similar reallocations of labor across sectors and occupations, quantitatively the bias in technological change across occupations is much more important than the bias across sectors.

Using detailed data on skill requirements in online vacancies, we estimate the demand for AI specialists across occupations, sectors, and firms. We document a dramatic increase in the demand for AI skills over 2010–2019 in the U.S. economy across most industries and occupations. The demand is highest in IT occupations, followed by architecture and engineering, scientific, and management occupations. Firms with larger market capitalization, higher cash holdings, and higher investments in R&D have a higher demand for AI skills. We also document a wage premium of 11% for job postings that require AI skills within the same firm and 5% within the same job title. Managerial occupations have the highest wage premium for AI skills. Firms demanding AI skills more intensively also offer higher salaries in non-AI jobs.

We argue that college students' field-of-study choices significantly influence how economies respond to labor market disruptions.To do so, we develop and estimate a framework featuring forward-looking students who choose a field of study when entering college, and subsequently make decisions over occupations after graduating and entering the labor market.Different fields endow workers with distinct comparative advantages and varying costs associated with switching occupations.Simulating both a trade war and wide scale adoption of AI, we use our model to make three points.First, relative to models that ignore how new cohorts adjust their field-of-study choices, our framework predicts larger aggregate income responses and greater distributional differences.Second, policies that enhance flexibility in field-of-study decisions-such as relaxing capacity constraints in high-demand programs-raise aggregate output.Finally, these policies also lessen the adverse distributional consequences of shocks, by affording more opportunities to students with lower earnings potential.

Innovation is a key driver of long-run economic growth. Studying innovation requires a clear view of the characteristics and behavior of the individuals who create new ideas. A general lack of rich, large-scale data has constrained such analyses. We address this by introducing a new dataset linking patent inventors to survey, census, and administrative microdata at the U.S. Census Bureau. We use this data to provide a first look at the demographic characteristics, employer characteristics, earnings, and employment dynamics of inventors. These linkages, which will be available to researchers with approved access, dramatically increase the scope of what can be learned about inventors and innovative activity.

Elite economics PhD programs aim to train graduate students for a lifetime of academic research. This paper asks how advising affects graduate students' post-PhD research productivity. Advising is highly concentrated: at the eight highly-selective schools in our study, a minority of advisors do most of the advising work. We quantify advisor attributes such as an advisor's own research output and aspects of the advising relationship like coauthoring and research field affinity that might contribute to student research success. Students advised by research-active, prolific advisors tend to publish more, while coauthoring has no effect. Student-advisor research affinity also predicts student success. But a school-level aggregate production function provides much weaker evidence of causal effects, suggesting that successful advisors attract students likely to succeed-without necessarily boosting their students' chances of success. Evidence for causal effects is strongest for a measure of advisors' own research output. Aggregate student research output appears to scale linearly with graduate student enrollment, with no evidence of negative class-size effects. An analysis of gender differences in research output shows male and female graduate students to be equally productive in the first few years post-PhD, but female productivity peaks sooner than male productivity.

We examine the concerns that new technologies will render labor redundant in a framework in which tasks previously performed by labor can be automated and new versions of existing tasks, in which labor has a comparative advantage, can be created. In a static version where capital is fixed and technology is exogenous, automation reduces employment and the labor share, and may even reduce wages, while the creation of new tasks has the opposite effects. Our full model endogenizes capital accumulation and the direction of research toward automation and the creation of new tasks. If the long-run rental rate of capital relative to the wage is sufficiently low, the long-run equilibrium involves automation of all tasks. Otherwise, there exists a stable balanced growth path in which the two types of innovations go hand-in-hand. Stability is a consequence of the fact that automation reduces the cost of producing using labor, and thus discourages further automation and encourages the creation of new tasks. In an extension with heterogeneous skills, we show that inequality increases during transitions driven both by faster automation and the introduction of new tasks, and characterize the conditions under which inequality stabilizes in the long run. (JEL D63, E22, E23, E24, J24, O33, O41)

A high proportion of skilled workers in the labor force implies a large market size for skill-complementary technologies, and encourages faster upgrading of the productivity of skilled workers. As a result, an increase in the supply of skills reduces the skill premium in the short run, but then it induces skill-biased technical change and increases the skill premium, possibly even above its initial value. This theory suggests that the rapid increase in the proportion of college graduates in the United States labor force in the 1970s may have been a causal factor in both the decline in the college premium during the 1970s and the large increase in inequality during the 1980s.

For many problems in macroeconomics, development economics, labour economics, and international trade, whether technical change is biased towards particular factors is of central importance. This paper develops a simple framework to analyse the forces that shape these biases. There are two major forces affecting equilibrium bias: the price effect and the market size effect. While the former encourages innovations directed at scarce factors, the latter leads to technical change favouring abundant factors. The elasticity of substitution between different factors regulates how powerful these effects are, determining how technical change and factor prices respond to changes in relative supplies. If the elasticity of substitution is sufficiently large, the long run relative demand for a factor can slope up.I apply this framework to develop possible explanations to the following questions: why technical change over the past 60 years was skill biased, and why the skill bias may have accelerated over the past 25 years? Why new technologies introduced during the late eighteenth and early nineteenth centuries were unskill biased? What is the effect of biased technical change on the income gap between rich and poor countries? Does international trade affect the skill bias of technical change? What are the implications of wage push for technical change? Why is technical change generally labour augmenting rather than capital augmenting? Copyright 2002, Wiley-Blackwell.

This paper investigates a possibly fundamental aspect of technological progress. If knowledge accumulates as technology advances, then successive generations of innovators may face an increasing educational burden. Innovators can compensate through lengthening educational phases and narrowing expertise, but these responses come at the cost of reducing individual innovative capacities, with implications for the organization of innovative activity—a greater reliance on teamwork—and negative implications for growth. Building on this “burden of knowledge” mechanism, this paper first presents six facts about innovator behaviour. I show that age at first invention, specialization, and teamwork increase over time in a large micro-data set of inventors. Furthermore, in cross-section, specialization and teamwork appear greater in deeper areas of knowledge, while, surprisingly, age at first invention shows little variation across fields. A model then demonstrates how these facts can emerge in tandem. The theory further develops explicit implications for economic growth, providing an explanation for why productivity growth rates did not accelerate through the 20th century despite an enormous expansion in collective research effort. Upward trends in academic collaboration and lengthening doctorates, which have been noted in other research, can also be explained in this framework. The knowledge burden mechanism suggests that the nature of innovation is changing, with negative implications for long-run economic growth.

This paper compares the changing skill structure of wage bills and employment in the United States with six other OECD countries (Denmark, France, Germany, Japan, Sweden, and the United Kingdom). We investigate whether a directly observed measure of technical change (R&D intensity) is closely linked to the growth in the importance of more highly skilled workers which has occurred in all countries. Evidence of a significant association between skill upgrading and R&D intensity is uncovered in all seven countries. These results provide evidence that skill-biased technical change is an international phenomenon that has had a clear effect of increasing the relative demand for skilled workers.

Journal Article Induced Bias in Innovation and the Theory of Distribution Get access Charles Kennedy Charles Kennedy University of the West Indies, Mona, Jamaica Search for other works by this author on: Oxford Academic Google Scholar The Economic Journal, Volume 74, Issue 295, 1 September 1964, Pages 541–547, https://doi.org/10.2307/2228295 Published: 01 September 1964

Abstract We argue theoretically and document empirically that aging leads to greater (industrial) automation, because it creates a shortage of middle-aged workers specializing in manual production tasks. We show that demographic change is associated with greater adoption of robots and other automation technologies across countries and with more robotics-related activities across U.S. commuting zones. We also document more automation innovation in countries undergoing faster aging. Our directed technological change model predicts that the response of automation technologies to aging should be more pronounced in industries that rely more on middle-aged workers and those that present greater opportunities for automation and that productivity should improve and the labor share should decline relatively in industries that are more amenable to automation. The evidence supports all four of these predictions.

We characterize the factors that determine who becomes an inventor in the United States, focusing on the role of inventive ability (“nature”) versus environment (“nurture”). Using deidentified data on 1.2 million inventors from patent records linked to tax records, we first show that children's chances of becoming inventors vary sharply with characteristics at birth, such as their race, gender, and parents' socioeconomic class. For example, children from high-income (top 1%) families are 10 times as likely to become inventors as those from below-median income families. These gaps persist even among children with similar math test scores in early childhood-which are highly predictive of innovation rates-suggesting that the gaps may be driven by differences in environment rather than abilities to innovate. We directly establish the importance of environment by showing that exposure to innovation during childhood has significant causal effects on children's propensities to invent. Children whose families move to a high-innovation area when they are young are more likely to become inventors. These exposure effects are technology class and gender specific. Children who grow up in a neighborhood or family with a high innovation rate in a specific technology class are more likely to patent in exactly the same class. Girls are more likely to invent in a particular class if they grow up in an area with more women (but not men) who invent in that class. These gender- and technology class-specific exposure effects are more likely to be driven by narrow mechanisms, such as role-model or network effects, than factors that only affect general human capital accumulation, such as the quality of schools. Consistent with the importance of exposure effects in career selection, women and disadvantaged youth are as underrepresented among high-impact inventors as they are among inventors as a whole. These findings suggest that there are many “lost Einsteins”-individuals who would have had highly impactful inventions had they been exposed to innovation in childhood-especially among women, minorities, and children from low-income families.

The authors develop a model of vintage human capital in which each technology requires vintage-specific skills. They examine the properties of a stationary equilibrium for their economy. The stationary equilibrium is characterized by an endogenous distribution of skilled workers across vintages. The distribution is shown to be single-peaked. Under general conditions, there is a lag between the appearance of a technology and its peak usage, a phenomenon known as diffusion. An increase in the rate of exogenous technological change shifts the distribution of human capital to more recent vintages, thereby increasing the diffusion rate. Copyright 1991 by University of Chicago Press.

This paper studies whether labor scarcity encourages technological advances, that is, technology adoption or innovation, for example, as claimed by Habakkuk in the context of nineteenth-century United States. I define technology as strongly labor saving if technological advances reduce the marginal product of labor and as strongly labor complementary if they increase it. I show that labor scarcity encourages technological advances if technology is strongly labor saving and will discourage them if technology is strongly labor complementary. I also show that technology can be strongly labor saving in plausible environments but not in many canonical macroeconomic models.

This paper considers a dynamic model of innovations in which firms can endogenously bias the direction of technological change. Both in a North–North and North–South context, we show that, when globalization triggers an increased threat of technological leapfrogging or imitation, firms tend to respond to that threat by biasing the direction of their innovations towards skilled-labor-intensive technologies. We show that this process of defensive skill-biased innovations generates an increase in wage inequalities in both regions. We then discuss suggestive empirical evidence of the existence of defensive skill-biased technical change.

We build an endogenous growth model with automation (the replacement of low-skill workers with machines) and horizontal innovation (the creation of new products). Over time, the share of automation innovations endogenously increases through an increase in low-skill wages, leading to an increase in the skill premium and a decline in the labor share. We calibrate the model to the US economy and show that it quantitatively replicates the paths of the skill premium, the labor share, and labor productivity. Our model offers a new perspective on recent trends in the income distribution by showing that they can be explained endogenously. (JEL D31, E25, J24, J31, O33, O41)

This paper investigates the impact of higher education on corporate innovation. To establish causality, we exploit a policy-induced exogenous shock in the supply of Chinese college-educated labor starting in 2003. Using a difference-in-differences approach, we find that Chinese firms in skilled industries generate better innovation outcomes as measured by patents and citations than those in unskilled industries. This effect is more pronounced among firms headquartered in a province with more science and engineering college graduates, young firms that are more likely to hire young graduates, and firms located near universities. Moreover, higher education expansion increases a firm's innovative human capital in terms of the number of educated employees and inventors. Finally, we show that technological innovation is a mechanism through which higher education affects productivity growth and, thus, the economy.

We study a dynamic general equilibrium model where innovation takes the form of the introduction of new goods whose production requires skilled workers. Innovation is followed by a costly process of standardization, whereby these new goods are adapted to be produced using unskilled labor. Our framework highlights a number of novel results. First, standardization is both an engine of growth and a potential barrier to it. As a result, growth is an inverse U-shaped function of the standardization rate (and of competition). Second, we characterize the growth and welfare maximizing speed of standardization. We show how optimal protection of intellectual property rights affecting the cost of standardization vary with the skill-endowment, the elasticity of substitution between goods and other parameters. Third, we show that, depending on how competition between innovating and standardizing firms is modelled and on parameter values, a new type of multiplicity of equilibria may arise. Finally, we study the implications of our model for the skill premium and we illustrate novel reasons for linking North-South trade to intellectual property rights protection. © 2010 Elsevier Inc.

Half of college students drop out before completing a degree. These low rates of college completion among young people should be viewed in the context of slow future growth in the educated labor force, as the well-educated baby boomers retire and new workers are drawn from populations with historically low education levels. This paper establishes a causal link between college costs and the share of workers with a college education. I exploit the introduction of two large tuition subsidy programs, finding that they increase the share of the population that completes a college degree by three percentage points. The effects are strongest among women, with white women increasing degree receipt by 3.2 percentage points and the share of nonwhite women attempting or completing any years of college increasing by six and seven percentage points, respectively. A cost-benefit analysis indicates that tuition reduction can be a socially efficient method for increasing college completion. However, even with the offer of free tuition, a large share of students continue to drop out, suggesting that the direct costs of school are not the only impediment to college completion.

The effects of automation on economic growth, education, and inequality are analyzed using an R&D-driven growth model with endogenous education in which high-skilled workers are complements to machines and low-skilled workers are substitutes for machines. The model predicts that automation leads to an increasing share of college graduates, increasing income and wealth inequality, and a declining labor share. We show that standard policy suggestions for the age of automation can trigger unintended side effects on inequality, growth, and welfare, irrespective of whether they are financed by progressive wage taxation or by a robot tax.

We examine the contribution of human capital to economy-wide technological improvements through the two channels of innovation and imitation. We develop a theoretical model showing that skilled labor has a higher growth-enhancing effect closer to the technological frontier under the reasonable assumption that innovation is a relatively more skill-intensive activity than imitation. Also, we provide evidence in favor of this prediction using a panel dataset covering 19 OECD countries between 1960 and 2000 and explain why previous empirical research had found no positive relationship between initial schooling level and subsequent growth in rich countries.

This study provides causal evidence that a shock to the relative supply of inputs to production can (1) affect the direction of technological progress and (2) lead to a rebound in the relative price of the input that became relatively more abundant (the strong induced-bias hypothesis). I exploit the impact of the U.S. Civil War on the British cotton textile industry, which reduced supplies of cotton from the Southern United States, forcing British producers to shift to lower-quality Indian cotton. Using detailed new data, I show that this shift induced the development of new technologies that augmented Indian cotton. As these new technologies became available, I show that the relative price of Indian/U.S. cotton rebounded to its pre-war level, despite the increased relative supply of Indian cotton. This is the first paper to establish both of these patterns empirically, lending support to the two key predictions of leading directed technical change theories.

Half of college students drop out before completing a degree. These low rates of college completion among young people should be viewed in the context of slow future growth in the educated labor force, as the well-educated baby boomers retire and new workers are drawn from populations with historically low education levels. This paper establishes a causal link between college costs and the share of workers with a college education. I exploit the introduction of two large tuition subsidy programs, finding that they increase the share of the population that completes a college degree by three percentage points. The effects are strongest among women, with white women increasing degree receipt by 3.2 percentage points and the share of nonwhite women attempting or completing any years of college increasing by six and seven percentage points, respectively. A cost-benefit analysis indicates that tuition reduction can be a socially efficient method for increasing college completion. However, even with the offer of free tuition, a large share of students continue to drop out, suggesting that the direct costs of school are not the only impediment to college completion.

Although microeconomic studies find a positive relationship between R&D and skill premia, much of the recent rise in U.S. wage inequality was accompanied by slowing labor-productivity growth and relatively slow introduction of new technologies. These conflicting observations are consistent with the effects of a skewed distribution of “absorptive capacities”—the rate at which technology-specific skills can be acquired—in a model of endogenous technological change. The framework is used to assess whether the productivity slowdown and the rise in wage inequality can be jointly accounted for by the contemporaneous decline in the growth rate of labor quality. (JEL E24, J31, O3)

This study revisits the important ideas proposed by Atkinson and Stiglitz's seminal 1969 paper on technological change. After linking these ideas to the induced innovation literature of the 1960s and the more recent directed technological change literature, it explains how these three complementary but different approaches are useful in the study of a range of current research areas – though they may also yield different answers to important questions. It concludes by highlighting several important areas where these ideas can be fruitfully applied in future work. Atkinson, A.B. and Stiglitz, J.E. (1969). 'A new view of technological change', Economic Journal, vol. 79(315), pp. 573–8.

Abstract Using a broad panel of advanced economies, we document that increases in GDP per capita are associated with a systematic shift in the composition of value added to sectors that are intensive in high-skill labour, a process we label as skill-biased structural change. It follows that further development in these economies leads to an increase in the relative demand for skilled labour. We develop a quantitative two-sector model of this process as a laboratory to assess the sources of the rise of the skill premium in the U.S. and a set of ten other advanced economies, over the period 1977 to 2005. For the U.S., we find that the sector-specific skill neutral component of technical change accounts for 18–24% of the overall increase of the skill premium due to technical change, and that the mechanism through which this component of technical change affects the skill premium is via skill-biased structural change.

Using detailed plant-le vel data from the 1988 and 1993 Surveys of Manufacturing Technology, this paper examines the impact of skill mix in U.S. local labor markets on the use and adoption of automation technologies in manufacturing.The level of automation differs widely across U.S. metropolitan areas.In both 1988 and 1993, in markets with a higher relative availability of lessskilled labor, comparable plants -even plants in the same narrow (4-digit SIC) industries -used systematically less automation.Moreover, between 1988 and 1993 plants in areas experiencing faster less-skilled relative labor supply growth adopted automation technology more slowly, both overall and relative to expectations, and even de-adoption was not uncommon.This relationship is stronger when examining an arguably exogenous component of local less-skilled labor supply derived from historical regional settlement patterns of immigrants from different parts of the world.These results have implications for two long-standing puzzles in economics.First, they potentially explain why research has repeatedly found that immigration has little impact on the wages of competing native-born workers at the local level.It might be that the technologies of local firms-rather than the wages that they offer-respond to changes in local skill mix associated with immigration.A modified two-sector model demonstrates this theoretical possibility.Second, the results raise doubts about the extent to which the spread of new technologies have raised demand for skills, one frequently forwarded hypothesis for the cause of rising wage inequality in the United States.Causality appears to at least partly run in the opposite direction, where skill supply drive s the spread of skill-complementary technology.

We treat rising inequality as an equilibrium outcome in which human capital investment fails to keep pace with rising demand for skills. Investment affects skill supply and prices on three margins: the type of human capital in which to invest, how much to acquire, and the intensity of use. The latter two represent the intensive margins of human capital acquisition and utilization. These choices are substitutes for the creation of new skilled workers, yet they are complementary with each other, magnifying inequality. When skill-biased technical change drives economic growth, greater inequality reduces growth.

In the presence of markup differences, externalities, and other social effects, the direction of innovation can be systematically distorted. I build a simple model of endogenous technology to study distortions in the direction of innovation. Empirical findings across a number of different areas are consistent with this framework's predictions. I use data from several studies to estimate the framework's key parameters and combine them with rough estimates of differential externalities and markups to provide suggestive evidence that innovation distortions can be substantial in the context of industrial automation, health care, and energy, and that correcting them could have sizable welfare benefits.

In this paper we argue that government spending played a significant role in stimulating the wave of innovation that hit the U.S. economy in the late 1970s and in the 1980s, as well as the simultaneous increase in inequality and in education attainments. Since the late 1970s U.S. policymakers began targeting commercial innovations more directly and explicitly. We focus on the shift in the composition of public demand toward high-tech goods, which, by increasing the market-size of innovative firms, functions as a de facto innovation policy tool. We build a quality-ladders non-scale growth model with heterogeneous industries and endogenous supply of skills, and show that an increase in the technological content of public spending stimulates R&D, raises the wage of skilled workers, and, at the same time, stimulates human capital accumulation. A calibrated version of the model suggests that government policy explains between 12% and 15% of the observed increase in wage inequality in the period 1976–1991. (JEL: E62, J31, O33, O41) \n \n

This Paper considers a dynamic model of innovations in which firms can endogenously bias the direction of technological change. Both in a North-North and North-South context, we show that, when globalization triggers an increased threat of technological leapfrogging or imitation, firms tend to respond to that threat by biasing the direction of their innovations towards skilled labour-intensive technologies. We show that this process of defensive skill biased innovations generates an increase in wage inequalities in both regions. We then discuss suggestive empirical evidence of the existence of defensive skill biased technical change.

In the presence of markup differences, externalities and other social considerations, the equilibrium direction of innovation can be systematically distorted. This paper builds a simple model of endogenous technology, which generalizes existing comparative static results and characterizes potential distortions in the direction of innovation. I show that empirical findings across a number of different areas are consistent with this framework's predictions and I use data from several studies to estimate its key parameters. Combining these numbers with rough estimates of differential externalities and markups, I provide suggestive evidence that equilibrium distortions in the direction of technology can be substantial in the context of industrial automation, health care, and energy, and correcting these distortions could have sizable welfare benefits.

We study how the change in the price of labor affects the direction of technological change using a novel measure decomposing innovations into products (new goods) and processes (lower production costs). Using the 1999 U.S.-China agreement as a shock that lowered effective labor cost, we find that U.S. firms operating in China decrease their share of process innovations by 9% and that this adjustment is driven by lower process innovation. We obtain the same results using a staggered loosening of restrictions on foreign ownership across industries in China over 1995-2012. This suggests that cheap abundant labor substitutes for labor-saving innovation.

This paper develops a model of vintage human capital in which each technology requires vintage specific skills. We examine the properties of a stationary equilibrium for our economy. The stationary equilibrium is characterized by an endogenous distribution of skilled workers across vintages. The distribution is shown to be single peaked and, under general conditions, there is a lag between the time when a technology appears and the peak of it's usage, a phenomenon known as diffusion. An increase in the rate of exogenous technological change shifts the distribution of human capital to more recent vintages thereby increasing the diffusion rate.

We use advances in natural language processing to construct new measures of workers' task-level exposure to artificial intelligence (AI) and machine learning from 2010 to 2023, capturing variation across firms, occupations, and time.Tasks with higher AI exposure subsequently experience reduced labor demand.To interpret these patterns, we develop a model that separates direct substitution from indirect reallocative effects of labor-saving technologies.Two variables summarize the impact of AI on within-firm labor demand: the mean exposure of an occupation's tasks, which depresses demand, and the concentration of exposure in a few tasks, which offsets losses by enabling workers to reallocate effort.Using an instrument based on historical university hiring networks, we find causal evidence consistent with these predictions.Despite strong substitution at the task level, overall employment effects are modest, as reduced demand in exposed occupations is offset by productivity-driven increases in labor demand at AI-adopting firms.

The recent wave of AI and automation has been argued to differ from previous General Purpose Technologies (GPTs), in that it may lead to rapid change in occupations' underlying task requirements and persistent technological unemployment. In this paper, we apply a novel methodology of dynamic task shares to a large dataset of online job postings to explore how exactly occupational task demands have changed over the past decade of AI innovation, especially across high, mid and low wage occupations. Notably, big data and AI have risen significantly among high wage occupations since 2012 and 2016, respectively. We built an ARIMA model to predict future occupational task demands and showcase several relevant examples in Healthcare, Administration, and IT. Such task demands predictions across occupations will play a pivotal role in retraining the workforce of the future.

David Ricardo initially believed machinery would help workers but revised his opinion, likely based on the impact of automation in the textile industry. Despite cotton textiles becoming one of the largest sectors in the British economy, real wages for cotton weavers did not rise for decades. As E.P. Thompson emphasized, automation forced workers into unhealthy factories with close surveillance and little autonomy. Automation can increase wages, but only when accompanied by new tasks that raise the marginal productivity of labor and/or when there is sufficient additional hiring in complementary sectors. Wages are unlikely to rise when workers cannot push for their share of productivity growth. Today, artificial intelligence may boost average productivity, but it also may replace many workers while degrading job quality for those who remain employed. As in Ricardo's time, the impact of automation on workers today is more complex than an automatic linkage from higher productivity to better wages.

We randomly assign applicants to over-subscribed programs to study the effects of teaching hard and soft skills in vocational training and examine their impacts on skills and labor market outcomes using both survey and administrative data. We find that providing vocational training that either emphasizes social or technical skills increases formal employment. We also find that admission to a vocational program that emphasizes technical relative to social skills increases overall employment and also days and hours worked in the short term. Yet, emphasis on softskills training helps applicants sustain employment and monthly wages over the longer term and allows them to catch up with those learning hard skills. Further, through a second round of randomization, we find that offering financial support for transportation and food increases the effectiveness of the program, indicating that resource constraints may be an obstacle for individuals considering vocational training.

Abstract How do innovation and education policy affect individual career choices and aggregate productivity? This paper analyses the effect of R&D subsidies and higher education policy on productivity growth through the supply of innovative talent. We put scarce talent, higher education attainment, and career choice at the centre of a new endogenous growth framework with individual-level heterogeneity in talent, financial resources, and preferences. We link the model to micro-level data from Denmark on the backgrounds of who obtains a PhD and becomes an inventor and the outcomes of a set of policy interventions. We find that R&D subsidies can be strengthened when combined with higher education subsidies, which enable talented but poor youth to pursue a career in research. Education and innovation policies not only alleviate different frictions, but also impact innovation at different time horizons. Education policy is more effective in societies with higher income inequality.

This Paper provides an interpretation for the recent rise in residual wage inequality which is consistent with the empirical observation that a sizeable part of this increase has a transitory nature, a feature that eludes standard models based on ex-ante heterogeneity in ability. In the model an acceleration in the rate of quality-improvement of equipment, like the one observed from the early 70's, reduces workers’ capacity to transfer skills from old to new machines. This force generates a rise in the cross-sectional variance of skills, and therefore of wages. Through calibration, the Paper shows that this mechanism can account for 30% of the surge in residual inequality in the US economy (or for most of its transitory component). Two key implications of the theory - faster within job wage growth and larger wage losses upon displacement - find empirical support in the data.

The misallocation of talent in innovation – “missing Einsteins” – has a first-order impact on growth and welfare. Surname-level empirical analysis combining inventor and census micro-data reveals people from richer backgrounds are more likely to become inventors, but those from high-education backgrounds become more prolific inventors. Motivated by this discrepancy, an endogenous growth model with financial frictions on the household side is developed. Individuals compete for scarce inventor training. The rich can become inventors even if mediocre through excessive credentialing spending. Shutting down credentialing spending raises innovation, growth, welfare, and inequality. Optimal progressive bequest taxes increase growth and welfare, but reduce inequality.

We develop a measure of workers' technology exposure that relies only on textual descriptions of patent documents and the tasks performed by workers in an occupation. Our measure appears to identify a combination of labor-saving innovations but also technologies that may require skills that incumbent workers lack. Using a panel of administrative data, we examine how subsequent worker earnings relate to workers' technology exposure. We find that workers at both the bottom but also the top of the earnings distribution are displaced. Our interpretation is that low-paid workers are displaced as their tasks are automated while the highest-paid workers face lower earnings growth as some of their skills become obsolete. Our calibrated model fits these facts and emphasizes the importance of movements in skill quantities, not just skill prices, for the link between technology and inequality.

We document the growth in higher education costs and tuition over the past 50 years. To explain these trends, we develop a general equilibrium model with skill- and sector-biased technical change. Finding the model’s parameters through a combination of estimation and calibration, we show that it can explain the rise in college costs between 1961 and 2009, along with the increase in college attainment and the change in the relative earnings of college graduates. The model predicts that if college costs had ceased to grow after 1961, enrollment in 2010 would have been 3%–6% higher.

We construct a search model with endogenous human capital and labor participation to study the growth effects of short-run frictions and the effectiveness of human capital policies. Employment, learning effort, and output growth increase with more effective learning, better labor-market matching, lower job separation, or less costly vacancy creation. Although output growth, employment, vacancy creation, and learning and search effort are most responsive to changes in a human capital policy that directly affects learning effort, such a policy need not be more beneficial for welfare. The effects of human capital policies become larger as the severity of labor-market frictions rises.

Studies of innovation have focused on the effects of patent laws on the number of innovations but ignored effects on the direction of technological change. This paper introduces a new data set of close to fifteen thousand innovations at the Crystal Palace World's Fair in 1851 and at the Centennial Exhibition in 1876 to examine the effects of patent laws on the direction of innovation. The paper tests the following argument: if innovative activity is motivated by expected profits, and if the effectiveness of patent protection varies across industries, then innovation in countries without patent laws should focus on industries where alternative mechanisms to protect intellectual property are effective. Analyses of exhibition data for twelve countries in 1851 and ten countries in 1876 indicate that inventors in countries without patent laws focus on a small set of industries where patents were less important, while innovation in countries with patent laws appears to be much more diversified. These findings suggest that patents help to determine the direction of technical change and that the adoption of patent laws in countries without such laws may alter existing patterns of comparative advantage across countries.

This paper formulates and estimates an open-economy overlapping generation general-equilibrium model of endogenous heterogeneous human capital in the form of schooling and on-the-job training. Physical capital accumulation is also analyzed. We use the model to explain rising wage inequality in the past two decades due to skill-biased technical change and to estimate investment responses. We compare an open economy version with a closed economy version. Using our empirically grounded general equilibrium model that explains rising wage inequality, we evaluate two policies often suggested as solutions to the problem of rising wage inequality: (a) tuition subsidies to promote skill formation and (b) tax policies. We establish that conventional partial equilibrium policy evaluation methods widely used in labor economics and public finance give substantially misleading estimates of the impact of national tax and tuition policies on skill formation. Conventional microeconomic methods for estimating the schooling response to tuition overestimate the response by an order of magnitude. Simulations of our model also reveal that move to a flat consumption tax raises capital accumulation and the real wages of all skill groups and barely affects overall measures of income inequality.

Abstract Economic developments of the past 3 decades posed new questions to economists: What are the causes of fluctuations in rates of return to human capital? What is the relation between the changing skill-wage structure and changing overall wage inequality? Does the widening of the wage structure produce an equilibrating supply response? What are the causes, dimensions and implications of the “technological cycle” for wages, unemployment, and its “natural rate”? Why is the long term trend of human capital formation relentlessly upward? My research of the past decade, among that of other economists, attempted to provide answers to these questions, as described above. In the course of the analysis several misconceptions are clarified: (1) The view of an increasing “wage gap” as a worsening “social divide” misses the incentive effects of the increased rates of return on furthering investments in human capital. These are empirically documented. (2) Growing overall wage inequality can conceal a declining inequality of opportunity as it did in recent decades. (3) Technological unemployment as an aggregate phenomenon appears to be a myth. (4) The concurrent supply response to increasing demand for human capital applies to investments, not to the stock. The accumulation of investments (such as enrollments) over time produces a lag in the response of the human capital stock. This lag is a basic cause of the “technological cycle”. Finally, it is worth noting that a positive skill bias is not inherent in technological changes. These may sometimes carry a negative effect on the demand for human capital. The implications of “deskilling” (the assembly line is an example) would be the opposite of what we found for the recent decades (1970–2000). However the long-term growth of human capital suggests a positive skill bias in the long run.

Using data on MIT bachelor's graduates from 1994 to 2012, this paper empirically examines the extent to which the inflow of elite talent into the financial industry affects the supply of innovators in science and engineering (S&E). I first show that finance does not systematically attract those who are best prepared at college graduation to innovate in S&E sectors. Among graduates who majored in S&E, cumulative GPA strongly and positively predicts long-term patenting; this result is robust to controlling for choices of major and career. In contrast, GPA negatively predicts the probability of taking a first job in finance after college. There is suggestive evidence that S&E and finance value different sets of skills: innovating in S&E calls for in-depth knowledge and/or interest in a specific subject area, whereas finance tends to value a combination of general analytic skills and social skills over academic specialization. I then provide evidence that anticipated career incentives influence students' acquisition of S&E human capital during college. The 2008-09 financial crisis, which substantially reduced the availability of jobs in finance and led to a worsening labor market in general, prompted some students to major in S&E instead of management or economics and/or to improve their academic performance. This response to the shock is driven by students with below-average academic credentials who were freshmen at the peak of the crisis.

This paper assesses the impact of investment- and education-specific technical change on occupational transition and the skill premium in a model with human capital. In this framework, human capital augments labor productivity and also facilitates the transition to skilled employment. In line with empirical evidence, this setup predicts that an increase in the productivity of physical capital (investment-specific change) leads to very small increases in the relative supply of skilled workers and to significant and rising increases in the skill premium. Additionally, reforms that improve the productivity of resources used in education (education-specific change) reduce wage inequality and increase mobility.

This paper considers how new technologies affect the returns to experience and how experience affects the adoption of new technologies. Whereas the traditional vintage model emphasizes skill obsolescence generated by imperfect transferability of skills across technologies, we consider the possibility that new technologies complement existing skills. Consistent with the vintage model, among college graduate men, young workers have adopted computers most intensively and the returns to experience have been flat. Among high school graduate men however, experienced workers have adopted new technologies most intensively and the returns to experience have increased, pointing toward complementarities between existing skills and new technologies.

Rising inequality in the relative wages of skilled and unskilled labor is often attributed to skill-biased technological progress. This paper presents a model in which the adoption of skill-biased or \\"unskilled-biased\\" technologies is endogenous. Conventional wisdom states that an increase in the supply of skilled labor lowers the relative wage of skilled to unskilled labor. In this paper, an increase in the supply of skilled labor leads to temporary stagnation in the wages of unskilled workers and an expanding gap between the wages of skilled and unskilled workers through an acceleration of skill-biased technological change.

I introduce endogenous human-capital accumulation into an infinite-horizon version of Chari and Hopenhayn's (1991) [4] vintage-human-capital model. Returns to skill and tenure premia are highest in young vintages, where skill is scarcest and agents accumulate human capital fastest. As the vintage ages, the skill premium decreases and vanishes entirely upon vintage death. Workers run through cycles of human-capital accumulation: their wages rise as they accumulate skill, undergo downward pressure as the technology ages, and finally drop sharply when the worker switches to a new technology. The results are in line with German linked employer-employee data: tenure premia are highest in young establishments, as well as in fast-growing industries, occupations and establishments. A calibration exercise suggests that human-capital accumulation is the most important determinant of workers' wage profiles, whereas changes to the price of skill and vintage productivity gains play a smaller quantitative role. © 2013 Elsevier Inc.

We construct a dynamic general equilibrium model with occupation mobility, human capital accumulation and endogenous assignment of workers to tasks to quantitatively assess the aggregate impact of automation and other task-biased technological innovations. We extend recent quantitative general equilibrium Roy models to a setting with dynamic occupational choices and human capital accumulation. We provide a set of conditions for the problem of workers to be written in recursive form and provide a sharp characterization for the optimal mobility of individual workers and for the aggregate supply of skills across occupations. We craft our dynamic Roy model in a production setting where multiple tasks within occupations are assigned to workers or machines. We solve for the balanced-growth path and characterize the aggregate transitional dynamics ensuing task-biased technological innovations. In our quantitative analysis of the impact of task-biased innovations in the U.S. since 1980, we find that they account for an increased aggregate output in the order of 75% and for a much higher dispersion in earnings. If the U.S. economy had larger barriers to mobility it would have experienced less job polarization but substantially higher inequality and lower output as occupation mobility has provided a "escape" for the losers from automation.

One long-standing hypothesis about science and engineering labor markets is that the supply of highly skilled workers is likely to be inelastic in the short run. We consider the market for computer scientists and electrical engineers (IT workers) and the evolution of wages and employment through two periods of increased demand. Relative to the boom of the 1970s, the demand shock in the 1990s generated relatively greater changes in employment and smaller changes in wages. The growth in the pool of skilled workers abroad, combined with increased immigration in high-skill fields, is central to this story.

The skill premium — the wage of skilled labor relative to the wage of unskilled labor — has increased substantially in the United States since the 1970s. The higher skill premium has been attributed to skill-biased factor-specific technical change or to capital accumulation when skilled labor is relatively more complementary to capital than is unskilled labor. The authors describe an economic mechanism through which factor-specific technical change and capital accumulation respond to changes in the rate of capital-embodied technical change, and they trace out how accelerated capital-embodied technical change increases the skill premium in the medium and long run.

This paper builds an endogenous growth model of directed technical change with vertical and horizontal R&D and scale effects at the industry level to study an analytical mechanism that is consistent with the observed crosscountry pattern in the skill structure, the technology structure and economic growth. We calibrate the model in order to uncover the effect of the skill structure on economic growth by studying how the former affects the technology structure. We find that the small positive elasticity of the economic growth rate regarding the ratio of high- to low-skilled workers that is empirically observed is explained by the combination of moderate levels of the market complexity costs related to vertical R&D and high entry costs in the high- vis-à-vis the low-tech sectors, which dampen the positive direct effect of the absolute productivity advantage of the high-skilled workers on growth.

Abstract This article generalizes central results from the theory of (endogenously) directed technical change to settings where technology does not take a labour-augmenting form and with arbitrarily many levels of skill. Building on simple notions of complementarity, the results remain intuitive despite their generality. The developed theory allows to study the endogenous determination of labour-replacing, that is, automation technology through the lens of directed technical change theory. In an assignment model with a continuum of differentially skilled workers and capital, where capital perfectly substitutes for labour in the production of tasks, any increase in the relative supply of skilled workers stimulates investment into improving the productivity of capital, potentially leading skill premia to increase in relative skill supply. Relatedly, trade with a skill-scarce country discourages improvements in capital productivity, potentially reversing the standard Heckscher–Ohlin effects.

We study how inequality, skills, and economic activity adjust over time to technological innovations. We develop a theory of technological transitions where economies adjust through two margins: (i) within-generation reallocation of workers with heterogeneous skills, and (ii) cross-generation changes in the skill distribution driven by entering generations investing in skills. We then characterize the equilibrium dynamics, showing that they resemble those of a qtheory of skill investment where q is lifetime inequality. Technological transitions are slower and more unequal whenever innovations are biased towards economic activities intensive in skills which differ more from those used in the rest of the economy-i.e., technology-skill specificity is higher. This is because the first margin is weaker and the second stronger. Lastly, we document that recent cognitive-biased innovations caused responses in occupational composition and training which were strong for younger generations but weak for older ones. This evidence is consistent with high technology-skill specificity, implying that cognitive-biased transitions are particularly slow and unequal because they are mainly driven by changes in the skill distribution across generations.

This paper focuses on the bi-directional interaction between technology adoption and labor market conditions. We examine cross-city differences in PC adoption, relative wages, and changes in relative wages over the period 1980-2000 to evaluate whether the patterns conform to the predictions of a neoclassical model of endogenous technology adoption. Our approach melds the literature on the effect of the relative supply of skilled labor on technology adoption to the often distinct literature on how technological change influences the relative demand for skilled labor. Our results support the idea that differences in technology use across cities and its effects on wages reflect an equilibrium response to local factor supply conditions. The model and data suggest that cities initially endowed with relatively abundant and cheap skilled labor adopted PCs more aggressively than cities with relatively expensive skilled labor, causing returns to skill to increase most in cities that adopted PCs most intensively. Our findings indicate that neoclassical models of endogenous technology adoption can be very useful for understanding where technological change arises and how it affects markets.

This study investigates the differential effects of technological change on general human capital acquired through schooling and technology-specific human capital acquired through training based on a life-cycle human capital investment model. Using data from the National Longitudinal Survey of Youth 79 (1987–2003), I find that for both high-ability and low-ability individuals, the net effect of technological change on training human capital is obsolescence, whereas that on schooling human capital is an increase in productivity in spite of the obsolescence. This finding is consistent with the view that individuals with more schooling may enjoy an advantage under rapid technological change over those with less schooling. I also find that technological change exerts differential impacts on individuals with different ability levels, which provides support for the skill-biased technical change theory.

Abstract Why is invention strongly positively correlated with parental income not only in the United States but also in Finland, which displays low income inequality and high social mobility? Using data on 1.45 M Finnish individuals and their parents, we find the following: (i) the positive association between parental income and off‐spring probability of inventing is greatly reduced when controlling for parental education; (ii) instrumenting for the parents having an MSc degree using distance to nearest university reveals a large causal effect of parental education on offspring probability of inventing; and (iii) the causal effect of parental education has been markedly weakened by the introduction in the early 1970s of a comprehensive schooling reform.

We document historical patterns of workers' transitions across occupations and over the life-cycle for different levels of exposure and complementarity to Artificial Intelligence (AI) in Brazil and the UK. In both countries, college-educated workers frequently move from high-exposure, low-complementarity occupations (those more likely to be negatively affected by AI) to high-exposure, high-complementarity ones (those more likely to be positively affected by AI). This transition is especially common for young college-educated workers and is associated with an increase in average salaries. Young highly educated workers thus represent the demographic group for which AI-driven structural change could most expand opportunities for career progression but also highly disrupt entry into the labor market by removing stepping-stone jobs. These patterns of “upward” labor market transitions for college-educated workers look broadly alike in the UK and Brazil, suggesting that the impact of AI adoption on the highly educated labor force could be similar across advanced economies and emerging markets. Meanwhile, non-college workers in Brazil face markedly higher chances of moving from better-paid high-exposure and low-complementarity occupations to low-exposure ones, suggesting a higher risk of income loss if AI were to reduce labor demand for the former type of jobs.

Abstract China has expanded the yearly quota on newly admitted college students by more than seven times since 1999. How did this massive education expansion affect firms' export and innovation choices? I document that after this expansion impacted the labor market, manufacturing firms' innovation increased considerably, especially among exporting firms, accompanied by sizable skill upgrading of exports. I then develop a multi‐industry spatial equilibrium model, featuring skill intensity differences across industries and heterogeneous firms' innovation and export choices. Quantitatively, the college expansion explained 72% of increases in China's manufacturing research and development (R&D) intensity between 2003 and 2018 and also triggered export skill upgrading.

Why has schooling not countered the pervasive rises in wage inequality driven by skill-biased technical change? Using data and a model of directed technical change in which developing countries acquire technology licenses from abroad, we show technological change is skill-biased in the South simply because it is in the North. This causes permanently rising wage inequality in the South. We model expanded schooling access as producing relatively educated new cohorts of labor market entrants. This makes the market for skill-biased technologies more attractive, which generates accelerated skill-biased technical change, which leads to higher wage inequality and possibly stagnant unskilled wages.

In this paper, we develop a model of technology adoption and economic growth in which households optimally obtain either a concept-based, "general" education or a skill-specific, "vocational" education. General education is more costly to obtain, but enables workers to operate new technologies incorporated into production. Firms weigh the cost of adopting and operating new technologies against increased revenues and optimally choose the level of adoption. We show that an economy whose policies favor vocational education will grow slower in equilibrium than one that favors general education. Moreover, the gap between their growth rates will increase with the growth rate of available technology. By characterizing the optimal Ramsey education subsidy policy we demonstrate that the optimal subsidy for general education increases with the growth rate of available technology.

David Ricardo initially believed machinery would help workers but revised his opinion, likely based on the impact of automation in the textile industry. Despite cotton textiles becoming one of the largest sectors in the British economy, real wages for cotton weavers did not rise for decades. As E.P. Thompson emphasized, automation forced workers into unhealthy factories with close surveillance and little autonomy. Automation can increase wages, but only when accompanied by new tasks that raise the marginal productivity of labor and/or when there is sufficient additional hiring in complementary sectors. Wages are unlikely to rise when workers cannot push for their share of productivity growth. Today, artificial intelligence may boost average productivity, but it also may replace many workers while degrading job quality for those who remain employed. As in Ricardo’s time, the impact of automation on workers today is more complex than an automatic linkage from higher productivity to better wages.

European economic growth has been weak, compared to the US, since the 80s. In previous work (Krueger and Kumar (2003)), we argued that the European focus on specialized, vocational education might have been effective during the 60s and 70s, but resulted in a growth gap relative to the US during the subsequent information age, when new technologies emerged more rapidly. In this paper, we extend this framework to assess the importance of education policy, when compared to labor market rigidity and product market regulation, which have also been suggested as reasons for US-Europe differences. Households decide between acquiring general education, which allows them to work in high-tech firms, and less costly skill-specific education, which is of value only to lowtech firms that use established production methods. High-tech firms draw a workforcespecific productivity for the new technology, and decide on whether to proceed with production and pay a portion of profits toward regulation costs, or fire the workers at a cost, and redraw a new productivity-workforce combination. Analytical characterization of the balanced growth equilibrium shows that lower firing or regulation cost increases expected growth, but causes the adopting firm to set a higher productivity threshold before it proceeds with production. A higher subsidy for general education can increase expected growth, but reduces the productivity threshold. An increased rate of technology availability can increase the gap in the growth rates of economies that differ in their policies. A “decomposition” exercise using a calibrated version of our model assigns a major role to education policy in explaining US-Europe growth differences.

Using a matched employer-employee dataset on the French manufacturing sector in the 1990s, we investigate how training incidence responds to technical and organizational changes. Using a difference-in-difference approach across age groups and types of firms, we find that older workers in low-skill occupations lag behind in terms of training (in computer skills and in teamwork) when firms implement advanced information technologies. By contrast, there is no significant difference between age groups in the training response to advanced IT among workers in high-skill occupations, or in the training response to new organizational practices (among all skill groups). These results suggest that a comparative disadvantage of older workers with regard to training in computer skills may be one cause of age-biased technical change. It severely affects low-skill older workers in firms implementing advanced information technologies. A partir de données appariées de salariés et d'entreprises industrielles en France dans les années 1990, on analyse comment l'accès à la formation continue évolue selon l'âge et en réponse aux changements technologiques et organisationnels

This paper draws on existing empirical literature and an original theoretical model to argue that globalization and skill supply affect the extent to which technology adoption in developing countries favors skilled workers. Developing countries are experiencing technical change that is skill-biased because skill-biased technologies are becoming relatively cheaper. Increased skill supply further biases technical change in favor of skilled labor. Free trade induces technology that favors skilled workers in skill-abundant developing countries and that favors unskilled workers in skill-scarce developing countries, and therefore amplifies the predicted wage effects of trade liberalization. These features aid our understanding of the observed rises in inequality within developing countries and the absence of a significant downward effect of expanded educational attainment on skill premia. They also help account for the large and differential effects of trade liberalization on inequality. These findings are pertinent for the Middle East and North Africa because of its recent increase in trade openness and remarkable rise in educational attainment.

Does capital accumulation increase labor demand and wages?Neoclassical production functions, where capital and labor are q-complements, ensure that the answer is yes, so long as labor markets are competitive.This result critically depends on the assumption that capital accumulation does not change the technologies being developed and used.I adapt the theory of endogenous technological change to investigate this question when technology also responds to capital accumulation.I show that there are strong parallels between the relationship between capital and wages and existing results on the conditions under which equilibrium factor demands are upward-sloping (e.g., Acemoglu, 2007).Extending this framework, I provide intuitive conditions and simple examples where a greater capital stock leads to lower wages, because it triggers more automation.I then offer an endogenous growth model with a menu of technologies where equilibrium involves choices over both the extent of automation and the rate of growth of labor-augmenting productivity.In this framework, capital accumulation and technological change in the long run are associated with wage growth, but an increase in the saving rate increases the extent of automation, and at first reduces the wage rate and subsequently depresses its long-run growth rate.

Abstract Does capital accumulation increase labor demand and wages? Neoclassical production functions, where capital and labor are q‐complements, ensure that the answer is yes, so long as labor markets are competitive. This result critically depends on the assumption that capital accumulation does not change the technologies being developed and used. I adapt the theory of endogenous technological change to investigate this question when technology also responds to capital accumulation. I show that there are strong parallels between the relationship between capital and wages and existing results on the conditions under which equilibrium factor demands are upward‐sloping (e.g., Acemoglu, Econometrica 75(5) (2007), 1371–410). Extending this framework, I provide intuitive conditions and simple examples where a greater capital stock leads to lower wages, because it triggers more automation. I then offer an endogenous growth model with a menu of technologies where equilibrium involves choices over both the extent of automation and the rate of growth of labor‐augmenting productivity. In this framework, capital accumulation and technological change in the long run are associated with wage growth, but an increase in the saving rate increases the extent of automation, and initially reduces the wage rate and can subsequently depress its long‐run growth rate.

Innovation booms are often fueled by easy financing that allows new technology firms to pay high wages that attracts skilled labor. Using the late 1990s Information and Communication Technology (ICT) boom as a laboratory, we show that skilled labor joining this new sector experienced sizeable long-term earnings losses. We show these earnings patterns are explained by faster skill obsolescence rather than either worker selection or the overall bust in the ICT sector. During the boom, financing flowed more to firms whose workers would experience the largest productivity declines, amplifying the negative effect of labor reallocation on aggregate human capital accumulation.

We investigate how individuals alter their educational investments in response to routine-biased technology. We find that individuals growing up in robot-impacted areas are more likely to complete a bachelor’s degree and experience a relative increase in earnings. Changes in the skill premium and opportunity cost appear to drive these effects. To interpret these findings, we estimate a model of endogenous skill acquisition where changes in the demand and supply of skills shape the path of earnings. Counterfactual simulations suggest that the endogenous skill response cannot fully undo the adverse earnings effects of automation unless there are sufficiently generous educational subsidies. (JEL I26, J22, J23, J24, J31)

Discussions of the future of jobs are concerned that technological change will displace labor and particularly jobs. In this work it is rarely remarked on the strangeness that some of the most globally scarce factors of high level technical expertise, capability to innovate, and entrepreneurial talent are devoted to economizing on--reducing the demand for--one of the most globally abundant factors: low to medium skill labor. I show that policy based barriers to the mobility of labor have created the largest single price distortion in history and that this price distortion induces biased technological change.Length: 36

ENDOGENOUS SKILL BIASED TECHNICAL CHANGE: TESTING FOR DEMAND PULL EFFECT* Abstract In this article we use the unification of Germany in 1990 to test the hypothesis that an increase in the supply of a production factor generates skill biased technical change. We test for this mechanism in the context of the model presented by Acemoglu and Autor (2011) that allows endogenous assignment of skills to tasks in the economy. We use cohorts of workers from comparable countries as a control group. After discussing the possible confounding factors, we conclude that this effect is absent. The differential pattern among the countries seems to be determined by labor market flexibilization and tax reform. SESGO ENDÓGENO DEL CAMBIO TÉCNICO HACIA LAS CALIFICACIONES: TESTEANDO EL EFECTO DE JALÓN DE LA DEMANDA Resumen En este artículo usamos la unificación de Alemania en 1990 para probar la hipótesis de que un incremento en la oferta de un factor de producción genera cambio técnico sesgado hacia el trabajo calificado. Probamos este mecanismo en el contexto del modelo presentado por Acemoglu y Autor (2011) que contempla la asignación endógena entre calificaciones y tareas dentro de la economía. Utilizamos cohortes de trabajadores de países comparables como grupo de control. Después de discutir los posibles factores de confusión, se concluye que este efecto está ausente. El patrón diferencial entre los países parece estar determinada por la flexibilización del mercado laboral y la reforma fiscal. * A version of this article is forthcoming on Industrial and Corporate Change. We wish to thank our research assistants, T. van der Veen and Laura Jiménez and the GINI team. A special acknowledgement to G. Corneo, V. Maestri, B. Nolan, M. Piva, P. Vanin and M. Vivarelli for discussion, to all those who attended the GINI workshop in Amsterdam, and to the participants of seminars at AIAS (Amsterdam) and Frei Universiteit (Berlin). The authors acknowledge comments by two anonymous referees, whose suggestions have helped to improve the previous version of this paper. The usual disclaimer applies.

This paper presents a stylised framework to examine how skill-biased technological \nchange and labour market frictions affect the relationship between economic \nexpansion and unskilled unemployment. The first part of the analysis focuses on the \ninvestment decisions in skill-acquisition and technology adoption activities faced by \nworkers and firms in response to the introduction of an innovative technology. The \nsecond part examines how endogenous two-sided heterogeneity in the labour market \naffects the macroeconomic outcomes in terms of unemployment, technological diffusion, and economic expansion. To conclude, the framework is used to discuss the effects of alternative forms of policy intervention on agents' investment decisions and on \nthe macroeconomic outcomes.

We propose a dynamic model of economic transition in which the supply constraint of skilled labor and skill premium are the focus. We argue that the constraint of skilled labor affect both the beginning date and the subsequent path of modern growth. The model matches the observed multiple paths of income inequality, such as U-shaped, inverted U-shaped or N-shaped paths. Hence, the model requires faster technology change and more investment on skill formation to account for the current income differences relative to models that focus only on steady states.

We develop a model of endogenous skill-biased technical change in developing countries. The endogenous response to a rise in skill supply counters the traditional substitution effect and dampens its role in reducing wage inequality. The model re-enforces consensus estimates of the elasticity of substitution between more/less educated workers by reconciling dispersed existing estimates. It also rationalizes estimates that were hitherto deemed implausible or model-inconsistent. We produce new estimates for developing countries with a novel global panel (finding values at or just above 2) and with Latin American data that facilitates analysis of dynamics (which reduce estimates to 1.7-1.8). We therefore shed new light on a parameter that is crucial for inequality, growth, and other key macroeconomic questions.

In this paper, we empirically address the hypothesis that there is a relationship between the supply of human capital and the rate and direction of skill-biased technical change (SBTC). Using country- and industry-level data on OECD countries, we find R&D to be positively related to the supply of human capital. There is, however, no indication that this translates into higher rates of SBTC, when SBTC is measured as changes in the wage bill share of skilled labour. Interestingly, both R&D and the rate of SBTC seem to be relatively high in low-skill industries in countries where the supply of human capital is relatively high.

We contribute to the debate on skill-biased technical change by studying the long-run dynamics of skill premia in an endogenous growth model in which technical change can be directed towards different factors. We show that R&D resources tend to be directed alternately towards skill-intensive and unskilled-intensive goods, creating cycles in skill premia. If resources were constantly directed towards the same sector, an innovation in a different sector would not be threatened by future innovators. Hence, researchers are incited to switch from one sector to another, in order to avoid the negative effect of innovations constantly occurring in the same sector.

Abstract We develop a simple growth model featuring individuals’ choices between general and specific skills, endogenous technological innovation, and a government subsidy for education. The two types of skills differ by their productivity and transferability: general skills are transferable across firms, while each firm-specific skill has a productivity advantage in the firm. Firms face uncertainty in their innovation activities, and the resulting heterogeneity in their labor demand makes the transferability of general skill valuable. We theoretically show that as a country catch up to the world technology frontier, firms invest more in innovation activities. This rises firms’ technological uncertainty and, thus, their demands for general skills increases. As a result, especially in more advanced economies, education subsidies may enhance GDP by increasing the supply of general skills. Using aggregated data for 12 European OECD counties, we calibrate the model and compare the theoretical prediction with the data. In cross-country comparisons, we find that the returns on general skills and the impact of general education expenditure on GDP are higher in countries with higher total factor productivity. These findings support our theoretical argument of the positive relationship between firms’ demand for general skills and countries’ stages of development.

Apprenticeships play a key role in enabling successful school‐to‐work transitions in many countries, but in the presence of imperfect information, the specificity of this type of training may entail important costs for those working outside their training fields. I study this issue in one of the most prominent training settings, the German apprenticeship system. Using administrative data and a broad occupational classification, I find that 40% of individuals work in occupations different from their training. I estimate the cost of mismatch using vacancy instruments and extend methodological approaches in high‐dimensional selection settings. Lacking training in one's occupation entails an average wage penalty of 14%, the equivalent of two years of work experience. The penalty increases with the task distance between training and occupation. My findings suggest that retraining is crucial to mitigate the adverse consequences from imperfect information in specialized training settings.

We construct a novel dataset linking academic publication records to U.S. Census employeremployee data to track 42,000 AI researchers over two decades.We document systematic changes in the allocation of AI talent.Industry increasingly attracts younger and foreign-born researchers, while gender representation improves more in academia.The top 1% of publishing industry scientists now earn $1.5 million more annually than comparable academics, a fivefold increase since 2001.Rising wage premia coincide with greater sorting into large incumbent firms.Researchers who move to industry publish less but patent more, consistent with a shift from open science toward proprietary innovation.

Abstract Research Summary This paper examines how firms' adoption of artificial intelligence (AI) relates to the demand for managers and managerial skills. Using a skills‐based measure of AI adoption derived from Lightcast job postings, we show that firms with greater AI adoption post more managerial vacancies and a higher share of such vacancies than less intensive adopters. These relationships are strongest in manufacturing and among firms with higher research & development intensity. Greater AI adoption is also associated with shifts in managerial skill requirements toward interpersonal and growth‐oriented skills, including stakeholder management, creativity, and sales management, and away from routine administrative skills such as budgeting, planning, staff management, and customer service. Overall, the results suggest a reconfiguration of managerial roles toward capabilities facilitating scaling, coordination, and adaptation in AI‐enabled environments. Managerial Summary As artificial intelligence (AI) becomes more prevalent within firms, managers and executives face a practical question about how managerial roles may change. Using US job postings data from 2010 to 2022, we find that firms with higher AI adoption exhibit relatively greater demand for managerial roles, especially in manufacturing and among more innovative firms. We also find that more intensive AI adoption is associated with changes in what managers are expected to do. Demand shifts away from routine administrative skills such as budgeting and planning and toward growth‐related skills such as sales, creativity, and stakeholder management. Overall, the evidence suggests a growing emphasis on managerial roles that relate to scaling, coordination, and organizational adaptation.

Generative AI is directional: it performs well in some task directions and poorly in others. Knowledge work is directional and endogenous as well: workers can satisfy the same job requirements with different mixes of tasks. We develop a high-dimensional model of AI adoption in which a worker uses a tool when it raises their output. Both the worker and the AI tool can perform a variety of tasks, which we model as convex production possibility sets. Because the tool requires supervision from the worker's own time and attention budget, adoption is a team-production decision, similar to hiring a coworker. The key sufficient statistics are the worker's pre-AI shadow prices: these equal the output gain from a small relaxation in each task direction, and they generally differ from the worker's observed activity mix. As AI capability improves, the set of adopted directions expands in a cone centered on these autarky prices. Near the entry threshold, small capability improvements generate large extensive-margin expansions in adoption. The model also delivers a structured intensive margin: between the entry and all-in thresholds, optimal use is partial. We parametrize the model in a simple but flexible way that nests most existing task-based models of technical change.

Generative AI is directional: it performs well in some task directions and poorly in others. Knowledge work is directional and endogenous as well: workers can satisfy the same job requirements with different mixes of tasks. We develop a high-dimensional model of AI adoption in which a worker uses a tool when it raises their output. Both the worker and the AI tool can perform a variety of tasks, which we model as convex production possibility sets. Because the tool requires supervision from the worker's own time and attention budget, adoption is a team-production decision, similar to hiring a coworker. The key sufficient statistics are the worker's pre-AI shadow prices: these equal the output gain from a small relaxation in each task direction, and they generally differ from the worker's observed activity mix. As AI capability improves, the set of adopted directions expands in a cone centered on these autarky prices. Near the entry threshold, small capability improvements generate large extensive-margin expansions in adoption. The model also delivers a structured intensive margin: between the entry and all-in thresholds, optimal use is partial. We parametrize the model in a simple but flexible way that nests most existing task-based models of technical change.

This paper studies how individuals sort into entrepreneurship and invention-related occupations and how their interactions shape innovation and economic growth. We develop an endogenous growth model in which occupational sorting jointly determines the supply of R&D talent and entrepreneurs’ demand for it. Empirically, using Danish microdata, we show that transformative entrepreneurs—those who hire R&D workers—tend to have higher IQ and education and build faster-growing firms than other entrepreneurs. Quantitatively, the estimated model indicates that financial barriers to education misallocate talent; alleviating them through education subsidies increases both demand and supply of R&D workers, raising innovation and long-run growth. Broad startup subsidies are ineffective.

This positional contribution has a twofold aim: the first is to explore the recent empirical literature developed around the issue of how the adoption of new technologies within the firm has changed the skill requirements of occupations; the second is to conjecture on the relationship, and on the relative sign, between technology adoption and firm sponsored onthe- job training. The basic idea is that the time-consuming dimension of the adoption process plays a direct role both in determining the profitability of the investment in new technology and in assessing the size of the productivity slowdown the firm eventually occurs after its introduction. On the extent that the timing of adoption depends on the workers’ skill composition and on the distance between the skills acquired for the job and the skills required by the job, the deep understanding of the interplay between the mechanisms of human capital accumulation can be helpful in order for the firm to set suitable and efficient job-training strategies. During the last two decades the discussion around the impact of technological change on workers’ human capital has been intense: the rapid diffusion of information and communication technologies (ICT) and computer-based machines (CNC, CAD), together with the large increase in the supply of highly-educated workers and rising returns to education, favoured the argument that technological change is characterized by a skill-biased nature (SBTC), leading to substantial changes in the division of labour and shifting labor demand towards employees with higher levels of education. On this purpose, different approaches have developed in the last decades that provide different evidence to a common research question. While a lot of national and international evidence still continues to support the SBTC hypothesis by employing ‘traditional’ aggregate measures of technological change and indirect measures of skill upgrading, a smaller literature is emerging that considers the heterogeneity of both technologies and skills at the workplace and aims at determining the demand of skills by the tasks occupations require. Even if new and interesting results emerge, many ‘black holes’ still remain, the most important of which seem to be the lack of theoretical and empirical models analyzing the role that school education and on-the-job training, and their interplay, can play in reducing the timing of new technology adoption.

<div> This paper documents degree completion patterns in programs associated with occupations highly exposed to generative AI and examines the educational background composition of workers in these occupations. Controlling for school-level shocks, event-study estimates show that by 2024 the number of bachelor’s degrees conferred in programs associated with high occupational AI exposure increases by 5% relative to programs associated with low occupational AI exposure. Programs linked to occupations where AI is more likely to automate tasks also experience a significant increase in bachelor’s degrees conferred. This increase is most pronounced at public and R1 institutions. So far, occupations with high AI exposure do not exhibit significant changes in the composition of workers across education levels or fields of study. These findings complement existing studies of AI’s employment effects and provide a more comprehensive understanding of AI’s impact on young workers. </div>

Abstract We investigate the impact of the academic environment on directed technical change and economic growth. We develop a heterogeneous Schumpeterian growth model in which innovation is categorized into radical and incremental types. A supportive environment for academic exploration enhances scientists’ autonomy utility in basic research, thereby motivating basic research and reducing the R&D difficulty of radical innovations through knowledge spillovers. We identify two major effects of the academic environment on economic growth: a positive directed technical change effect fostering growth through radical innovation, and a negative applied research crowding-out effect. Numerical analysis based on Chinese data reveals a negative autonomy utility (−0.37), indicating insufficient autonomy in basic research exploration. Promoting economic growth necessitates institutional reforms. The optimal autonomy utility for maximizing growth is 0.80. Welfare analysis further shows that the optimal autonomy utility is 0.87 for basic research labor and 0.75 for non-basic research labor.

This study explores how allowing remote work affects firms' innovation activity. I investigate the effect of the significant shift toward working from home that occurred in response to the COVID-19 pandemic on the direction of innovation. I find that work from home has shifted the path of innovation toward technology that supports non–face-to-face communication. Using a sample of patent-holding firms, I document that firms that adopted work from home applied for more patents of non–face-to-face technologies than firms that did not offer work from home. The results are driven by smaller, younger firms in the ICT-related industry, suggesting that these firms had greater digital resilience and were better able to adjust to the unprecedented shock to the working environment. I discuss how these facts contribute to our understanding of the impact of work from home on the direction of innovation, and how we may design policy responses to future shocks.

We study the supply-side direction of innovation in human enhancement technologies whose effects are inheritable across generations or confined to the treated individual. In a three-generation framework with heterogeneous agents, binding budget constraints, uncertainty about outcomes, and the possibility of imitation by entrants, inheritability creates a distinctive appropriability problem: a single purchase generates benefits for future descendants, but firms cannot charge those future beneficiaries. This intergenerational non-appropriability can reverse the private ranking of projects relative to their social value, leading decentralized research to favor non-inheritable enhancements even when inheritable ones maximize welfare, under high risk and limited entry to tilt excessively toward inheritability. We characterize equilibrium timing, the conditions under which innovation is inefficiently directed, and how inequality shapes adoption and profits. We also extend the model to competing researchers and show that standard congestion effects do not eliminate the core wedge between private incentives and social efficiency.

Strategy and innovation scholars have long emphasized the positive role of market demand in driving innovation within a technological domain. This study sheds light on an indirect negative spillover effect of market demand on technological progress: whereas increased downstream market demand within a domain generally drives increased technological progress in that domain (i.e., the demand-relevant domain), it may also adversely affect the technological progress of firms in adjacent domains. This occurs because the increased technological progress within the demand-relevant domain, driven by the downstream market demand, can intensify competition for skilled knowledge workers—a critical innovation resource whose supply is often inelastic in the short term. Empirically, I test these arguments by exploiting an unexpected environmental policy shock—the zero emission vehicle (ZEV) mandate—which led to an exogenous increase in demand for electric vehicle (EV) technologies. Following the ZEV mandate, I find evidence of increased inventive activities in the EV domain by EV firms. However, firms in adjacent (non-EV) domains were more likely to lose knowledge workers to EV firms following the ZEV mandate. Consequently, these affected firms produced 22% fewer inventions, particularly in their core technological areas, and became 19% less likely to explore new technological areas. Notably, affected firms in growing technological domains, such as renewable energy, and smaller, younger firms were more adversely (or at least equally) impacted. Supplemental Material: The online appendix is available at https://doi.org/10.1287/orsc.2023.18181 .

Growth in this model is driven by technological change that arises from intentional investment decisions made by profit-maximizing agents. The distinguishing feature of the technology as an input is that it is neither a conventional good nor a public good; it is a nonrival, partially excludable good. Because of the nonconvexity introduced by a nonrival good, price-taking competition cannot be supported. Instead, the equilibrium is one with monopolistic competition. The main conclusions are that the stock of human capital determines the rate of growth, that too little human capital is devoted to research in equilibrium, that integration into world markets will increase growth rates, and that having a large population is not sufficient to generate growth.

We apply an understanding of what computers do to study how computerization alters job skill demands. We argue that computer capital (1) substitutes for workers in performing cognitive and manual tasks that can be accomplished by following explicit rules; and (2) complements workers in performing nonroutine problem-solving and complex communications tasks. Provided that these tasks are imperfect substitutes, our model implies measurable changes in the composition of job tasks, which we explore using representative data on task input for 1960 to 1998. We find that within industries, occupations, and education groups, computerization is associated with reduced labor input of routine manual and routine cognitive tasks and increased labor input of nonroutine cognitive tasks. Translating task shifts into education demand, the model can explain 60 percent of the estimated relative demand shift favoring college labor during 1970 to 1998. Task changes within nominally identical occupations account for almost half of this impact.

A central organizing framework of the voluminous recent literature studying changes in the returns to skills and the evolution of earnings inequality is what we refer to as the canonical model, which elegantly and powerfully operationalizes the supply and demand for skills by assuming two distinct skill groups that perform two different and imperfectly substitutable tasks or produce two imperfectly substitutable goods. Technology is assumed to take a factor-augmenting form, which, by complementing either high or low skill workers, can generate skill biased demand shifts. In this paper, we argue that despite its notable successes, the canonical model is largely silent on a number of central empirical developments of the last three decades, including: (1) significant declines in real wages of low skill workers, particularly low skill males; (2) non-monotone changes in wages at different parts of the earnings distribution during different decades; (3) broad-based increases in employment in high skill and low skill occupations relative to middle skilled occupations (i.e., job "polarization"); (4) rapid diffusion of new technologies that directly substitute capital for labor in tasks previously performed by moderately skilled workers; and (5) expanding offshoring in opportunities, enabled by technology, which allow foreign labor to substitute for domestic workers specific tasks. Motivated by these patterns, we argue that it is valuable to consider a richer framework for analyzing how recent changes in the earnings and employment distribution in the United States and other advanced economies are shaped by the interactions among worker skills, job tasks, evolving technologies, and shifting trading opportunities. We propose a tractable task-based model in which the assignment of skills to tasks is endogenous and technical change may involve the substitution of machines for certain tasks previously performed by labor. We further consider how the evolution of technology in this task-based setting may be endogenized. We show how such a framework can be used to interpret several central recent trends, and we also suggest further directions for empirical exploration. © 2011 Elsevier B.V.

We present a framework for understanding the effects of automation and other types of technological changes on labor demand, and use it to interpret changes in US employment over the recent past. At the center of our framework is the allocation of tasks to capital and labor—the task content of production. Automation, which enables capital to replace labor in tasks it was previously engaged in, shifts the task content of production against labor because of a displacement effect. As a result, automation always reduces the labor share in value added and may reduce labor demand even as it raises productivity. The effects of automation are counterbalanced by the creation of new tasks in which labor has a comparative advantage. The introduction of new tasks changes the task content of production in favor of labor because of a reinstatement effect, and always raises the labor share and labor demand. We show how the role of changes in the task content of production—due to automation and new tasks—can be inferred from industry-level data. Our empirical decomposition suggests that the slower growth of employment over the last three decades is accounted for by an acceleration in the displacement effect, especially in manufacturing, a weaker reinstatement effect, and slower growth of productivity than in previous decades.

This essay discusses the effect of technical change on wage inequality. I argue that the behavior of wages and returns to schooling indicates that technical change has been skill-biased during the past sixty years. Furthermore, the recent increase in inequality is most likely due to an acceleration in skill bias. In contrast to twentieth-century developments, much of the technical change during the early nineteenth century appears to be skill-replacing. I suggest that this is because the increased supply of unskilled workers in the English cities made the introduction of these technologies profitable. On the other hand, the twentieth century has been characterized by skill-biased technical change because the rapid increase in the supply of skilled workers has induced the development of skill-complementary technologies. The recent acceleration in skill bias is in turn likely to have been a response to the acceleration in the supply of skills during the past several decades.

The recent rise in wage inequality is usually attributed to skill-biased technical change (SBTC), associated with new computer technologies. We review the evidence for this hypothesis, focusing on the implications of SBTC for overall wage inequality and for changes in wage differentials between groups. A key problem for the SBTC hypothesis is that wage inequality stabilized in the 1990s despite continuing advances in computer technology; SBTC also fails to explain the evolution of other dimensions of wage inequality, including the gender and racial wage gaps and the age gradient in the return to education.

Empirical work has been limited in its ability to directly study whether skill requirements in the workplace have been rising and whether these changes have been related to technological change. This article answers these questions using a unique data set from West Germany that enabled me to look at how skill requirements have changed within occupations. I show that occupations require more complex skills today than in 1979 and that the changes in skill requirements have been most pronounced in rapidly computerizing occupations. Changes in occupational content account for about 36% of the recent educational upgrading in employment.

Current concern with relationships among particular technologies, capital, and the wage structure motivates this study of the origins of technology-skill complementarity in manufacturing. We offer evidence of the existence of technology-skill and capital-skill (relative) complementarities from 1909 to 1929, and suggest that they were associated with continuous-process and batch methods and the adoption of electric motors. Industries that used more capital per worker and a greater proportion of their horsepower in the form of purchased electricity employed relatively more educated blue-collar workers in 1940 and paid their blue-collar workers substantially more from 1909 to 1929. We also infer capital-skill complementarity using the wage-bill for non-production workers and find that the relationship was as large from 1909-19 as it has been recently. Finally, we link our findings to those on the high-school movement (1910 to 1940). The rapid increase in the supply of skills from 1910 to 1940 may have prevented rising inequality with technological change.

In the human capital model with perfect labor markets, firms never invest in general skills and all cost of general training are borne by workers. When lobor market frictions compress the structure of wages, firms may pay for these investments. The distortion in the wage structure turns "technologically" general skills into de facto "specific " skills. Credit market imperfections are neither neccessary nor sufficient for firm‐sponsored training. Since labor market frictions and insititutions shape the wage structure, they may have an important impact on the financing and amount of human capital investments and account for some international differences in training practices.

We study the use and economic impact of AI technologies. We propose a new measure of firm-level AI investments using employee resumes. Our measure reveals a stark increase in AI investments across sectors. AI-investing firms experience higher growth in sales, employment, and market valuations. This growth comes primarily through increased product innovation. Our results are robust to instrumenting AI investments using firms' exposure to universities' supply of AI graduates. AI-powered growth concentrates among larger firms and is associated with higher industry concentration. Our results highlight that new technologies like AI can contribute to growth and superstar firms through product innovation.

Abstract We test the hypothesis that information and communication technologies (ICT) polarize labor markets by increasing demand for the highly educated at the expense of the middle educated, with little effect on low-educated workers. Using data on the United States, Japan, and nine European countries from 1980 to 2004, we find that industries with faster ICT growth shifted demand from middle-educated workers to highly educated workers, consistent with ICT-based polarization. Trade openness is also associated with polarization, but this is not robust to controlling for R&D. Technologies account for up to a quarter of the growth in demand for highly educated workers.

This study focuses on the risk of automation and its interaction with training and the use of skills at work. Building on the expert assessment carried out by Carl Frey and Michael Osborne in 2013, the paper estimates the risk of automation for individual jobs based on the Survey of Adult Skills (PIAAC). The analysis improves on other international estimates of the individual risk of automation by using a more disaggregated occupational classification and identifying the same automation bottlenecks emerging from the experts' discussion. Hence, it more closely aligns to the initial assessment of the potential automation deriving from the development of Machine Learning. Furthermore, this study investigates the same methodology using national data from Germany and United Kingdom, providing insights into the robustness of the results. The risk of automation is estimated for the 32 OECD countries that have participated in the Survey of Adult Skills (PIAAC) so far. Beyond the share of jobs likely to be significantly disrupted by automation of production and services, the accent is put on characteristics of these jobs and the characteristics of the workers who hold them. The risk is also assessed against the use of ICT at work and the role of training in helping workers transit to new career opportunities.

Many technologies used by the LDCs are developed in the OECD economies and are designed to make optimal use of the skills of these richer countries' workforces. Differences in the supply of skills create a mismatch between the requirements of these technologies and the skills of LDC workers, and lead to low productivity in the LDCs. Even when all countries have equal access to new technologies, this technology-skill mismatch can lead to sizable differences in total factor productivity and output per worker. We provide evidence in favor of the cross-industry productivity patterns predicted by our model, and also show that technology-skill mismatch could account for a large fraction of the observed output per worker differences in the data.

This paper develops a model to analyse how skill premia differ over time and across countries, and uses this model to study the impact of international trade on wage inequality. Skill premia are determined by technology, the relative supply of skills, and trade. Technology is itself endogenous, and responds to profit incentives. An increase in the relative supply of skills, holding technology constant, reduces the skill premium. But an increase in the supply of skills over time also induces a change in technology, increasing the demand for skills. The most important result of the paper is that increased international trade induces skill-biased technical change. As a result, trade opening can cause a rise in inequality both in the U.S. and the less developed countries, and thanks to the induced skill-biased technical change, this can happen without a rise in the relative prices of skill-intensive goods in the U.S., which is the usual intervening mechanism in the standard trade models.

We investigate the potential implications of large language models (LLMs), such as Generative Pre-trained Transformers (GPTs), on the U.S. labor market, focusing on the increased capabilities arising from LLM-powered software compared to LLMs on their own. Using a new rubric, we assess occupations based on their alignment with LLM capabilities, integrating both human expertise and GPT-4 classifications. Our findings reveal that around 80% of the U.S. workforce could have at least 10% of their work tasks affected by the introduction of LLMs, while approximately 19% of workers may see at least 50% of their tasks impacted. We do not make predictions about the development or adoption timeline of such LLMs. The projected effects span all wage levels, with higher-income jobs potentially facing greater exposure to LLM capabilities and LLM-powered software. Significantly, these impacts are not restricted to industries with higher recent productivity growth. Our analysis suggests that, with access to an LLM, about 15% of all worker tasks in the US could be completed significantly faster at the same level of quality. When incorporating software and tooling built on top of LLMs, this share increases to between 47 and 56% of all tasks. This finding implies that LLM-powered software will have a substantial effect on scaling the economic impacts of the underlying models. We conclude that LLMs such as GPTs exhibit traits of general-purpose technologies, indicating that they could have considerable economic, social, and policy implications.

We document that between 50% and 70% of changes in the U.S. wage structure over the last four decades are accounted for by relative wage declines of worker groups specialized in routine tasks in industries experiencing rapid automation. We develop a conceptual framework where tasks across industries are allocated to different types of labor and capital. Automation technologies expand the set of tasks performed by capital, displacing certain worker groups from jobs for which they have comparative advantage. This framework yields a simple equation linking wage changes of a demographic group to the task displacement it experiences. We report robust evidence in favor of this relationship and show that regression models incorporating task displacement explain much of the changes in education wage differentials between 1980 and 2016. The negative relationship between wage changes and task displacement is unaffected when we control for changes in market power, deunionization, and other forms of capital deepening and technology unrelated to automation. We also propose a methodology for evaluating the full general equilibrium effects of automation, which incorporate induced changes in industry composition and ripple effects due to task reallocation across different groups. Our quantitative evaluation explains how major changes in wage inequality can go hand‐in‐hand with modest productivity gains.

Motivated by the large differences in labor market outcomes across college majors, we survey the literature on the demand for and return to high school and postsecondary education by field of study. We combine elements from several papers to provide a dynamic model of education and occupation choice that stresses the roles of the specificity of human capital and uncertainty about preferences, ability, education outcomes, and labor market returns. The model implies an important distinction between the ex ante and ex post returns to education decisions. We also discuss some of the econometric difficulties in estimating the causal effects of field of study on wages in the context of a sequential choice model with learning. Finally, we review the empirical literature on the choice of curriculum and the effects of high school courses and college major on labor market outcomes.

This paper explores a one-agent Bayesian model of learning by doing and technological choice.To produce output, the agent can choose among various technologies.The beneficial effects of learning by doing are bounded on each technology, and so long-run growth in output can take place only if the agent repeatedly switches to better technologies.As the agent repeatedly uses a technology, he learns about its unknown parameters, and this accumulated expertise is a form of human capital.But when the agent switches technologies, part of this human capital is lost.It is this loss of human capital that may prevent the agent from moving up the quality ladder of technologies as quickly as he can, since the loss is greater the bigger is the technological leap.We analyze the global dynamics.We find that a human-capital-rich agent may find it optimal to avoid any switching of technologies, and therefore to experience no long-run growth.On the other hand, a human-capital-poor agent, who because of his lack of skill is not so attached to any particular technology, can find it optimal to switch technologies repeatedly, and therefore enjoy long-run growth in output.Thus the model can give rise to overtaking.

There have been striking postwar changes in the supply and price of skilled labor relative to unskilled labor. The relative quantity of skilled labor has increased substantially, and the skill premium, which is the wage of skilled labor relative to unskilled labor, has grown significantly since 1980. Many studies have found that it is difficult to account for the increase in the skill premium on the basis of observable variables and have concluded that latent &quot;skill-biased technological change&quot; is the main factor responsible for the increase. This paper develops a framework that provides a simple, explicit economic mechanism for understanding skill-biased technological change in terms of observable variables and uses the framework to evaluate the fraction of variation in the skill premium that can be accounted for by changes in observed factor quantities. We use a version of the neoclassical growth model in which the key feature of the aggregate technology is capital-skill complementar...

Conventional wisdom holds that the social rate of return to R&D significantly exceeds the private rate of return and, therefore, R&D should be subsidized. In the U.S., the government has directly funded a large fraction of total R&D spending. This paper shows that there is a serious problem with such government efforts to increase inventive activity. The majority of R&D spending is actually just salary payments for R&D workers. Their labor supply, however, is quite inelastic so when the government funds R&D, a significant fraction of the increased spending goes directly into higher wages. Using CPS data on wages of scientific personnel, this paper shows that government R&D spending raises wages significantly, particularly for scientists related to defense such as physicists and aeronautical engineers. Because of the higher wages, conventional estimates of the effectiveness of R&D policy may be 30 to 50% too high. The results also imply that by altering the wages of scientists and engineers even for firms not receiving federal support, government funding directly crowds out private inventive activity.

Abstract The current wave of technological change based on advancements in artificial intelligence (AI) has created widespread fear of job loss and further rises in inequality. This paper discusses the rationale for these fears, highlighting the specific nature of AI and comparing previous waves of automation and robotization with the current advancements made possible by a widespread adoption of AI. It argues that large opportunities in terms of increases in productivity can ensue, including for developing countries, given the vastly reduced costs of capital that some applications have demonstrated and the potential for productivity increases, especially among the low skilled. At the same time, risks in the form of further increases in inequality need to be addressed if the benefits from AI-based technological progress are to be broadly shared. For this, skills policies are necessary but not sufficient. In addition, new forms of regulating the digital economy are called for that prevent further rises in market concentration, ensure proper data protection and privacy, and help share the benefits of productivity growth through the combination of profit sharing, (digital) capital taxation, and a reduction in working time. The paper calls for a moderately optimistic outlook on the opportunities and risks from AI, provided that policymakers and social partners take the particular characteristics of these new technologies into account.

Abstract This article studies the impact of changing job skills on career earnings dynamics for college graduates. We measure changes in the skill content of occupations between 2007 and 2019 using detailed job descriptions from a near universe of online job postings. We then develop a simple model where the returns to work experience are a race between on-the-job learning and skill obsolescence. Obsolescence lowers the return to experience, flattening the age-earnings profile in faster-changing careers. We show that the earnings premium for college graduates majoring in technology-intensive subjects such as computer science, engineering, and business declines rapidly, and that these graduates sort out of faster-changing occupations as they gain experience.

This paper presents three sets of results about equilibrium bias of technology. First, I show that when the menu of technological possibilities only allows for factor-augmenting technologies, the increase in the supply of a factor induces technological change relatively biased toward that factor—meaning that the induced technological change increases the relative marginal product of the factor becoming more abundant. Moreover, this induced bias can be strong enough to make the relative marginal product of a factor increasing in response to an increase in its supply, thus leading to an upward-sloping relative demand curve. I also show that these results about relative bias do not generalize when more general menus of technological possibilities are considered. Second, I prove that under mild assumptions, the increase in the supply of a factor induces technological change that is absolutely biased toward that factor—meaning that it increases its marginal product at given factor proportions. The third and most important result in the paper establishes the possibility of and conditions for strong absolute equilibrium bias—whereby the price (marginal product) of a factor increases in response to an increase in its supply. I prove that, under some regularity conditions, there will be strong absolute equilibrium bias if and only if the aggregate production function of the economy fails to be jointly concave in factors and technology. This type of failure of joint concavity is possible in economies where equilibrium factor demands and technologies result from the decisions of different agents.

We develop a model where the allocation of human resources, intergenerational social mobility, and technological growth are jointly determined. High growth endogenously increases the equilibrium return to innate cognitive ability and makes the allocation of individuals depend more on innate ability and less on social background. Individuals with a higher level of innate cognitive ability can deal better with less known, but more productive, technologies and thus choose a higher rate of technological growth. A social allocation based on innate ability and high growth will thus reinforce each other, implying the possibility of multiple endogenous growth equilibria. (JEL J62, O1)

Abstract Human capital accumulation is introduced in a growth model with R&D‐driven expansion in variety and quality of intermediate goods and knowledge spillovers from both research activities. Economic growth is no longer uniquely tied to population growth as previous growth models without scale effects suggest. The model predicts that economic growth depends positively on the rate of human capital accumulation and positively or negatively on population growth and is therefore supported by empirical evidence to a greater extent than previous models. In particular, long‐run growth is compatible with a stable population.

While the importance of innovation to economic development is widely understood, the conditions conducive to it remain the focus of much attention. This volume offers new contributions to fundamental questions relating to the economics of innovation and technological change. Central to the development of new technologies are institutional environments, and among the topics discussed are the roles played by universities and other nonprofit research institutions and the ways in which the allocation of funds between the public and private sectors affects innovation. Other essays examine the practice of open research and how the diffusion of information technology influences knowledge accumulation.

This paper consolidates two previously disconnected literatures. It integrates R&D-based innovations into a unified growth framework with micro-founded fertility and schooling behavior. The theory suggests a refined view on the human factor in productivity growth. It helps to explain the historical emergence of R&D-based growth and the subsequent emergence of mass education and the demographic transition. The model predicts that the erstwhile positive correlation between population growth and innovative activity turns negative during economic development. This "population-productivity reversal" explains why innovative modern economies are usually characterized by low or negative population growth. Because innovations in modern economies are based on the education of the workforce, the medium-run prospects for future economic growth-when fertility is going to be below replacement level in virtually all developed countries-are better than suggested by conventional R&D-based growth theories. © 2013 Springer Science+Business Media New York.

We examine an endogenous growth model in which market frictions are an integral part of the economic environment. Workers invest in education when young, which raises their productivity once employed. The level of schooling also acts as a key determinant of the rate of economic growth by influencing workers' ability to accumulate additional human capital on-the-job. Once schooling is completed, workers search for employment. The division of the surplus between vacancies and searching workers is characterized, as is the optimal level of education. The economy may display multiple steady-state growth paths.

Zvi Griliches was a modern master of empirical economics. In this short book, he recounts what he and others have learned about sources of economic growth. This book conveys way he tackled research problems. For Griliches, economic theorizing without measurement is merely fashioning of parables, but measurement without theory is blind. Judgement enables one to strike right balance. The book begins with economists' first attempts to productivity growth systematically in 1930s. In mid-1950s these efforts culminated in a startling puzzle. The growth of measured inputs like labour and capital explained only a fraction of growth of national output. Economists called this phenomenon efficiency or technical change or the residual. However, Griliches observes that most accurate name was a measure of our ignorance. What explained rest of economic growth quickly became one of most important questions in economics. Over next 30 years, Griliches and his colleagues and students looked for various components of residual in education (the formation of human capital), investment (the formation of physical capital), and research and development. In 1973, after oil price shocks, productivity growth slowed and residual almost disappeared. Since shocks were a short-term phenomenon, they could not account for slowdown. A main focus on this book is therefore puzzle of productivity slowdown and how to date it and how to explain it.

R&D-based models of growth predict an unrealistic degree of responsiveness of long-run growth rates to policy changes. The present paper "exogenizes" the equilibrium growth rate in the Grossman-Helpman model by endogenizing human capital along the lines proposed by Uzawa and Lucas: the pace of long-run growth is unaffected by R&D subsidies, flat-rate taxes, basic research, and - in highly developed countries - by cross-border knowledge spillovers and international trade. A complete dynamic analysis is performed.

Abstract We answer three core questions about the hypothesized role of newly emerging job categories (“new work”) in counterbalancing the erosive effect of task-displacing automation on labor demand: what is the substantive content of new work, where does it come from, and what effect does it have on labor demand? We construct a novel database spanning eight decades of new job titles linked to U.S. Census microdata and to patent-based measures of occupations’ exposure to labor-augmenting and labor-automating innovations. The majority of current employment is in new job specialties introduced since 1940, but the locus of new-work creation has shifted from middle-paid production and clerical occupations over 1940–1980 to high-paid professional occupations and secondarily to low-paid services since 1980. New work emerges in response to technological innovations that complement the outputs of occupations and demand shocks that raise occupational demand. Innovations that automate tasks or reduce occupational demand slow new-work emergence. Although the flow of augmentation and automation innovations is positively correlated across occupations, the former boosts occupational labor demand while the latter depresses it. The demand-eroding effects of automation innovations have intensified in the past four decades while the demand-increasing effects of augmentation innovations have not.

Abstract In 2018, Paul Romer and William Nordhaus shared the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel. Romer was recognized “for integrating technological innovations into long‐run macroeconomic analysis”. This article reviews his prize‐winning contributions. Romer, together with others, rejuvenated the field of economic growth. He developed the theory of endogenous technological change, in which the search for new ideas by profit‐maximizing entrepreneurs and researchers is at the heart of economic growth. Underlying this theory, he pinpointed that the nonrivalry of ideas is ultimately responsible for the rise in living standards over time.

We examine the role of technological change in explaining the large and persistent decline in earnings following job loss. Using detailed skill requirements from the near universe of online vacancies, we estimate technological change by occupation and find that technological change accounts for 45 percent of the decline in earnings after job loss. Technological change lowers earnings after job loss by requiring workers to have new skills to perform newly created jobs in their prior occupation. When workers lack the required skills, they move to occupations where their skills are still employable but are paid a lower wage. (JEL J24, J31, J63, O33)

The article presents a model of endogenous innovation and growth, in which technological change is path dependent. The historical pattern of technological development plays a central role in determining the pace of future technological change. Path dependence is explained using a distinction between fundamental and secondary knowledge. The economy moves endogenously between periods of drastic and nondrastic innovation. Technological lock‐in is shown to be a special case of path dependence. The model provides a rationale for cycles in technological leadership. This rationale exists in equilibria with positive levels of fundamental research and in a world with no imitation.

This paper develops a dynamic supply and demand model of occupational choice and applies it to the engineering profession. The model is largely successful in understanding data in the U.S. engineering labor market. The engineering market responds strongly to economic forces. The demand for engineers responds to the price of engineering services and demand shifters. More important, supply and enrollment decisions are remarkably sensitive to career prospects in engineering. Also a rational model, in which students use some forward‐looking elements to forecast future demand for engineers, fits the data reasonably well. These findings suggest that subsidies to build technical talent ahead of demand are misplaced unless public policy makers have better information on future market conditions than the market participants do.

This essay discusses the effect of technical change on wage inequality. I argue that the behavior of wages and returns to schooling indicates that technical change has been skill-biased during the past sixty years. Furthermore, the recent increase in inequality is most likely due to an acceleration in skill bias. In contrast to twentieth century developments, most technical change during the nineteenth century appears to be skill-replacing. I suggest that this is because the increased supply of unskilled workers in the English cities made the introduction of these technologies profitable. On the other hand, the twentieth-century has been characterized by skill-biased technical change because the rapid increase in the supply of skilled workers has induced the development of skill-complementary technologies. The recent acceleration in skill bias is in turn likely to have been a response to the acceleration in the supply of skills during the past several decades.

What determines the earnings of a worker relative to his peers in the same occupation? What makes a worker fail in one occupation but succeed in another? More broadly, what are the factors that determine the productivity of a worker-occupation match? To help answer questions like these, we propose an empirical measure of multidimensional skill mismatch that is based on the discrepancy between the portfolio of skills required by an occupation and the portfolio of abilities possessed by a worker for learning those skills. This measure arises naturally in a dynamic model of occupational choice and human capital accumulation with multidimensional skills and Bayesian learning about one’s ability to learn skills. Not only does mismatch depress wage growth in the current occupation, it also leaves a scarring effect—by stunting skill acquisition—that reduces wages in future occupations. Mismatch also predicts different aspects of occupational switching behavior. We construct the empirical analog of our skill mismatch measure from readily available US panel data on individuals and occupations and find empirical support for these implications. The magnitudes of these effects are large: moving from the worst- to best-matched decile can improve wages by 11 percent per year for the rest of one’s career. (JEL E24, J24, J31, J41)

Science and engineering skills as well as management and leadership skills are often referred to as sources of innovative activities within companies. Broken down into sectoral innovation patterns, this article examines the role of formal education, actual occupation and work experience in the innovation performance in manufacturing firms within a probit model. It uses unique micro data for Germany (LIAB) that contain information about corporate innovation activities and the qualification of employees in terms of formal education, actual professional status and work experience. We find clear differences in the human capital endowment between sectors according to the Pavitt classification. Sectors with a high share of highly skilled employees engage in above average product innovation (specialised suppliers and science-based industries). However, according to our estimation results, across as well as within these sectors a large share of highly skilled employees does not substantially increase the probability of a firm being innovative.

Engineers, as implementers of technology, are highly complementary to the intangible knowledge assets that firms accumulate. This paper seeks to address whether technical talent is a source of rents for corporate employers, both in general and in the specific case of the surprising open-source launch of TensorFlow, a deep learning software package, by Google. First, I present a simple model of how employers can use job design as a tool to exercise monopsony power by partially allocating employee time to firm-specific tasks. Then, using over 180 million position records and over 52 million skill records from LinkedIn, I build a panel of firm-level investment in technological human capital (information technology, research, and engineering talent quantities) to measure the market value of technological talent. I find that on average, an additional engineer at a firm is correlated with approximately $854,000 more market value. Firm fixed effects and instrumental variables analyses provide mixed evidence on the marginal causal value of engineers in general. Specifically for AI talent, the value of engineering skills is clearer. AI skills are strongly correlated with market value, though variation in AI skills from 2014-2017 does not explain contemporaneous revenue productivity within firms. AI-intensive companies rapidly gained market value following the launch of TensorFlow, while companies with opportunities to automate relatively larger quantities of labor with machine learning did not. Using a differencein- differences approach, I show that the TensorFlow launch is associated with an approximate market value increase of 4-7% for AI-using firms. Firms outside the top quintile of AI use (as measured by skill counts on LinkedIn) grow by approximately $3.56 million for a 1% increase in AI skill. AI superstar firms in the top quintile also appear to benefit, but show pre-trends in market value growth.

In this paper, we construct a parsimonious overlapping-generations model of human capital accumulation and study its quantitative implications for the evolution of the U.S. wage distribution from 1970 to 2000. A key feature of the model is that individuals differ in their ability to accumulate human capital, which is the main source of wage inequality in this model. We examine the response of this model to skill-biased technical change (SBTC), which is modeled as an increase in the trend growth rate of the price of human capital starting in the early 1970s. The model displays behavior that is consistent with several important trends observed in the US data, including the rise in overall wage inequality; the fall and subsequent rise in the college premium, as well as the fact that this behavior was most pronounced for younger workers; the rise in within-group inequality; the stagnation in median wage growth; and the small rise in consumption inequality despite the large rise in wage inequality. We consider different scenarios regarding how individuals' expectations evolve during SBTC. Specifically, we study the case where individuals immediately realize the advent of SBTC (perfect foresight), and the case where they initially underestimate the future growth of the price of human capital (pessimistic priors), but learn the truth in a Bayesian fashion over time. Lack of perfect foresight appears to have little effect on the main results of the paper. Overall, the model shows promise for explaining a diverse set of wage distribution trends observed since the 1970s in a unifying human capital framework.

Poorer US cities were catching up with richer ones at an annual rate of roughly 1.4% between 1940 and 1980. However, wage convergence across US cities went from 1.4% a year between 1940 and 1980 to 0% a year between 1980 and 2010. This paper quantifies the contributions of skill-biased technical change (SBTC) and agglomeration economies to the end of cross-cities wage convergence within the US between 1980 and 2010. I develop and estimate a dynamic spatial equilibrium model that looks at the causes of the decline in spatial wage convergence. The model choice is motivated by novel empirical regularities regarding the evolution of the skill premium and migration patterns over time and across space. The model successfully matches the quantitative features of the decline in US regional wage convergence, as well as other stylized facts on US economic growth. Moreover, the model also reproduces the convergence and the divergence in the skill ratio across US cities and other features on quantities, such as the secular decline in within US migration after 1980. Finally, the counterfactual analysis suggests that SBTC explains the approximately the 80% of the decline of regional convergence between 1980 and 2010 among high skill workers.

The paper tests whether technological change has been neutral in Canadian manufacturing industries, using a system of translog cost share equations for 1962 through 1986. The model features two classes of labor treated as distinct inputs. Tests rejected homotheticity in all industries. Hicks neutrality was also rejected in 16 of 18 industries. The most common pattern of nonneutral technical change was a bias away from blue-collar workers. Formal tests for skill-neutral innovation rejected the hypothesis in ten industries in favor of skill-using technical change. The results suggest that in studies of Canadian manufacturing, aggregation across labor inputs is inappropriate.

Abstract This paper surveys the last two and a half decades of non‐neoclassical literature on endogenous technical change and the functional income distribution. We distinguish between classical‐Marxian and post‐Keynesian models, and analyze them under three different assumptions on the determinants of technical change: capital accumulation, income distribution, and labor market tightness. The balanced growth implications of alternative models are compared with neoclassical exogenous and endogenous growth theories. Despite the strong differences in the assumptions regarding the substitutability between capital and labor, the role of different classes in society, and whether or not productive factors are fully employed, the various alternative models can be classified in a way that highlights remarkable similarities with their neoclassical counterparts. Both neoclassical and alternative theories of endogenous growth: (i) have shown that long‐run growth is sensitive to investment decisions, and (ii) rely on a linear spillover from the stock of knowledge to the production of innovations. The comparison highlights the different channels emphasized by competing theories: saving behavior and market structure in the neoclassical theories, as opposed to income distribution, the state of the labor market, and investors' behavior in alternative theories.

Many empirical studies document a positive correlation between workplace computerization and the employment of skilled labor in production.Does this mean that computers necessarily substitute for the tasks performed by less educated workers and complement the tasks performed by more educated workers?We explore this question by positing that computerization leads to the automation of tasks that can be fully described in terms of procedural or 'rules-based' logic.This process typically leaves many tasks to be performed by humans.Management decisions play a key role -at least in the short run -in determining how these tasks are organized into jobs, with potentially significant implications for skill demands.We illustrate how this conceptual framework helps to interpret the consequences of the introduction of digital check imaging in two back office departments of a large bank.We argue that the model has applicability to many organizations and helps to reconcile differences between the approaches economists and sociologists typically take to studying the consequences of technological changes.Two recent trends have rekindled interest in the questions of how technological change impacts the skills that workers use at their jobs and the way in which these skills are remunerated.The first is the increase in earnings inequality.Since 1980, the earnings of highly educated workers have increased relative to those with less education, a phenomenon that economists attribute primarily to large changes in relative demands for the skills supplied by these groups of workers.The second trend is the remarkable proliferation of computers and information technology, beginning with the spread of mainframe applications during the 1970s, moving to greater use of personal computers in the 1980s, and to enormous growth in applications of networked computers in the 1990s.A number of studies have documented positive correlations between the use of computers and the use of more educated labor in production across detailed industries and across plants within industries in the U.S. Similar relationships are present in data from other industrialized countries.Some analysts cite these correlations as evidence that computers embody skill-biased technical change, meaning that computers substitute for less educated workers in performing some tasks and complement more educated workers in performing other tasks.Other observers reject this conclusion as unduly deterministic.Based on analyses of case studies, they argue that equating computers with skill-based technical change ignores management's role in job and organizational design and relies on simplistic definitions of skill.In this paper, we argue that the introduction of computer-based technology creates strong economic pressure to substitute machinery for people in carrying out tasks that can be fully described in terms of procedural or 'rules-based' logic and hence performed by a computer.This process typically leaves many tasks to be performed by humans, and management decisions play a key roleat least in the short runin determining how these tasks are organized into jobs, with potentially significant implications for skill demands.We show how this model helps to interpret how the introduction of digital check imaging in two back office departments of a large bank, one (downstairs) department processing deposits, the other (upstairs) department handling exceptions (e.g.overdrafts or stop check requests).The introduction of check imaging led to the computerization of certain tasks in both departments.However, the re-organization of the remaining tasks differed across the two departments.In deposit processing, check imaging led to greater specialization in jobs; in exceptions processing, it led to the integration of tasks, with a resulting increase in the demand for problem-solving skills.We use ideas articulated by Autor, Levy and Murnane (2001), Lindbeck and Snower (2000), and Osterman (1994) to explain why the introduction of check imaging led managers to reorganize work so differently in the two departments.We begin with a brief discussion of the literature that bears on our case.We then turn to a discussion of what computers actually dothe execution of rules-based logic.This provides the background for describing and interpreting the evidence from our case study.Though we present only one case, we believe that the model -computers displacing humans in some tasks with management decisions reshaping othersoffers a potentially valuable framework for future studies of computerization's skill impacts. Previous ResearchThe explanation favored by many economists for the positive relationship between computer use and the demand for educated labor is computer-skill complementarity or skill-biased technological change.1The essence of this hypothesis is that technological change involving computers increases the productivity of highly educated workers more than it increases the productivity of less-educated workers.An alternative explanation popularized in books such as The End of Work (Rifkin, 1995) is computer-labor substitution: computers substitute for low skilled labor in carrying out a variety of tasks.Both of these explanations imply an increase in relative demand for highly skilled workers.Some social scientists, especially those trained in disciplines other than economics, find the concept of skill-biased technological change troubling.For example, in a thought-provoking article Paul Attewell (1990) pointed out that there are many different ways to think about the concept of skill.In one tradition (positivism), skill is treated as an attribute ofjobs, and jobs that are substantively complex are viewed as skilled.A difficulty here is the arbitrariness of ranking in a single dimension of complexity jobs that are very different, for example, conducting biological research versus managing a large organization.Perhaps all that these highly skilled jobs have in common is that both require a great deal of conscious thought.A different research tradition, ethnomethodology, sees a variety of human activities such as walking across a crowded room or carrying on a conversation with many voices in the background as extremely skilled tasks that humans learn to do without conscious thought, an insight validated by decades of research in artificial intelligence (Pinker, 1997).Consistent with this view, many of these "simple" activities cannot be performed by computers, a point we return to below.These very different conceptions of skill raise questions about exactly what economists mean when they refer to "skill-biased technological change" and what the predictions are about

In an analysis connecting labor market earnings to college major choices, we find statistically significant relationships between changes in wages by occupation and subsequent changes in college majors completed in related fields of college study between 1982 and 2012. College majors (defined at a detailed level) are most strongly related to wages observed three years earlier, when students were college freshmen. The responses to wages vary depending on the extent to which there is a strong mapping of majors into particular occupations. We also find that women, blacks, Hispanics, and students with low test scores are less likely to respond to wage changes. These findings have implications for policy interventions designed to align students' major choices with labor market demand.

We investigate the hypothesis that the combination of three related innovations—1) information technology (IT), 2) complementary workplace reorganization, and 3) new products and services—constitute a significant skill-biased technical change affecting labor demand in the United States. Using detailed firm-level data, we find evidence of complementarities among all three of these innovations in factor demand and productivity regressions. In addition, firms that adopt these innovations tend to use more skilled labor. The effects of IT on labor demand are greater when IT is combined with the particular organizational investments we identify, highlighting the importance of IT-enabled organizational change.

The nonrivalry of ideas gives rise to increasing returns, a fact celebrated in Paul Romer's recent Nobel Prize. An implication is that the long-run rate of economic growth is the product of the degree of increasing returns and the growth rate of research effort; this is the essence of semi-endogenous growth theory. This review interprets past and future growth from a semi-endogenous perspective. For 50+ years, US growth has substantially exceeded its long-run rate because of rising educational attainment, declining misallocation, and rising (global) research intensity, implying that frontier growth could slow markedly in the future. Other forces push in the opposite direction. First is the prospect of “finding new Einsteins”: How many talented researchers have we missed historically because of the underdevelopment of China and India and because of barriers that discouraged women inventors? Second is the longer-term prospect that artificial intelligence could augment or even replace people as researchers. Throughout, the review highlights many opportunities for further research.

The authors describe the evolution of \n relative wages in five Latin American countries-Argentina, \n Brazil, Chile, Colombia, and Mexico. They use repeated \n cross-sections of household surveys, and decompose the \n evolution of relative wages into factors associated with \n changes in relative supply and relative demand. The authors \n have three main conclusions: 1) Increases in the relative \n wages of the most skilled (university-educated) workers took \n place concurrently with increases in their relative \n abundance in all of the countries except Brazil. This is \n strong evidence of increases in the demand for skilled \n workers. 2) Increases in the wage bill of skilled workers \n occurred largely within sectors, and in the same sectors in \n different countries, which is consistent with skill-biased \n technological change. 3) Trade appears to be an important \n transmission mechanism. Increases in the demand for the most \n skilled workers took place at a time when countries in Latin \n America considerably increased the penetration of imports, \n including imports of capital goods. The authors show that \n changes in the volume and research and development intensity \n of imports are significantly related to changes in the \n demand for more skilled workers in Latin America. Their \n research complements earlier work on the effects of \n technology transmitted through trade on productivity and on \n the demand for skilled labor.

We develop a microeconomic model of endogenous growth where clean and dirty technologies compete in production and innovation-in the sense that research can be directed to either clean or dirty technologies. If dirty technologies are more advanced to start with, the potential transition to clean technology can be difficult both because clean research must climb several rungs to catch up with dirty technology and because this gap discourages research effort directed towards clean technologies. Carbon taxes and research subsidies may nonetheless encourage production and innovation in clean technologies, though the transition will typically be slow. We characterize certain general properties of the transition path from dirty to clean technology. We then estimate the model using a combination of regression analysis on the relationship between R&D and patents, and simulated method of moments using microdata on employment, production, R&D, firm growth, entry and exit from the US energy sector. The model's quantitative implications match a range of moments not targeted in the estimation quite well. We then characterize the optimal policy path implied by the model and our estimates. Optimal policy makes heavy use of research subsidies as well as carbon taxes. We use the model to evaluate the welfare consequences of a range of alternative policies.

In Caselli and Coleman (2000) we developed a framework that separately identifies the efficiency units embodied in unskilled labor, skilled labor, and capital in a country’s aggregate production function. Applying that framework to cross-country data, we showed that countries where unskilled labor is relatively abundant are those with the most efficient unskilled workers, while countries where skilled labor and capital are abundant are the most efficient users of these inputs. We interpreted these findings as evidence of appropriatetechnology adoption: in each country firms choose from a menu of technologies; different technologies imply different combinations of values for the efficiency units embodied in the three factors of production; in each country the technology is chosen that makes the most of the most abundant factors. In this paper we apply the same framework to time-series data from the United States over the period 1963–1992. We find that throughout this period the efficiencies of skilled labor and capital have risen. The efficiency of unskilled labor has risen in tandem with those of the other factors in the early part of the sample, but surprisingly, it has been falling since sometime in the 1970’s (the exact turning point depends somewhat on some parametric assumptions). In analogy with the cross-country evidence, these changes are closely associated with changes in the relative abundance of skilled labor and capital, which increased rather dramatically. In this sense, the recent history of technologies in use by U.S. firms may mimic the choice of technologies around the world today: as skills and capital become more abundant, technologies are chosen that maximize the efficiency of these inputs. As we discuss below, however, in the U.S. context the converse story (relative labor supplies adjusting to exogenous changes in technology) is also consistent with the data. Besides its relevance to models of technology adoption, this evidence also sheds new light on the widely documented recent increase in the skilled wage premium in the United States. Lawrence F. Katz and Kevin M. Murphy (1992) and David H. Autor et al. (1998) used the relative wage and the relative labor supply series to show that the efficiency of unskilled labor relative to skilled labor must have declined over time. Our framework allows us to go further and show that, indeed, the absolute efficiency of unskilled labor has fallen (after 1970), while the efficiency of skilled labor and capital have increased. A similar result was obtained with different techniques by Marta Ruiz Arranz (2001).

College students select majors for a variety of reasons, including expected returns in the labor market. This paper demonstrates an empirical method linking a census of US degrees and fields of study with measures of the knowledge content of jobs. The study combines individual wage and employment data from the Current Population Survey (CPS) with ratings on 27 knowledge content areas from the Occupational Information Network (O*NET), thus providing measures of the economy-wide knowledge content of jobs. Fields of study and corresponding BA degree data from the Digest of Education Statistics for 1976-1977 through 2001-2002 are linked to these 27 content areas. We find that the choice of college major is responsive to changes in the knowledge composition of jobs and, more problematically, the wage returns to types of knowledge. Women's degree responsiveness to knowledge content appears to be stronger than men's, but their response to wage returns is weak. © 2007 Elsevier Ltd. All rights reserved.

We generalize a search, matching, and bargaining model to allow individuals to acquire productivity‐enhancing schooling prior to labor market entry. In general, search frictions and weakness in bargaining position contribute to underinvestment in schooling from an efficiency perspective. Using estimates of a general equilibrium version of the model in which firm vacancy creation decisions are included, we find that minimum wages and schooling subsidies improve aggregate welfare, but have very different welfare impacts across the ability distribution. In particular, policies that maximize the average welfare of workers have strongly negative effects on the welfare of the least able.

The decline in middle-wage occupations and rise in automation over the last decades are at the center of policy discussions. We develop an empirically relevant general equilibrium model that features endogenous labor force participation, occupational choice, and automation capital. We use the model to consider two types of policies: the retraining of workers who were adversely affected by automation, and redistribution policies that transfer resources to these workers. Our framework emphasizes general equilibrium effects such as displacement effects of retraining programs, complementarities between the factors of production, and the effects of distortionary taxation that is required to fund these programs.

Abstract Although workers in science, technology, engineering, and math (STEM) fields earn above‐average wages, the number of college graduates prepared for STEM jobs lags behind employer demand. A key question is how to recruit and retain college students in STEM majors. We offer new evidence on the role of financial aid in supporting STEM attainment. Exploiting a regression discontinuity that allows for causal inference, we find that eligibility for need‐based financial aid increased STEM credit completion by 20 to 35 percent among academically‐ready students in a large, public higher education system. These results appear to be driven by shifting students into STEM‐heavy course loads, suggesting aid availability impacts the academic choices students make after deciding to enroll. We also find suggestive evidence that aid offers increase degree attainment in STEM fields, although we cannot rule out null impacts on STEM degree production.

Deciding which field to study is one of the most consequential decisions college students make, but most research on the topic focuses on students attending four-year colleges. To understand how students attending community colleges make field of study decisions, the author links administrative educational records of recent high school graduates with local mass layoff and plant closing announcements. He finds that declines in local employment deter students from entering closely related community college programs and instead induce them to enroll in other vocationally-oriented programs. He further documents that students predominantly shift enrollment between programs that lead to occupations requiring similar skills.

This paper presents a novel stylized fact and analyzes its contribution to the skill bias of technical change in U.S. manufacturing. The share of skilled labor embedded in intermediate inputs correlates strongly with the skill share employed in final production. This finding points towards an intersectoral technology-skill complementarity (ITSC). Together with input-output linkages, the observed complementarity delivers a multiplier that reinforces skill demand along the production chain. Reduced-form estimates suggest that the effect is quantitatively important, explaining about as much skill upgrading as outsourcing. Empirical evidence suggests that one channel through which this complementarity works is product innovation. I also analyze the importance of different drivers of skill upgrading over time. While foreign outsourcing and IT capital is associated with skill demand particularly strongly from the 1980s onwards (a period of rapidly increasing skill premia), R&amp;D contributed stably throughout the period 1958-2005. The same is true for ITSC, which augmented within-sector skill bias in a stable fashion throughout the last 5 decades.

In most OECD-countries, labour demand has shifted from unskilled to skilled over time.\nMany analyses of this phenomenon focus either on technical change, capital-skill\ncomplementarity or mutual labour substitution. Applying a more general approach enables\nus to explore the relative importance of different factors behind the shift in labour demand\nin Norwegian manufacturing. A multivariate error-correction model of the cost-shares of\nskilled and unskilled labour, materials and energy is estimated. The results show that\nskilled-biased technical change, primarily due to a positive effect on skilled labour and less\ndue to a negative effect on unskilled labour, explains much of the shift in labour demand.\nIn addition, mutual labour substitution and capital stock growth are important.\nKeywords: Skilled-biased technical change; Factor demand; Industry level panel data

Timely and accurate measurement of AI use by firms is both challenging and crucial for understanding the impacts of AI on the U.S. economy.We provide new, real-time estimates of current and expected future use of AI for business purposes based on the Business Trends and Outlook Survey for September 2023 to February 2024.During this period, bi-weekly estimates of AI use rate rose from 3.7% to 5.4%, with an expected rate of about 6.6% by early Fall 2024.The fraction of workers at businesses that use AI is higher, especially for large businesses and in the Information sector.AI use is higher in large firms but the relationship between AI use and firm size is non-monotonic.In contrast, AI use is higher in young firms although, on an employmentweighted basis, is U-shaped in firm age.Common uses of AI include marketing automation, virtual agents, and data/text analytics.AI users often utilize AI to substitute for worker tasks and equipment/software, but few report reductions in employment due to AI use.Many firms undergo organizational changes to accommodate AI, particularly by training staff, developing new workflows, and purchasing cloud services/storage.AI users also exhibit better overall performance and higher incidence of employment expansion compared to other businesses.The most common reason for non-adoption is the inapplicability of AI to the business.

We analyze the theoretical effects on growth and welfare in transition economies of emigration of educated and uneducated labor, of higher emigration probability, etc. Using a Grossman-Helpman growth model, we show that the prospects of labor market integration with the EU raises the expected returns to education, stimulate human capital formation and thus raise the growth rate in the candidate countries. However, given this expected returns, emigration of educated workers tends to lower growth and welfare of those remaining. Thus, while the brain drain reduces welfare, the effects of labor market integration could nevertheless be positive. Emigration of low skilled workers also reduces growth via adverse effects on education. Higher tuition fees, common in transition countries, counteract positive growth effects of market determined wages.

We analyze the effect of technological change in a novel framework that integrates an economy's skill distribution with its occupational and industrial structure. Individuals become managers or workers based on their managerial vs. worker skills, and workers further sort into a continuum of tasks (occupations) ranked by skill content. Our theory dictates that faster technological progress for middle-skill tasks not only raises the employment shares and relative wages of lower-and higher-skill occupations among workers (horizontal polarization), but also raises those of managers over workers as a whole (vertical polarization). Both dimensions of polarization are faster within sectors that depend more on middle-skill tasks and less on managers. This endogenously leads to faster TFP growth of such sectors, whose employment and value-added shares shrink if sectoral goods are complementary (structural change). We present several novel facts that support our model, followed by a quantitative analysis showing that task-specific technological progress--which was fastest for occupations embodying routine-manual tasks but not interpersonal skills--is important for understanding changes in the sectoral, occupational, and organizational structure of the U.S.

Abstract We identify phrases associated with novel technologies using textual analysis of patents, job postings, and earnings calls, enabling us to identify four stylized facts on the diffusion of jobs relating to new technologies. First, the development of economically impactful new technologies is geographically highly concentrated, more so even than overall patenting: 56% of the most economically impactful technologies come from just two U.S. locations, Silicon Valley and the Northeast Corridor. Second, as the technologies mature and the number of related jobs grows, hiring spreads geographically. This process is very slow, taking around 50 years to disperse fully. Third, while initial hiring in new technologies is highly skill-biased, over time the mean skill level in new positions declines, drawing in an increasing number of lower-skilled workers. Finally, the geographic spread of hiring is slowest for higher-skilled positions, with the locations where new technologies were pioneered remaining the focus for the technology's high-skill jobs for decades.

This Paper proposes a model of endogenous growth where innovating requires both researchers, who produce inventions, and entrepreneurs who implement them. As research and entrepreneurship compete in the allocation of aggregate resources, the relation between growth and research effort is hump-shaped. When entrepreneurs appropriate too little rents from innovation, too few resources are allocated to entrepreneurship and returns to R&D are low because of this lack of entrepreneurial skills. When so, innovation should be promoted by encouraging entrepreneurship rather than research.

We use the American Community Survey (ACS) to investigate the extent to which college major decisions were affected during and after the Great Recession with special attention to business and science, technology, engineering, and mathematics (STEM) fields, as well as the heterogeneity across demographic groups. Several conclusions are reached. First, the Great Recession increased the frequency of STEM majors but decreased the frequency of business majors. Second, the increase for STEM fields spreads across several detailed STEM majors, while the decrease in business majors is especially concentrated among finance and management. Third, we find strong heterogeneous effects of the Great Recession by gender and race/ethnicity. ( JEL I20, J24)

While many preceding studies discuss the equilibrium intensity of R&D, this article focuses on its equilibrium direction. There can be a pure-strategy equilibrium in which multiple firms "cluster," i.e., attempt to develop the same technology even if (i) potential technologies are ex ante equally promising, (ii) each technology can be patented by no more than one firm, and (iii) there are no informational spillovers among firms. Economic applications of this clustering result are not confined to R&D. Any situation where agents are racing in search of exclusive economic opportunities can be an example of this model.

Abstract This paper examines the role of lifelong learning in counteracting skill depreciation and obsolescence. We differentiate between occupations with more hard skills versus more soft skills and draw on representative job advertisement data that contain machine‐learning categorized skill requirements and cover the Swiss job market in great detail across occupations (from 1950 to 2019). We examine lifelong learning effects for “harder” versus “softer” occupations, thereby analyzing the role of training in counteracting skill depreciation in occupations that are differently affected by skill depreciation. Our results reveal novel empirical patterns regarding the benefits of lifelong learning, which are consistent with theoretical explanations based on structurally different skill depreciation rates: In harder occupations, with large shares of fast‐depreciating hard skills, the role of lifelong learning is primarily as a hedge against unemployment risks rather than a boost to wages. By contrast, in softer occupations, in which workers build on more value‐stable soft‐skill foundations, the role of lifelong learning instead lies mostly in acting as a boost for upward career mobility and leads to larger wage gains.

In this paper, I ask the question: <it>Does the output-mix of countries change in response to changes in factor endowments?</it> If so: <it>How long does it take?</it> Using data on capital, as well as skilled and unskilled labour employed in three-digit International Standard Industrial Classification (ISIC) manufacturing industries for a sample of 27 developing and developed countries over the 1973–1990 period, I find that the output-mix of countries does not change in response to endowment changes, even after 15 years. This answer raises another question: <it>How then do countries absorb changes in factor endowments?</it> The data show that in both the short and long runs, an increase in the supply of a production factor reduces its rate of return and makes it more intensively used in all sectors of the economy: changes in production techniques. In the long run, the point estimate is that the reduction in the rate of return is more than 50% larger than in the short run. This is consistent with induced innovations being predominantly biased towards the scarce factor.

This paper proposes a new and unified explanation for the following trends observed over the last 25 years: (1) the increased returns to education, (2) the slow measured growth in TFP in an economy undergoing massive changes in its methods of production, and (3) the poor wage performance, relative to TFP growth, of both young high school and college educated workers. The explanation we propose downplays the role of exogenous skill-biased technological change and instead emphasizes how the endogenous choice of modes of organization, influenced by changes in factor supplies, can generate the above observations. For example, we argue that increased education attainment, through its effect of the choice production techniques, may have been the major cause for the increased differential between more and less educated workers over the last quarter of a century. The evidence we examine to test our hypothesis is based on US and Canadian data over the period 1971 - 95. We pay particular attention to explaining the difference between our results and those associated with the skill-biased technical change hypothesis.

Abstract This paper offers the first systematic historical evidence on the role of a central actor in modern growth theory: the engineer. We construct a database on the share of engineers in the labor force during the Second Industrial Revolution (1870–1914) at the county level for the United States and the state and national levels for the Americas. These measures are robustly correlated with income today after controlling for literacy, other types of higher-order human capital (college graduates, lawyers, physicians, patenting) and demand-side factors, as well as after instrumenting engineering using the 1862 US Land Grant Colleges program. Differences in engineering density in 1880 accounted for 10% of the higher US county incomes today, while national disparities in engineering density can explain approximately a quarter of the income divergence in the Americas. To document the mechanisms through which engineering density works, we show how it is correlated with higher rates of technology adoption and structural transformation across intermediate time periods and with numerous measures of the knowledge economy today.

Using case study data gathered between 1993 and 1996, the authors investigate how automation of information handling and materials handling affected employment distribution, skill acquisition, work activities, and compensation in 23 semiconductor plants in four countries. Information handling automation is skill-biased technical change, which leads to use of relatively more technicians and engineers. In the sample studied, it widened the skill gap across occupations, and coincided with higher initial wages for all employees and shorter career ladders for engineers. Materials handling automation also widened the skill gap, but coincided with employment of relatively more operators and with lower pay across all occupations. Although technological change widened the skill gap between occupations and was biased toward employment of high-skill workers in the sample, the authors do not find that it led to increased wage inequality in the semiconductor plants they examine.

This paper examines shifts over time in the relative demand for skilled labor in the United States. Although de-skilling in the conventional sense did occur overall in nineteenth century manufacturing, a more nuanced picture is that occupations hollowed out: the share of middle-skill jobs - artisans - declined while those of high-skill - white collar, non-production workers - and low-skill - operatives and laborers increased. De-skilling did not occur in the aggregate economy; rather, the aggregate shares of low skill jobs decreased, middle skill jobs remained steady, and high skill jobs expanded from 1850 to the early twentieth century. The pattern of monotonic skill upgrading continued through much of the twentieth century until the recent polarization of labor demand since the late 1980s. New archival evidence on wages suggests that the demand for high skill (white collar) workers grew more rapidly than the supply starting well before the Civil War.

In the mid-2010s, as computing and other digital technologies matured (), researchers began to speculate about a new era of innovation—with artificial intelligence (AI) as the standard-bearer of a “Fourth Industrial Revolution” (). The release of generative AI (Gen-AI) technologies (e.g., ChatGPT) in late 2022 reignited the discussion, prompting us to wonder: what are the barriers, risks, and potential rewards to using gen-AI for design and manufacturing? As Gen-AI has entered the mainstream, geopolitics and business practices have shifted. Covid-19 disrupted global supply chains, tensions with import partners have risen, and military conflicts introduce new uncertainties. As companies consider propositions like ‘reshoring’ or ‘nearshoring/friendshoring’ production (), we recognize other hindrances: suboptimal resource allocation, labor market volatility and trends toward an older and geographically mismatched workforce, and highly concentrated tech markets that foster anticompetitive business practices. As the United States expands domestic production capacity (e.g., semiconductors and electric vehicles), Gen-AI could help us overcome those challenges. To investigate the current and potential usefulness of Gen-AI in design and manufacturing, we interviewed industry experts—including engineers, manufacturers, tech executives, and entrepreneurs. They have identified many opportunities for the deployment of Gen-AI: (1) reducing the incidence of costly late-stage design changes when scaling production; (2) providing information to designers and engineers, including identifying suitable design spaces and material formulations and incorporating consumer preferences; (3) improving test data interpretation to enable rapid validation and qualification; (4) democratizing workers’ access and usage of data to enable real-time insights and process adjustment; and (5) empowering less-skilled workers to be more productive and do more-expert work. Current Gen-AI solutions (e.g., ChatGPT, Claude) cannot accomplish these goals due to several key deficiencies, including the inability to provide robust, reliable, and replicable output; lack of relevant domain knowledge; unawareness of industry-standards requirements for product quality; failure to integrate seamlessly with existing workflow; and inability to simultaneously interpret data from different sources and formats. We propose a development framework for the next generation of Gen-AI tools for design and manufacturing (“NextGen-AI”): (1) provide better information about engineering tools, repositories, search methods, and other resources to augment the creative process of design; (2) integrate adherence to first principles when solving engineering problems; (3) leverage employees’ experiential knowledge to improve training and performance; (4) empower workers to perform new and more-expert productive tasks rather than pursue static automation of workers’ current functions; (5) create a collaborative and secure data ecosystem to train foundation models; and (6) ensure that new tools are safe and effective. These goals are extensive and will require broad-based buy-in from business leaders, operators, researchers, engineers, and policymakers. We recommend the following priorities to enable useful AI for design and manufacturing: (1) improve systems integration to ethically collect real-time data, (2) regulate data governance to ensure equal opportunity in development and ownership, (3) expand the collection of worker-safety data to assess industry-wide AI usage, (4) include engineers and operators in the development and uptake of new tools, and (5) focus on skills-complementary deployments to maximize productivity upside.

In this paper, we construct a parsimonious overlapping-generations model of human capital accumulation and study its quantitative implications for the evolution of the U.S. wage distribution from 1970 to 2000. A key feature of the model is that individuals differ in their ability to accumulate human capital, which is the main source of wage inequality in this model. We examine the response of this model to skill-biased technical change (SBTC), which is modeled as an increase in the trend growth rate of the price of human capital starting in the early 1970s. The model displays behavior that is consistent with several important trends observed in the US data, including the rise in overall wage inequality; the fall and subsequent rise in the college premium, as well as the fact that this behavior was most pronounced for younger workers; the rise in within-group inequality; the stagnation in median wage growth; and the small rise in consumption inequality despite the large rise in wage inequality. We consider different scenarios regarding how individuals' expectations evolve during SBTC. Specifically, we study the case where individuals immediately realize the advent of SBTC (perfect foresight), and the case where they initially underestimate the future growth of the price of human capital (pessimistic priors), but learn the truth in a Bayesian fashion over time. Lack of perfect foresight appears to have little effect on the main results of the paper. Overall, the model shows promise for explaining a diverse set of wage distribution trends observed since the 1970s in a unifying human capital framework.

The author investigates the effects of \n technology on the employment and wages of differently \n skilled Mexican manufacturing workers using firm panel data \n from 1992-99. She analyzes the relationship between \n technology and skill demand. Findings support the \n skill-biased technical change hypothesis. She then examines \n the temporal relationship of technology adoption to firm \n productivity and worker wages. The author finds that skilled \n labor increases after technology adoption. And wages of both \n skilled and semi-skilled workers exhibit markedly increased \n growth rates compared with the growth rate of low-skilled \n workers. The results show that investment in human capital \n improves technology-driven productivity gains.

Abstract The employability of an aging population in a world of continuous and biased technical change is top of the political agenda. Due to endogenous human capital depreciation the effective retirement age is often below statutory retirement age resulting in permanent non-employability of older workers. We analyze this phenomenon in a putty-putty human capital vintage model and focus on education and the speed of human capital depreciation. Introducing a two-stage education system with initial schooling and lifelong learning, not even lifelong learning turns out to be capable of aligning economic and statutory retirement. However, well-designed education programs will keep more workers in highly productive activities at the end of their working life, and hence will substitute for simple social transfers, or for an early switch towards very low paid jobs.

The Korean vocational education and training (VET) system is heralded as one of the key factors contributing to the countries past economic growth. VET has played an important role in developing a skilled labor force during Korea's economic development. However, with the increasing importance of higher education and general education, the status of VET in the country is declining. This paper explores recent Korean data to analyze the labor market outcomes of pre-employment VET institutions. The findings show that current vocational high school education is not associated with better labor market outcomes, in terms of employment rate, wage levels, prospect of permanent employment, and transition to the first job, when compared to general high school education. Among VET programs, the authors find that graduates of higher level, more comprehensive VET programs experience greater labor market achievements than graduates of less competitive, shorter programs. The authors also find that the VET institutes play an important role in supplying technical labor to small and medium enterprises (SMEs).

A start‐up engages in an investment portfolio problem by choosing how much to invest in a “non‐rival” project and a “rival” project that threatens an incumbent. Anticipating its acquisition, the start‐up distorts its investment portfolio in order to raise acquisition rents. This may improve or worsen the direction of innovation and consumer surplus. The bigger the difference in social surplus appropriability across the two projects, the more likely it is that the direction of innovation improves and consumers benefit from an acquisition. These results also hold if the acquirer takes over the research facilities of the start‐up.

How are inventors allocated in the US economy and does that allocation affect innovative capacity? To answer these questions, we first build a model where an inventor with a new idea has the possibility to work for an entrant or incumbent firm. Strategic considerations encourage the incumbent to hire the inventor, offering higher wages, and then not implement her idea. We then combine data on 760 thousand U.S. inventors with the LEHD data. We find that when an inventor is hired by an incumbent, their earnings increases by 12.6 percent and their innovative output declines by 6 to 11 percent.

We build a model of technology choice with heterogeneous firms and workers to study how imported technology affects wages. Imported machines increase the productivity of worker-firm matches, but are more expensive than domestic ones. More productive firms and more skilled workers are hence more likely to use an imported machine. We study trade liberalization in the model, which makes imported machines cheaper. Both the direct and the equilibrium implications of trade liberalization increase the returns to skill. We use linked employer-employee data on Hungarian machine operators for 1992-2003 to test the predictions of the model. Machine operators exposed to imported machines earn higher wages than similar workers at similar firms. The returns to skill have increased in our sample between 1992 and 2000. A quarter of the increase can be attributed to greater exposure to imported machines. Our results suggest that imported machines can help propagate skill-biased technical change.

This paper examines whether the sector bias of skill-biased technical change (SBTC) explains changing skill premia within countries in recent decades. First, using a two-factor, two-sector, two-country model we demonstrate that in many cases it is the sector bias of SBTC that determines SBTC’s effect on relative factor prices, not its factor bias. Thus, rising (falling) skill premia are caused by more extensive SBTC in skill-intensive (unskill-intensive) sectors. Second, we test the sector-bias hypothesis using industry data for many countries in recent decades. An initial consistency check strongly supports the hypothesis. Among ten countries we find a strong correlation between changes in skill premia and the sector bias of SBTC during the 1970s and 1980s. The hypothesis is also strongly supported by more structural estimation on UK and US data of the economy-wide wage changes ‘mandated’ to maintain zero profits in all sectors in response to the sector bias of SBTC. The suggestive mandated-wage estimates match the direction of actual wage changes in both countries during both the 1970s and the 1980s. Thus, the empirical evidence strongly suggests that the sector bias of SBTC can help explain changing skill premia.

This study examined the effects of the misallocation of talent between the government and private enterprise sectors on innovation. By using the 2005 inter-census population survey and patent database, we find a negative correlation between misallocation of talent and innovation intensity. Exploring possible mechanisms, we conclude that this negative correlation between misallocation of talent and innovation was best explained by the negative impact of such misallocation upon entrepreneurship and R&D spending. That is, misallocations of talent reduce the willingness of people to be productive Schumpeterian entrepreneurs, and the majority of companies affected by such misallocation are reluctant to increase R&D spending. Most importantly, we find that excessive talent enters government departments in prefectures worse the suppression of innovation. This result sheds new light on the important role of allocation of talent on innovation for scholars and policymakers. Our findings have important implications for how to effectively allocate talent between the government vs.enterprise sectors in order to encourage more productive, value-creating activities.

We develop measures of labor-saving and labor-augmenting technology exposure using textual analysis of patents and job tasks.Using US administrative data, we show that both measures negatively predict earnings growth of individual incumbent workers.While labor-saving technologies predict earnings declines and higher likelihood of job loss for all workers, laboraugmenting technologies primarily predict losses for older or highly-paid workers.However, we find positive effects of labor-augmenting technologies on occupation-level employment and wage bills.A model featuring labor-saving and labor-augmenting technologies with vintage-specific human capital quantitatively matches these patterns.We extend our analysis to predict the effect of AI on earnings.

Abstract When job tasks are automated, does this augment or diminish the value of labor in the tasks that remain? We argue the answer depends on whether removing tasks raises or reduces the expertise required for remaining non-automated tasks. Since the same task may be relatively expert in one occupation and inexpert in another, automation can simultaneously replace experts in some occupations while augmenting expertise in others. We propose a conceptual model of occupational task bundling that predicts that changing occupational expertise requirements have countervailing wage and employment effects: automation that decreases expertise requirements reduces wages but permits the entry of less expert workers; automation that increases expertise requirements increases wages but reduces the set of qualified workers. We develop a novel, content-agnostic method for measuring job-task expertise, and we use it to quantify changes in occupational expertise demands over four decades attributable to job task removal and addition. We document that automation has raised wages and decreased employment in occupations where it eliminated inexpert tasks, but reduced wages and increased employment in occupations where it eliminated expert tasks. These effects are distinct from—and in the case of employment, opposite to—the effects of changing task quantities. The expertise framework resolves the puzzle of why routine task automation has decreased employment but often increased wages in routine task-intensive occupations. It provides a general tool for analyzing how task automation and new task creation reshape the scarcity value of human expertise within and across occupations.

The share of high-skilled workers in U.S. cities is positively correlated with city size, and this correlation strengthened between 1980 and 2010. Furthermore, during the same time period, the U.S. economy experienced a significant structural transformation with regard to industrial composition, most notably in the decline of manufacturing and the rise of highskilled service industries. To decompose and investigate these trends, this paper develops and estimates a spatial equilibrium model with heterogeneous firms and workers that allows for both industry-specific and skill-specific technology changes across cities. The estimates imply that both supply and demand of high-skilled labor have increased over time in big cities. In addition, demand for skilled labor in large cities has increased somewhat within all industries. However, this aggregate increase in skill demand in cities is highly concentrated in a few industries. The finance, insurance, and real estate sectors alone account for 35 percent of the net change over time.

This study investigates the relationship between occupational choice and unemployment over the professional career with German administrative linked employer–employee data that track more than 800,000 graduates from vocational education over 25 years. Using short-run fluctuations in local and sectoral occupation-specific labor demand as instruments, it finds that choosing an occupation that later turns out to suffer from low or negative employment growth has a statistically and economically significant impact on unemployment over the professional career. On average, an unanticipated one-standard deviation decrease in occupation-specific employment growth raises unemployment by about 116 days over the life cycle.

We study the impact of financial incentives on higher education decisions and the choice of major. We rely on a reform whereby Israeli kibbutzim shifted from their traditional policy of equal sharing to productivity-based wages. We use for identification the staggered implementation of this reform in different kibbutzim. In this setting of very low initial returns to education, we find that the dramatic increase in the rate of return and its sharp variation across fields of study led to a large increase in the probability of receiving a Bachelor degree, especially in STEM fields of study that are expected to yield higher financial returns. For men this increase was largely in computer science and engineering, and for women in biology, chemistry and computer science. Our findings suggest that investment in higher education and the choice of major are responsive to increases in the return to education for both men and women.

We study the impact of financial incentives on higher education decisions and the choice of major. We rely on a reform whereby Israeli kibbutzim shifted from their traditional policy of equal sharing to productivity-based wages, using for identification the staggered implementation of this reform. In this setting of very low initial returns to education, we find that the dramatic increase in the rate of return and its sharp variation across fields of study led to a large increase in the probability of receiving a bachelor’s degree, especially in STEM fields of study that are expected to yield higher financial returns.

We introduce publicly funded education in R&D-based economic growth theory. The framework allows us to i) incorporate a realistic process of human capital accumulation for industrialized countries, ii) reconcile R&D-based growth theory with the empirical evidence on the relationship between economic prosperity and population growth, iii) revise the policy invariance result of semi-endogenous growth frameworks, and iv) show that the transitional effects of an education reform tend to be qualitatively different from its long-run impact.

Abstract This article reviews recent work on how automation and task displacement have contributed to labour share declines and inequality in the US labour market. We summarize the basic building blocks of a task-based framework in which a set of tasks is allocated between capital, skilled labour and unskilled labour. Automation, which corresponds to the use of new technologies expanding the set of tasks that can be performed by capital, always reduces the labour share in value added and may depress overall wages and employment. The negative effects of automation on labour share and its potentially adverse consequences for labour demand can be counterbalanced by the creation of new labour-intensive tasks, which can reinstate labour into the production process. We also show that when automation displaces unskilled labour from the tasks in which they used to specialize (which has been its modal impact so far), it increases the demand for skills and inequality. New tasks may or may not limit the increase in the demand for skills depending on whether they are mostly targeted at skilled workers. We then provide a range of evidence supporting the basic predictions and implications of this framework. Most importantly, the decline in the share of labour in national income and the increase in the demand for skills appear to be related to an acceleration in the pace of automation and a deceleration in technological changes complementing humans during the last 30 years. We end with a discussion of the potential bias towards automation in the development and adoption of digital technologies, and how this will affect the nature of work in the face of recent advances in artificial intelligence.

Abstract This paper studies the long-term effects of technology and know-how transfers on structural transformations. In the 1950s, the Soviet Union supported the construction of the 156 Projects, which were large-scale, capital-intensive industrial clusters in China. These projects included a technology transfer, consisting of state-of-the-art Soviet machinery and equipment, and a know-how transfer, via the training of Chinese engineers, production supervisors, and high-skilled technicians by Soviet experts. We use newly assembled data that follow steel plants for over four decades, and we exploit natural variation in the transfers they eventually received. We find that, while production advantages stemming from Soviet technology faded away if not complemented with training, the know-how transfer had a long-lasting impact on plant performance, stimulated technology upgrade when China was a closed economy, and increased exports to the Western world when China engaged in international trade. The know-how transfer also generated productivity and technology spillovers onto complementary establishments.

Abstract German establishments heavily rely on the apprenticeship system for skill supply. With one in four apprenticeship contracts ending before successful completion, it is in the interest of establishments and policy-makers to determine factors, which reduce non-completion. This paper investigates the role of apprenticeship wages and income prospects after completion for apprenticeship non-completion in Germany. For this purpose, this study identifies incidences of apprenticeship non-completion in a large sample of administrative data on employment biographies and estimates a piecewise exponential model of the non-completion hazard with shared frailties by occupations. The results suggest a robust and significant association with both apprenticeship wages and skilled worker wages. All else at means, apprenticeships which are paid 5% more than the mean apprenticeship wage, on average have a 0.8 percentage points higher estimated survival rate. In turn, an apprenticeship expected to lead to a skilled job that is paid 5% above average, has an estimated survival rate, which is 3.1 percentage points higher on average. These findings highlight the importance of income prospects for apprenticeship non-completion.

We propose a model to show that when innovation in a given field becomes more lucrative, its direction can be distorted even though its rate rises. Higher payoffs attract innovators, making the R&D supply side more competitive. This competition endogenously shifts effort toward less promising but quicker-to-invent projects. We empirically quantify the magnitude of this distortion, in the context of pharmaceutical innovation during the Covid-19 pandemic. In the social planner solution, 74 percent more firms would have worked on vaccines and 17 percent more on novel compounds. Policy remedies include advance purchase commitments based on ex-ante value, targeted research subsidies, and antitrust exemptions for joint research ventures.

The article proposes a new conceptual framework for capturing production, R&D, and economic growth in aggregative models which extend their horizon into the digital era. Two key factors of production are considered: hardware, including physical labor, traditional physical capital and programmable hardware, and software, encompassing human cognitive work, pre-programmed software, and artificial intelligence (AI). Hardware and software are complementary in production whereas their constituent components are mutually substitutable. The framework generalizes, among others, the standard model of production with capital and labor, models with capital–skill complementarity and skill-biased technical change, and unified growth theories embracing also the pre-industrial period. It offers a clear conceptual distinction between mechanization and automation as well as between robotization and the development of AI. It delivers sharp, economically intuitive predictions for long-run growth, the evolution of factor shares, and the direction of technical change

This paper presents an endogenous growth model of directed technical change with vertical and horizontal R&D to study an analytical mechanism that is consistent with the coexistence of aggregate stability and structural change. We focus on changes in the share of the high- versus the low-tech sectors in the context of a slow, but flexible, transitional dynamics that arises from a dynamic system with a three-dimensional stable manifold. Under the hypothesis of a positive shock in the proportion of high-skilled labour, the technological-knowledge bias channel leads to nonbalanced sectoral growth, while the aggregate variables remain approximately constant and thus consistent with the Kaldor facts. With prevailing market-scale effects, a calibration exercise shows that the model is able to account for around two-thirds of the increase in the share of the high-tech manufacturing sectors observed in European data from 1995 to 2007.

This paper examines the impact of stricter high school math requirements on the likelihood of completing a degree in STEM fields. Exploiting cross-state variation in the timing of math reforms, I find that stricter math curriculum requirements significantly increased the proportion of the college-educated population earning a STEM degree. Within STEM, the increases in degree completion are concentrated in math and science while there is little discernible impact in technology and engineering. Further analysis suggests that high school graduation, college attendance, and overall degree completion are largely unaffected by the implementation of math reforms. Instead, stricter math curriculum requirements appear to have shifted some students away from non-STEM fields into STEM fields.

We describe alignment between high school career and technical education (CTE) and local labor markets across five states—Massachusetts, Michigan, Montana, Tennessee, and Washington. We find that CTE is partially aligned with local labor markets. A 10-percentage-point higher share of local jobs related to a CTE career cluster is associated with a 3-point higher rate of CTE concentration in that cluster. Women and students from racial or ethnic minority groups are better aligned with local employment than men, in part due to their selection of CTE fields like Education & Training, Health Science, and Hospitality & Tourism, which correspond with a large portion of the workforce in almost every metro area. We find more limited evidence of dynamic, short-term adjustments in CTE after changes in local labor markets. A small degree of realignment lags the labor market by 2 to 3 years and is only observed following changes in college-level employment.

Both the macroeconomic and the microeconomic evidence from U. S. economy’s experience over the past two centuries leads to a view of technological change (broadly conceived) as having not been “neutral” in its effects upon growth. The specific meaning of “non-neutrality” in this context is that technical and organizational innovation had effects upon the derived demands for factors of production, and these tended to alter the relative prices of the heterogeneous array of productive assets in the economy. By directly and indirectly impinging on relative real rates of remuneration established in the markets for particular types of human labor and skill, and for the services of specific tangible and intangible capital, “technological change” altered key conditions governing the absolute and relative growth rates of the various macroeconomic factors of production. On the other hand, because innovation exhibited strong cumulative features reflecting the influence of “localized learning, ” past domestic factor market conditions exerted a persisting influence upon the globally non-neutral trajectory of American technological and organizational development. This essay thus explores two broad and related historical themes. Firstly, the nonneutrality of the impacts of innovations on the demand side of the markets for productive inputs implies that “innovation” should be understood as contributing to complex interactions among all the proximate “sources of growth.” Even though the latter are usually presented by exercises in “growth accounting” as distinct and separate dynamic elements contributing to the rise of labor productivity and per capita real output, the identification of the total factor productivity “residual” as the “contribution” of technological change is mistaken in ignoring the quantitatively important effect of successive capital-deepening “traverses” to the growth of labor productivity. The second theme underscores a fundamental contrast between the twentieth and the nineteenth century growth processes, in regard to the impacts of the predominant “bias” of the direction of innovation: the relative shift away from the accumulation of stocks of tangible reproducible capital and towards the formation of intangible productive assets by in investments in education, training and the search for new scientific and technological knowledge.

Skill-biased technical change is usually interpreted in terms of the efficiency parameters of skilled and unskilled labor. This implies that the relative productivity of skilled workers changes proportionally in all tasks. In contrast, we argue that technical changes also affect the curvature of the distribution of relative productivity. Building on Rosen's (1978) tasks assignment model, this implies that not only the efficiency parameters of skilled and unskilled workers change, but also the elasticity of substitution between skill-types of labor. Using data for the United States between 1963 and 2002, we find significant empirical support for a decrease in the elasticity of substitution at the end of the 70s followed by an increase at the end of the 80s. This pattern of the elasticity of substitution has contributed to the slowdown in labor productivity in the late 70s through the 80s and to a speedup in the 90s

In this paper we test the hypothesis that the sector bias of skill-biased technical change is important in explaining the rising relative wage of skilled workers in the manufacturing sector in three Central and Eastern European transition countries. The evidence for Hungary and Poland is consistent with the sector bias being important in explaining the rising wage premium; the hypotheses is however not confirmed for the Czech Republic.

To study the short-run and long-run implications on wage inequality, we introduce directed technical change into a Ricardian model of offshoring. A unique final good is produced by combining a skilled and an unskilled product, each produced from a continuum of intermediates (tasks). Some of these tasks can be transferred from a skill-abundant West to a skill-scarce East. Profit maximization determines both the extent of offshoring and technological progress. Offshoring induces skill-biased technical change because it increases the relative price of skill intensive products and induces technical change favoring unskilled workers because it expands the market size for technologies complementing unskilled labor. In the empirically more relevant case, starting from low levels, an increase in offshoring opportunities triggers a transition with falling real wages for unskilled workers in the West, skill-biased technical change and rising skill premia worldwide. However, when the extent of offshoring becomes sufficiently large, further increases in offshoring induce technical change now biased in favor of unskilled labor because offshoring closes the gap between unskilled wages in the West and the East, thus limiting the power of the price effect fueling skill-biased technical change. The unequalizing impact of offshoring is thus greatest at the beginning. Transitional dynamics reveal that offshoring and technical change are substitutes in the short run but complements in the long run. Finally, though offshoring improves the welfare of workers in the East, it may benefit or harm unskilled workers in the West depending on elasticities and the equilibrium growth rate.

When job tasks are automated, does this augment or diminish the value of labor in the tasks that remain?We argue the answer depends on whether removing tasks raises or reduces the expertise required for remaining non-automated tasks.Since the same task may be relatively expert in one occupation and inexpert in another, automation can simultaneously replace experts in some occupations while augmenting expertise in others.We propose a conceptual model of occupational task bundling that predicts that changing occupational expertise requirements have countervailing wage and employment effects: automation that decreases expertise requirements reduces wages but permits the entry of less expert workers; automation that raises requirements raises wages but reduces the set of qualified workers.We develop a novel, content-agnostic method for measuring job task expertise, and we use it to quantify changes in occupational expertise demands over four decades attributable to job task removal and addition.We document that automation has raised wages and reduced employment in occupations where it eliminated inexpert tasks, but lowered wages and increased employment in occupations where it eliminated expert tasks.These effects are distinct from-and in the case of employment, opposite to-the effects of changing task quantities.The expertise framework resolves the puzzle of why routine task automation has lowered employment but often raised wages in routine task-intensive occupations.It provides a general tool for analyzing how task automation and new task creation reshape the scarcity value of human expertise within and across occupations.

Technological progress renders various skills obsolete, however, the rate of skill obsolescence will vary according to the worker's human capital investments. Workers heavily invested in general skills, such as education, will not suffer high rates of obsolescence, while less-educated workers who invest more in “technology-specific” skills will suffer more when the technology is changed. Consistent with this framework, this chapter demonstrates that increasing randomness is the primary source of inequality growth within uneducated workers, whereas inequality growth within educated workers is determined more by predictable factors. Furthermore, this chapter shows that increasing randomness generates a “precautionary” demand for education.

We study the impact of techies—engineers and other technically trained workers—on firm-level productivity. We first report new facts on the role of techies in the firm by using French administrative data and unique surveys. Techies are STEM-skill intensive and are associated with innovation, as well as with technology adoption, management, and diffusion within firms. Using structural econometric methods, we estimate the causal effect of techies on firm-level Hicks-neutral productivity in both manufacturing and non-manufacturing industries. We find that techies raise firm-level productivity, and this effect goes beyond the employment of R&D workers, extending to ICT and other techies. In non-manufacturing firms, the impact of techies on productivity operates mostly through ICT and other techies, not R&D workers. Engineers have a greater effect on productivity than technicians.

This paper investigates the determination and consequences of organizational changes (OC) in a panel of British and French establishments. Organizational changes include the decentralization of authority, delayering of managerial functions, and increased multitasking. We argue that OC and skills are complements. We offer support for the hypothesis of "skill-biased" organizational change with three empirical findings. First, organizational changes reduce the demand for unskilled workers in both countries. Second, OC is negatively associated with increases in regional skill price differentials (a measure of the relative supply of skill). Third, OC leads to greater productivity increases in establishments with larger initial skill endowments. Technical change is also complementary with human capital, but the effects of OC is not simply due to its correlation with technological change but has an independent role.

This paper examines the effect of information and communication technologies (ICT) on the return paid to two different types of skill: general skills, acquired through schooling and work experience, and job-specific skills, acquired by experience in a particular job. Using the UK Labour Force Survey we estimate skill returns in different industries over the period 1994-2001. We evaluate the marginal effect on these returns of the ICT intensity of industry capital and find that the shift towards ICT capital has been associated with a rise in the return to general skills and a reduction in the return to job-specific experience. JEL Classification: J30; J31; O30. Key Words: Skill-biased technical change; returns to human capital; technology adoption; skills obsolescence. Acknowledgements: We are grateful to Mary O’Mahony for constructing and advising on the ICT and capital stock data and to participants at the conference of the Royal Economic Society,

In this paper we present an analytically tractable general equilibrium overlapping-generations model of human capital accumulation, and study its implications for the evolution of the U.S. wage distribution from 1970 to 2000. The key feature of the model, and the only source of heterogeneity, is that individuals differ in their ability to accumulate human capital. Therefore, wage inequality results only from differences in human capital accumulation. We examine the response of this model to skill-biased technical change (SBTC) theoretically. We show that in response to SBTC, the model generates behavior consistent with the U.S. data including (i) a rise in overall wage inequality in both the short run and long run, (ii) an initial fall in the education premium followed by a strong recovery, leading to a higher premium in the long run, (iii) the fact that most of this fall and rise takes place among younger workers, (iv) stagnation in median wage growth (and a slowdown in aggregate labor productivity), and (v) a rise in consumption inequality that is much smaller than the rise in wage inequality. These results suggest that the heterogeneity in the ability to accumulate human capital is an important feature for understanding the effects of SBTC, and interpreting the transformation that the U.S. economy has gone through since the 1970s.

Abstract This paper proposes a theoretical analysis and an empirical investigation on simultaneous choices of an enrolment and discipline field, comparing science, technology, engineering, and mathematics (STEM) to non‐STEM fields, to enlighten economic variables influencing students. The cross‐country analysis of 35 countries in the period 2013–2016 tries to disentangle factors shifting students from one choice to another and why countries differ in directing students toward specific disciplines. The results show that expenditures in R&D convince more students to choose STEM disciplines, whereas population density and the expected years of schooling have a negative impact on the percentage of STEM enrolments. As a whole, it appears that STEM enrolments are chosen for investment motives, whereas non‐STEM enrolments are less guided by economic factors and more dependent on a consumption effect.

We present a model where firms conduct R&D in both a safe and a risky direction. As patentability standards rise, an innovation in the risky direction is less likely to receive a patent, which decreases the static incentive for new entrants to conduct risky R&D but can increase their dynamic incentive. These, together with a strategic substitution and a market structure effect, result in an inverted‐U shape in the risky direction but a U shape in the safe direction for the relationship between R&D intensity and patentability standards. R&D is biased toward (against) the risky direction under lower (higher) standards.

This paper considers the allocation of innovators between two research lines that differ in their values of innovation and their probabilities of discovery. Innovators choose a research line to maximize their expected utility, and the high value research line may attract more or fewer innovators than the low value research line, depending on the difficulty of discovery. The equilibrium allocation is not efficient, as innovators ignore the effects of their choice of research lines on other innovators.

Where does adaptation to innovation take place? The supply of educated workers and local industry structure matter for the subsequent location of new work?that is, new types of labor-market activities that closely follow innovation. Using census 2000 microdata, the author shows that regions with more college graduates and a more diverse industrial base in 1990 are more likely to attract these new activities. Across metropolitan areas, initial college share and industrial diversity account for 50% and 20%, respectively, of the variation in selection into new work unexplained by worker characteristics. He uses a novel measure of innovation output based on new activities identified in decennial revisions to the U.S. occupation classification system. New work follows innovation, but unlike patents, it also represents subsequent adaptations by production and labor to new technologies. Further, workers in new activities are more skilled, consistent with skill-biased technical change.

Labor market frictions are seen in many extensions of the classical human capital theory as a prerequisite for firms financing general training. The labor market reforms in Germany at the beginning of the millennium have therefore been seen by many as a danger to the firms’ willingness to support the apprenticeship training system. This paper analyzes the training strategies German firms deployed to cope with the greater labor market flexibility as a result of the labor market reform. Switzerland where no reforms had taken place serves as the counterfactual. The results show that firms successfully reduced the net-costs of training by involving apprentices in more work and reducing non-productive tasks, like practicing. Contrary to the widespread fear, this adapted training strategy resulted also in a substantial increase in work-related competencies and productivity of apprentices.

Demography and structural transformation are interrelated, and depend critically on education. At the turn of the twentieth century, US parents began having fewer children while increasing educational investment per child. This quantity-quality tradeoff facilitated job reallocation from the low-skilled agricultural sector to the high-skilled nonagricultural sector. This transformation is examined in a heterogeneous agent model with a nondegenerate human capital distribution, focusing on how fertility and education decisions affect structural transformation. The result shows that the quantity-quality decisions account for up to approximately one-third of the decline in the agricultural employment share. (JEL E24, I21, I26, J11, J13, J24, N31)

In this paper we present a tractable general equilibrium overlapping-generations model of human capital accumulation which is consistent with several key features of the evolution of the U.S. wage distribution from 1970 to 2000. The key feature of the model, and the only source of heterogeneity, is that individuals differ in their ability from all kinds of idiosyncratic uncertainty, and thus, wage inequality results only from differences in human capital accumulation. We examine the response of this model to skill-biased technical change (SBTC) both theoretically and quantitatively. First, we theoretically show that in response to SBTC, the model generates behavior consistent with the U.S. data including (i) a rise in total wage inequality, (ii) an initial fall in the education (skill) premium followed by a strong recovery, leading to a higher premium in the long-run, (iii) the fact that most of this fall and rise takes place among younger workers, (iv) a rise in within-group inequality, (v) an increase in educational attainment, (vi) stagnation in median wage growth (and a slowdown in aggregate labor productivity), and (vii) a rise in consumption inequality that is much smaller than the rise in wage inequality. We then calibrate the model to the U.S. data before 1970, and find that the evolution of these variables closely track their empirical counterparts during SBTC (from 1970 on). These results suggest that the heterogeneity in the ability to accumulate human capital is a key feature for understanding the effects of SBTC and interpreting the transformation that the US economy has gone through since the 1970’s.

This paper relates changes in the high-skilled / medium-skilled relative wage of full-time male wage-earners in France with changes in supply and skill-biased demand shifts, including skill-biased technical change (SBTC). Using annual employer-employee administrative data matched with Census data from 1967 to 2009, we document a strong decrease in this relative wage concomitantly with a strong increase in the relative supply. Estimating a labor supply and demand model, in which experience groups are imperfect substitutes, we show that the strong increase in educational attainment in France has hidden so far the effects of SBTC. The magnitude of these effects is however between half and 100% of what is usually found in the U.S. for the same period. Then, education supply has stabilized since the mid-1990s for young cohorts. Our simulations show that if this stabilization goes on whereas the skill-biased demand keeps increasing or even remains constant, there will be a rise in wage inequality in the next two decades for the cohorts of workers who experienced the supply stabilization. Finally, we discuss why this increase in wage inequality may even be stronger.

This paper studies how education choices and labor market frictions interact in shaping wage inequality. The wage premium of college graduates relative to high school graduates (between-group inequality) has tripled since 1980 in the U.S., and the variance of log wages conditional on educational attainments (within-group inequality) has become about 50% larger across the board. To understand the source of this change, we construct a model with schooling investments and labor market frictions that generates supply and demand of skills and frictional wage differentials as equilibrium objects. The model features a two-sided sorting: education sorting of skilled workers into college education and labor market sorting of productive firms into the labor market for college graduates − together implying an assortative matching of high skilled workers to productive firms. A novel model-based wage decomposition of both the between- and within-group inequalities is obtained. Calibrating the model to the U.S. data, we find that the inequality trend is accounted for by worker composition and labor market friction. If there were no skill- biased technological change, the variance of log wages would be smaller, mainly due to lower within-group inequality.

We study how periodic uncertainty about skilled labor supply affects corporate investment using the H-1B visa program for skilled workers as an empirical setting. Exploiting cross-regional variation in H-1B labor flows based on historical immigrant enclaves, we find that firms in regions that attract more H-1B workers concentrate their investments in the quarter after uncertainty about visa access is resolved by the H-1B lottery. Consistent with a skilled labor uncertainty channel, we find that the investment spikes are confined to industries where invested capital cannot be easily redeployed, to firms that cannot easily find domestic substitutes for lost H-1B workers, and to firms that cannot easily arrange alternative employment visas for their foreign employees. This paper was accepted by Camelia Kuhnen, finance. Funding: S.-J. Xu gratefully acknowledges research support from the University of Alberta. Supplemental Material: The internet appendix and data files are available at https://doi.org/10.1287/mnsc.2023.03486 .

Firm price-cost markups may reflect (a) bigger step sizes from quality innovations that confer significant knowledge spillovers onto other firms, and/or (b) higher process efficiency than competing firms or other factors which bear no obvious knowledge externality.We write down an endogenous growth model with innovation step size and process efficiency as alternative sources of markup heterogeneity.Compared with the laissez-faire equilibrium, the social planner wants to reallocate research towards high step size firms but not high process efficiency firms.We then use price and productivity data across firms in French manufacturing to infer firm step sizes and process efficiency.We find that the planner could achieve faster growth by reallocating research toward high step size firms, and more so if high step size firms could freely license their innovations to high process efficiency firms.

We study how large shocks impact individuals’ skilling decisions using data from a large, government-sponsored, online learning platform in Saudi Arabia. The onset of the COVID-19 pandemic brought about a massive increase in online skilling, and demand shifted towards courses that offered skills, such as telework, likely to be immediately valuable during the pandemic. Consistent with a model where individuals trade off reskilling costs with their expectations of future labor market conditions and their duration of work, we find that shifts into telework courses were largest for older workers. In contrast, younger workers increased enrollments in courses related to new skills, such as general, occupation-specific, and computer-related skills. Using national administrative employment data, we provide descriptive evidence that these investments in skills in early 2020 helped users maintain employment over the course of the pandemic.

We identify phrases associated with novel technologies using textual analysis of patents, job postings, and earnings calls, enabling us to identify four stylized facts on the diffusion of jobs relating to new technologies. First, the development of economically impactful new technologies is geographically highly concentrated, more so even than overall patenting: 56% of the most economically impactful technologies come from just two U.S. locations, Silicon Valley and the Northeast Corridor. Second, as the technologies mature and the number of related jobs grows, hiring spreads geographically. But this process is very slow, taking around 50 years to disperse fully. Third, while initial hiring in new technologies is highly skill biased, over time the mean skill level in new positions declines, drawing in an increasing number of lower-skilled workers. Finally, the geographic spread of hiring is slowest for higher-skilled positions, with the locations where new technologies were pioneered remaining the focus for the technology’s high-skill jobs for decades.

Abstract We study the interplay between the decision of firms to innovate and human capital. Based on a dynamic evolutionary model, we show that in the presence of a high stock of human capital, an advanced economy can remain caught in an “innovation trap”. Following the literature on endogenous growth, R&D investments and human capital are modeled as strategic complements. Skilled workers increase productivity and enjoy a wage premium if they are employed in the R&D sector, while they receive the same wage as unskilled workers if they are employed in the production sector. We model the evolutionary dynamics of the share of innovative firms and human capital to determine the conditions under which an economy converges to a high, low or mixed state of innovation.

Abstract Employers regularly complain of a shortage of qualified scientists and advocate that to remain competitive more scientists need to be trained. However, using a survey of graduates from British universities, I report that 3 years after graduation less than 50% of graduates from science subjects are working in a scientific occupation. Accounting for selection into major and occupation type, I estimate the wages of graduates and report that the wage premium of science graduates only occurs when these graduates are matched to a scientific occupation – and not because science skills are in demand in all occupations. I also provide additional evidence to assess whether science graduates are pushed or pulled into non-scientific occupations. Altogether, the evidence does not support the claim that science graduates are pulled by better conditions, financial or otherwise, into non-scientific jobs.

While job polarization was a salient feature in European economies in the decade up to 2010, this phenomenon has all but disappeared, except in a handful of Southern-European economies. The decade following 2010 is characterized by occupational upgrading, where low-paid jobs shrink and high paid jobs expand. We show that this is associated with automation: employment shares in low paid, highly automatable jobs shrinks, while employment shares of better paid jobs that are unlikely to be automated expands. Techies (engineers and technicians with strong STEM skills) help explain cross country variation in occupational upgrading: economies that are abundant in techies or exhibit high growth of techies see strong skill upgrading; in contrast, polarization is observed in economies with few techies. Robotization is associated with skill upgrading in manufacturing. We discuss the additional roles of globalization, structural change and labor market institutions in driving these phenomena. Hitherto, artificial intelligence (AI) seems to have similar impacts as other automation technologies. However, there is uncertainty about what new AI technologies harbor.

This paper investigates the shift in demand away from low-skilled and towards high-skilled labour in the Netherlands over the 1990s. Making the distinction between the effects of technical change on job type and job level, the conclusion is that skill-biased technical change based on job level is the chief cause for this shift.

We use recent advances in natural language processing and large language models to construct novel measures of technology exposure for workers that span almost two centuries.Combining our measures with Census data on occupation employment, we show that technological progress over the 20th century has led to economically meaningful shifts in labor demand across occupations: it has consistently increased demand for occupations with higher education requirements, occupations that pay higher wages, and occupations with a greater fraction of female workers.Using these insights and a calibrated model, we then explore different scenarios for how advances in artificial intelligence (AI) are likely to impact employment trends in the medium run.The model predicts a reversal of past trends, with AI favoring occupations that are lower-educated, lowerpaid, and more male-dominated.

Chandler (1977) shows that large firms require hierarchies of white-collar workers to coordinate complex production.We document that this insight continues to hold globally today, and we show that low education levels in developing countries limit the supply of white-collar workers and constrain firm size.We extend the occupational choice model of Lucas (1978) to allow entrepreneurs to reorganize their firms by allocating administrative tasks to hired professionals, which brings the firm closer to constant returns to scale.We calibrate the model to be consistent with cross-sectional microdata and validate it using quasi-experimental and experimental evidence on the effects of educational expansions and management training interventions.Skills explain twothirds of the reorganization of production into large firms with economic development, while structural transformation and reductions in barriers are needed to explain the remaining shift.

With the majority of all H-1B visas going to Indians, we study how US immigration policy coupled with the internet boom affected both the US and Indian economies, and in particular both countries’ IT sectors. The H-1B scheme led to a tech boom in both countries, inducing substantial gains in firm productivity and consumer welfare in both the United States and India. We find that the US-born workers gained $431 million in 2010 as a result of the H-1B scheme. In India, the H-1B program induced Indians to switch to computer science (CS) occupations, increasing the CS workforce and raising overall IT output in India by 5 percent. Indian students enrolled in engineering schools to gain employment in the rapidly growing US IT industry via the H-1B visa program. Those who could not join the US workforce, due to the H-1B cap, remained in India, and along with return-migrants, enabled the growth of an Indian IT sector, which led to the outsourcing of some production to India. The migration and rise in Indian exports induced a small number of US workers to switch to non-CS occupations, with distributional impacts. Our general equilibrium model captures firm-hiring across various occupations, innovation and technology diffusion, and dynamic worker decisions to choose occupations and fields of major in both the United States and India. Supported by a rich descriptive analysis of the changes in the 1990s and 2000s, we match data moments and show that our model captures levels and trends of key variables in validation tests. We perform counter-factual exercises and find that on average, workers in each country are better off because of high-skill migration.

Previous studies assume that labor market skills are either fully general or specific to a firm. This paper uses patterns in mobility and wages to the transferability of specific skills across occupations. The empirical analysis combines information on tasks performed in different occupations with a large panel on complete work histories and wages. Our results demonstrate that labor market skills are partially transferable across occupations. We find that individuals move to occupations with similar task requirements, and that the distance of moves declines with time in the labor market. Further, tenure in the last occupation affects current wages, and the effect is stronger if the two occupations are similar. We calculate that task-specific human capital is an important source of wage growth, especially for university graduates

The dissertation consists of an introductory chapter and three essays that apply search-matching theory to study the interaction of labor market frictions, technological change and macroeconomic fluctuations. \n\nThe first essay studies the impact of capital-embodied growth on equilibrium unemployment by extending a vintage capital/search model to incorporate vintage human capital. In addition to the capital obsolescence (or creative destruction) effect that tends to raise unemployment, vintage human capital introduces a skill obsolescence effect of faster growth that has the opposite sign. Faster skill obsolescence reduces the value of unemployment, hence wages and leads to more job creation and less job destruction, unambiguously reducing unemployment. \n\nThe second essay studies the effect of skill biased technological change on skill mismatch and the allocation of workers and firms in the labor market. By allowing workers to invest in education, we extend a matching model with two-sided heterogeneity to incorporate an endogenous distribution of high and low skill workers. We consider various possibilities for the cost of acquiring skills and show that while unemployment increases in most scenarios, the effect on the distribution of vacancy and worker types varies according to the structure of skill costs. When the model is extended to incorporate endogenous labor market participation, we show that the unemployment rate becomes less informative of the state of the labor market as the participation margin absorbs employment effects. \n\nThe third essay studies the effects of labor taxes on equilibrium labor market outcomes and macroeconomic dynamics in a New Keynesian model with matching frictions. Three policy instruments are considered: a marginal tax and a tax subsidy to produce tax progression schemes, and a replacement ratio to account for variability in outside options. In equilibrium, the marginal tax rate and replacement ratio dampen economic activity whereas tax subsidies boost the economy. The marginal tax rate and replacement ratio amplify shock responses whereas employment subsidies weaken them. The tax instruments affect the degree to which the wage absorbs shocks. We show that increasing tax progression when taxation is initially progressive is harmful for steady state employment and output, and amplifies the sensitivity of macroeconomic variables to shocks. When taxation is initially proportional, increasing progression is beneficial for output and employment and dampens shock responses.

Unlike most empirical cross-country analysis of the determinants of innovation, which focus mainly on developed countries, this study analyzes the transition countries of Eastern Europe and the Former Soviet Union.

This paper develops a new framework to analyze the incidence of labor market shocks, focusing on automation and artificial intelligence. Central to our theory is the distance-dependent elasticity of substitution (DIDES), where worker mobility between occupations declines with their distance in skill space. Mapping 306 occupations into cognitive, manual, and interpersonal skill dimensions, we estimate a low-dimensional latent skill model that preserves granular substitution patterns. We show that both automation and artificial intelligence cluster within skill-adjacent occupations, constraining employment adjustment and amplifying wage effects. The clustering nature of technologies generates unequal outcomes: 20--50% of labor demand shocks translate to wages (versus 30% under standard models), while mobility recovers only 20\% of losses (versus 30% from standard estimates).

We document that nearly half of the global decline in agricultural employment was driven by new cohorts entering the labor market. A new dataset of policy reforms supports an interpretation of these cohort effects as human capital. Using a model of frictional labor reallocation, we conclude that human capital growth led to a sharp decline in the agricultural labor supply, accounting, at fixed prices, for 40% of the decrease in agricultural employment. This aggregate effect is halved in general equilibrium and it reflects the role of human capital as both a mediating factor and an independent driver of labor reallocation.

This paper examines the regulation of technological innovation direction under uncertainty about potential harms. We develop a model with two competing technological paths and analyse various regulatory interventions. The optimal regulatory approach depends critically on the magnitude of potential harm relative to technological benefits. Our analysis reveals a motive to double down on harmful technologies in resource allocation across research paths, challenging common intuitions about diversification. We demonstrate that ex post regulatory instruments, particularly liability regimes, outperform ex ante restrictions in many scenarios. These insights have important implications for regulating emerging technologies like artificial intelligence, suggesting the need for informationally-responsive regulatory frameworks.

ABSTRACT We model a duopolistic game where firms first choose the direction of their innovation, then invest in the chosen direction, and finally, compete in the product market. Investments occur either in overlapping or non‐overlapping territories. We show that, in the presence of a generous patent regime that allows the protection of innovations of little value, firms tend to invest in overlapping technologies; stricter requirements for patentability may induce firms to operate in different technological areas, thereby increasing market efficiency. We illustrate our general theory using two stylized models of Cournot competition with product and process innovations, respectively.

Why is invention strongly positively correlated with parental income not only in the US but also in Finland which displays low income inequality and high social mobility?Using data on 1.45M Finnish individuals and their parents, we find that: (i) the positive association between parental income and off-spring probability of inventing is greatly reduced when controlling for parental education; (ii) instrumenting for the parents having a MSc-degree using distance to nearest university reveals a large causal effect of parental education on offspring probability of inventing; and (iii) the causal effect of parental education has been markedly weakened by the introduction in the early 1970s of a comprehensive schooling reform.

This study explores how educational investment impacts technology adoption and firm productivity in Vietnam. While higher education is linked to better adaptability to new technologies, its direct effect on business efficiency remains unclear. This research investigates the impact of industry-specific education and skills on wage growth and firm performance, particularly in sectors undergoing technological transformation. This study utilizes the Vietnam Household Living Standards Survey (VHLSS) and Historical Adoption of Technology (HATCH) data to employ instrumental variable analysis, assessing the causal impact of education on technology adoption and wage growth. Sectoral and worker-level data help identify long-term productivity trends. Using linked VHLSS worker records and an industry-year technology exposure index from HATCH, we demonstrate that higher educational attainment is associated with greater absorptive capacity; however, increases in technology exposure are followed by short-term declines in earnings growth for incumbent workers. Differences by education are modest in our estimates and are not systematically smaller for higher-educated workers. These results reconcile the role of education in capability building with the adjustment costs associated with technological transitions. We do not report a separate productivity model, so we refrain from making quantitative claims about province-level productivity effects in this paper. The study relies on secondary data, limiting control over unobserved variables affecting technology adoption. This study provides new causal evidence on the link between education, technology adoption, and firm productivity in a developing economy. • Educational investment increases workers' absorptive capacity for new technology. • Technology exposure reduces short-term earnings growth for incumbent workers. • Differences in adjustment costs between education groups are modest. • Education supports capability building but transitions create temporary wage losses. • Study provides causal evidence from VHLSS and HATCH data for Vietnam.

This paper generalizes the canonical model of human capital accumulation through schooling to endogenize the process of academic specialization. It provides the solution to a class of dynamic investment problems with switching and stopping under sequential uncertainty. Under mild assumptions, the model's optimal policy has a particularly simple form that can be reduced to the comparison of independent indices. The optimal policy implies that schooling should begin with a period of general education, common to all students, followed by a period of gradual academic specialization before graduation. At the microeconomic level, it is consistent with the dynamics of student course taking observed in the data and the outcomes of educational interventions studied in the literature. At the macroeconomic level, its predictions are consistent with models of how education should adapt to changes in the speed and scope of technological change in labor markets.

This paper studies how the speed-quality tradeoff in innovation interacts with firm dynamics, concentration, and economic growth. Empirically, we document long-run trends in the increasing speed of innovation alongside declining quality at large firms. Leveraging variation from an exogenous policy change, we document the existence of the speed-quality tradeoff both at the firm and aggregate level. We develop an endogenous growth model that incorporates the speed-quality tradeoff and show that allocating less labor towards speed increases growth, particularly in the presence of private benefits to innovation and spillovers from heterogeneous innovations. We quantify the model to link firms’ decisions across speed and quality to aggregate outcomes. Quantitatively, the recent growth slowdown is mainly due to changes in the innovation production function, while the allocation of inventors between speed and quality within firms has a modest impact. When spillovers across firms are taken into account, the effect becomes significantly larger; the shift to speed over the last 30 years explains up to one-quarter of the decrease in growth.

This study empirically investigates the effects of a policy change in the Republic of Korea that lifted the restriction requiring universities to admit students to groups of departments rather than to single departments. Entry cohorts affected by the policy change had to specify their majors from the beginning of their college education. Using this policy-driven change in the timing of specialization, we find that early specialization lowered wages during early career. This negative wage effect was not driven by selection at admissions or students choosing lower-paying majors and occupations, although graduated majors shifted toward less popular majors. Rather, it is related to their employment in lower-paying industries, which is closely linked to weak local labor demand at the time of graduation. These findings are consistent with less adaptive skills to navigate unexpected changes in the labor market.

What role does education play in economic growth? Conventional wisdom and macroeconomic theories posit that as a nation becomes more educated, they become wealthier. The basic argument says a more educated populace is more productive (i.e. the quality of human capital increases) thereby increasing economic output. However, the majority of empirical work done on this topic has not found a strong relationship between education and economic growth. The purpose of this paper will be to identify where this body of research went wrong and offer theoretical insights ignored by this literature based in market-process theory tradition. It will draw upon an existing body of research (both empirical and theoretical) that fits within this theoretical paradigm as well as suggest a path forward for researchers in this field.

We introduce publicly funded education into R&D-based economic growth theory. Our framework allows us to i) explicitly describe a realistic process of human capital accumulation within these types of growth models, ii) reconcile semi-endogenous growth theory with the empirical evidence on the relationship between economic development and population growth, and iii) revise the policy invariance result of semi-endogenous growth frameworks. In particular, we show that the model supports a negative association between economic growth and population growth if the education sector is well developed and the population growth rate is low, that is, for modern industrialized countries. Furthermore, within our framework, changes in public educational investments have the potential to affect the long-run balanced growth rate.

Science, Technology, Engineering, and Mathematics (STEM) dalam sistem pendidikan merupakan salah satu bidang yang sangat penting kepada para pelajar dalam meyediakan peluang untuk mengalami peluang belajar dalam tetapan kerjaya sebenar. Kepentingan pengintegrasian STEM dalam kurikulum matematik diberi perhatian apabila terdapat permintaan untuk kerjaya bidang STEM dalam industri pekerjaan. Justeru, data empirikal yang berkaitan dengan kecenderungan pemilihan kursus STEM dalam tempoh 2019 hingga Mac 2023 dikumpul dan disintesis untuk melaksanakan kajian literatur yang bersistematik ini. Tujuan kajian ini adalah untuk mengetahui kecenderungan pemilihan kursus STEM dalam kalangan pelajar sekolah menengah. Pangkalan data Web of Science (WOS) dan SCOPUS digunakan untuk mendapatkan artikel yang berkaitan tentang kecenderungan pemilihan kursus STEM. Empat kriteria dalam kajian ini adalah tahun penerbitan, negara, pendekatan kajian serta faktor-faktor yang mempengaruhi pemilihan kursus STEM. Terdapat 18 artikel dipilih berdasarkan empat kriteria yang dinyatakan dan dianalisis. Model PRISMA dijadikan sebagai panduan dalan pencarian bahan literatur. Hasil analisis menunjukkan bahawa pelajar sekolah menengah memiliki kecenderungan yang tinggi dan positif terhadap pemilihan kursus STEM semasa sambung belajar ke peringkat universiti.

This dissertation offers a comprehensive analysis of the wide-ranging economic implications and labour market consequences of autonomous technologies, while also considering the role of the circular economy in sustaining economic growth. Spanning three thematic areas across nine self-contained essays, the research provides a combination of macroeconomic modelling and microeconomic evidence. The thesis is divided into three main parts, leaving aside the introduction and the concluding remarks. The first part focuses on macroeconomic theoretical modelling of a dual traditional-autonomous economy, employing dynamic general equilibrium models to evaluate the impact of autonomous technologies on the economy. The second part provides microeconomic evidence regarding the influence of autonomous technologies on labour markets. Finally, the third part presents a novel perspective on the integration of circular economy concepts into a neoclassical dynamic general equilibrium model. The thesis initiates its exploration through the lens of a dual traditional-autonomous economy model to study the economic implications of automation, discovering that the effects are largely determined by the adoption rate of autonomous capital and its elasticity of substitution with traditional technology. A significant finding suggests the existence of an adoption rate threshold, beyond which the process of automation can lead to a complete shift from traditional capital and labour. Furthermore, the study uncovers the necessity for a profound reform of current tax systems by observing a reduction in the government’s size due to the substitution of traditional tax-bearing inputs with autonomous technology. Further exploration into the optimal tax policy for maintaining the social security contributions to GDP ratio by taxing autonomous capital suggests that a robots’ social security tax paid by employers of autonomous capital is the most efficient long-term strategy. The dissertation then pivots to provide microeconomic evidence, examining the impact of digitalization on labour markets, using data from the US and European countries. The research provides a view of the effects of digitalization on the US employment landscape, presenting mappings that classify occupations based on their relationship with automation and AI. The study also investigates the implications of the automation process and AI on early retirement decisions across 26 European countries, revealing a significant role of technological change in these decisions. Moreover, this part analyses the role of computerization, AI, machine learning, and occupational reorganization capacity in unemployment probabilities among older workers, indicating the heterogeneity in the impact of new technologies on the labour market. Finally, this section explores the implications of digitalization for worker mobility, illustrating a significant influence on the relocation of displaced workers. Finally, the thesis presents a novel mathematical description of a circular economy by incorporating the concept into a neoclassical dynamic general equilibrium linear economy model. The study reveals a positive S-shaped relationship between the optimal recycling rate and economic development, concluding that increasing the circularity of the economy is a necessary condition for enhancing social welfare in a growing economy. Overall, the dissertation presents a comprehensive examination of the broad economic impacts of autonomous technologies, while introducing the concept of the circular economy into macroeconomic modelling. The findings have important implications for policy-making, contributing to a better understanding of the technology-driven economic changes.

We examine the role of technological change in shaping insurance to the unemployed. We integrate technological change, occupation choice, and employment risk into a Bewley-style economy to examine the optimal combination of retraining subsidies and public insurance transfers for unemployed workers. We find that the optimal policy introduces a retraining subsidy to unemployed workers and increases the generosity of transfers to the unemployed relative to current U.S. policy. The utilitarian government incorporates retraining subsidies as part of an optimal policy as they provide additional, longer run, consumption insurance after job loss while imposing only modest increases in distortionary taxes. In the absence of technological change, a utilitarian government sets a lower subsidy on the tuition cost of retraining as earnings declines after job loss are less persistent.

This paper presents a fully specified model of long-run growth in which knowledge is assumed to be an input in production that has increasing marginal productivity. It is essentially a competitive equilibrium model with endogenous technological change. In contrast to models based on diminishing returns, growth rates can be increasing over time, the effects of small disturbances can be amplified by the actions of private agents, and large countries may always grow faster than small countries. Long-run evidence is offered in support of the empirical relevance of these possibilities.

Invention is here interpreted broadly as the production of knowledge. From the viewpoint of welfare economics, the determination of optimal resource allocation for invention will depend on the technological characteristics of the invention process and the nature of the market for knowledge.

Journal Article The Economic Implications of Learning by Doing Get access Kenneth J. Arrow Kenneth J. Arrow Stanford Search for other works by this author on: Oxford Academic Google Scholar The Review of Economic Studies, Volume 29, Issue 3, June 1962, Pages 155–173, https://doi.org/10.2307/2295952 Published: 01 June 1962

Traditional growth theory emphasizes the incentives for capital accumulation rather than technological progress. Innovation is treated as an exogenous process or a by-product of investment in machinery and equipment. Grossman and Helpman develop a unique approach in which innovation is viewed as a deliberate outgrowth of investments in industrial research by forward-looking, profit-seeking agents.

A simple supply and demand framework is used to analyze changes in the U. S. wage structure from 1963 to 1987. Rapid secular growth in the demand for more-educated workers, "more-skilled" workers, and females appears to be the driving force behind observed changes in the wage structure. Measured changes in the allocation of labor between industries and occupations strongly favored college graduates and females throughout the period. Movements in the college wage premium over this period appear to be strongly related to fluctuations in the rate of growth of the supply of college graduates.

"Volumes 2A and 2B of The Handbook of Economic Growth summarize recent advances in theoretical and empirical work while offering new perspectives on a range of growth mechanisms, from the roles played by institutions and organizations to the ways factors beyond capital accumulation and technological change can affect growth. Written by research leaders, the chapters summarize and evaluate recent advances while explaining where further research might be profitable. With analyses that are provocative and controversial because they are so directly relevant to public policy and private decision-making, these two volumes uphold the standard for excellence in applied economics set by Volumes 1A and 1B (2005)"--Publisher description

We study the effects of industrial robots on US labor markets. We show theoretically that robots may reduce employment and wages and that their local impacts can be estimated using variation in exposure to robots—defined from industry-level advances in robotics and local industry employment. We estimate robust negative effects of robots on employment and wages across commuting zones. We also show that areas most exposed to robots after 1990 do not exhibit any differential trends before then, and robots’ impact is distinct from other capital and technologies. One more robot per thousand workers reduces the employment-to-population ratio by 0.2 percentage points and wages by 0.42%.

This handbook comprises two volumes: 'Developments in research methods and their applications' (volume 4A) and 'New developments and research on labor markets' (volume 4B). The purpose is to examine what new tools and models economists can use to understand how individuals participate in labor markets. Using a mixture of conceptual models and empirical research, the contributors demonstrate how better data and advanced experiments help them apply economic theory, producing better analyses and conclusions. The chapters show how labor economists are developing new and innovative ways to measure key parameters and test important hypotheses. Volume 4A contains the following chapters: Decomposition methods in economics / Nicole Fortin, Thomas Lemieux, Sergio Firpo; Field experiments in labor economics / John A. List, Imran Rasul; Lab labor: what can labor economists learn from the lab? / Gary Charness, Peter Kuhn; The structural estimation of behavioral models: discrete choice dynamic programming methods and applications / Michael P. Keane, Petra E. Todd, Kenneth I. Wolpin; Program evaluation and research designs / John DiNardo, David S. Lee; Identification of models of the labor market / Eric French, Christopher Taber; Search in macroeconomic models of the labor market / Richard Rogerson, Robert Shimer; Extrinsic rewards and intrinsic motives: standard and behavioral approaches to agency and labor markets / James B. Rebitzer, Lowell J. Taylor. Volume 4B contains: Earnings, consumption and life cycle choices / Costas Meghir, Luigi Pistaferri; Racial inequality in the 21st century: the declining significance of discrimination / Roland G. Fryer; Imperfect competition in the labor market / Alan Manning; Skills, tasks and technologies: implications for employment and earnings / Daren Acemoglu, David Autor; Institutional reforms and dualism in European labor markets / Tito Boeri; Local labor markets / Enrico Moretti; Human capital development before age five / Douglas Almond, Janet Currie; Recent developments in intergenerational mobility / Sandra E. Black, Paul J. Devereux; New perspectives on gender / Marianne Bertrand; Great expectations: law, employment contracts, and labor market performance / W. Bentley MacLeod; Human resource management and productivity / Nicholas Bloom, John Van Reenen; Personnel economics: hiring and incentives / Paul Oyer, Scott Schaefer.

This paper argues that the 'scale effects' prediction of many recent R&D-based models of growth is inconsistent with the time-series evidence from industrialized economies. A modified version of the Romer model that is consistent with this evidence is proposed, but the extended model alters a key implication usually found in endogenous growth theory. Although growth in the extended model is generated endogenously through R&D, the long-run growth rate depends only on parameters that are usually taken to be exogenous, including the rate of population growth. Copyright 1995 by University of Chicago Press.

Endogenous growth theory explains long-run growth as emanating from economic activities that create new technological knowledge. This article sketches the outlines of the theory, especially the ‘Schumpeterian’ variety, and briefly describes how the theory has evolved in response to empirical discoveries.

Whole eras of technical progress and growth appear to be driven by a few ‘General Purpose Technologies’ (GPT's), such as the steam engine, the electric motor, and semiconductors. GPT's are characterized by pervasiveness, inherent potential for technical improvements, and ‘innovational complementarities’, giving rise to increasing returns-to-scale. However, a decentralized economy will have difficulty in fully exploiting the growth opportunities of GPT's: arms-length market transactions between the GPT and its users may result in ‘too little, too late’ innovation. Likewise, difficulties in forecasting the technological developments of the other side can lower the rate of technical advance of all sectors.

A recent “revisionist” literature characterizes the pronounced rise in U.S. wage inequality since 1980 as an “episodic” event of the first half of the 1980s driven by nonmarket factors (particularly a falling real minimum wage) and concludes that continued increases in wage inequality since the late 1980s substantially reflect the mechanical confounding effects of changes in labor force composition. Analyzing data from the Current Population Survey for 1963 to 2005, we find limited support for these claims. The slowing of the growth of overall wage inequality in the 1990s hides a divergence in the paths of upper-tail (90/50) inequality—which has increased steadily since 1980, even adjusting for changes in labor force composition—and lower-tail (50/10) inequality, which rose sharply in the first half of the 1980s and plateaued or contracted thereafter. Fluctuations in the real minimum wage are not a plausible explanation for these trends since the bulk of inequality growth occurs above the median of the wage distribution. Models emphasizing rapid secular growth in the relative demand for skills—attributable to skill-biased technical change—and a sharp deceleration in the relative supply of college workers in the 1980s do an excellent job of capturing the evolution of the college/high school wage premium over four decades. But these models also imply a puzzling deceleration in relative demand growth for college workers in the early 1990s, also visible in a recent “polarization” of skill demands in which employment has expanded in high-wage and low-wage work at the expense of middle-wage jobs. These patterns are potentially reconciled by a modified version of the skill-biased technical change hypothesis that emphasizes the role of information technology in complementing abstract (high-education) tasks and substituting for routine (middle-education) tasks.

This paper investigates the relationship between product market competition (PMC) and innovation. A Schumpeterian growth model is developed in which firms innovate ѳtep-by-stepҬ and where both technological leaders and their followers engage in R&D activities. In this model, competition may increase the incremental profit from innovating; on the other hand, competition may also reduce innovation incentives for laggards. This model generates four main predictions which we test empirically. First, the relationship between product market competition (PMC) and innovation is an inverted U-shape: the escape competition effect dominates for low initial levels of competition, whereas the Schumpeterian effect dominates at higher levels of competition. Second, the equilibrium degree of technological Ѯeck-and-neckness' among firms should decrease with PMC. Third, the higher the average degree of Ѯeck-and-neckness' in an industry, the steeper the inverted-U relationship between PMC and innovation in that industry. Fourth, firms may innovate more if subject to higher debt-pressure, especially at lower levels of PMC. We confront these four predictions with a new panel data set on UK firms' patenting activity at the US patenting office. The inverted U relationship, the neck and neck, and the debt pressure predictions are found to accord well with observed behavior in the data.

The supply and price of skilled labor relative to unskilled labor have changed dramatically over the postwar period. The relative quantity of skilled labor has increased substantially, and the skill premium, which is the wage of skilled labor relative to that of unskilled labor, has grown significantly since 1980. Many studies have found that accounting for the increase in the skill premium on the basis of observable variables is difficult and have concluded implicitly that latent skill-biased technological change must be the main factor responsible. This paper examines that view systematically. We develop a framework that provides a simple, explicit economic mechanism for understanding skill-biased technological change in terms of observable variables, and we use the framework to evaluate the fraction of variation in the skill premium that can be accounted for by changes in observed factor quantities. We find that with capital-skill complementarity, changes in observed inputs alone can account for most of the variations in the skill premium over the last 30 years.

This paper examines the effect of skill-biased technological change as measured by computerization on the recent widening of U. S. educational wage differentials. An analysis of aggregate changes in the relative supplies and wages of workers by education from 1940 to 1996 indicates strong and persistent growth in relative demand favoring college graduates. Rapid skill upgrading within detailed industries accounts for most of the growth in the relative demand for college workers, particularly since 1970. Analyses of four data sets indicate that the rate of skill upgrading has been greater in more computer-intensive industries.

This paper makes the case that purposive, profit-seeking investments in knowledge play a critical role in the long-run growth process. First, the authors review the implications of neoclassical growth theory and the more recent theories of ‘endogenous growth.’ Then they discuss the empirical evidence that bears on the modeling of long-run growth. Finally, the authors describe in more detail a model of growth based on endogenous technological progress and discuss the lessons that such models can teach us.

We analyze an economy where firms undertake both innovation and adoption of technologies from the world technology frontier. The selection of high-skill managers and firms is more important for innovation than for adoption. As the economy approaches the frontier, selection becomes more important. Countries at early stages of development pursue an investment-based strategy, which relies on existing firms and managers to maximize investment but sacrifices selection. Closer to the world technology frontier, economies switch to an innovation-based strategy with short-term relationships, younger firms, less investment, and better selection of firms and managers. We show that relatively backward economies may switch out of the investment-based strategy too soon, so certain policies such as limits on product market competition or investment subsidies, which encourage the investment-based strategy, may be beneficial. However, these policies may have significant long-run costs because they make it more likely that a society will be trapped in the investment-based strategy and fail to converge to the world technology frontier.

Many writers have claimed that research and development (R&D) has two faces. In addition to the conventional role of stimulating innovation, R&D enhances technology transfer (absorptive capacity). We explore this idea empirically using a panel of industries across twelve OECD countries. We find R&D to be statistically and economically important in both technological catch-up and innovation. Human capital also plays an major role in productivity growth, but we only find a small effect of trade. In failing to take account of R&D-based absorptive capacity, existing U.S.-based studies may underestimate the return to R&D.

Can directed technical change be used to combat climate change? We construct new firm-level panel data on auto industry innovation distinguishing between “dirty” (internal combustion engine) and “clean” (e.g., electric, hybrid, and hydrogen) patents across 80 countries over several decades. We show that firms tend to innovate more in clean (and less in dirty) technologies when they face higher tax-inclusive fuel prices. Furthermore, there is path dependence in the type of innovation (clean/dirty) both from aggregate spillovers and from the firm’s own innovation history. We simulate the increases in carbon taxes needed to allow clean technologies to overtake dirty technologies.

Demand for less-skilled workers plummeted in developed countries in the 1980s. In open economies, <it>pervasive</it> skill-biased technological change (SBTC) can explain this decline. SBTC tends to increase the domestic supply of unskill-intensive goods by releasing less-skilled labor. The more countries experiencing a SBTC, the greater its potential to decrease the relative wages of less-skilled labor by increasing the <it>world</it> supply of unskill-intensive goods. We find strong evidence for pervasive SBTC in developed countries. Most industries <it>increased</it> the proportion of skilled workers <it>despite</it> generally rising or stable relative wages. Moreover, the <it>same</it> manufacturing industries simultaneously increased demand for skills in <it>different</it> countries. Many developing countries also show increased skill premiums, a pattern consistent with SBTC.

Although the college-high school wage gap for younger U. S. men has doubled over the past 30 years, the gap for older men has remained nearly constant. In the United Kingdom and Canada the college-high school wage gap also increased for younger relative to older men. Using a model with imperfect substitution between similarly educated workers in different age groups, we argue that these shifts reflect changes in the relative supply of highly educated workers across age groups. The driving force behind these changes is the slowdown in the rate of growth of educational attainment that began with cohorts born in the early 1950s in all three countries.

This paper investigates the determination and consequences of organizational changes (OC) in a panel of British and French establishments. Organizational changes include the decentralization of authority, delayering of managerial functions, and increased multitasking. We argue that OC and skills are complements. We offer support for the hypothesis of “skill-biased” organizational change with three empirical findings. First, organizational changes reduce the demand for unskilled workers in both countries. Second, OC is negatively associated with increases in regional skill price differentials (a measure of the relative supply of skill). Third, OC leads to greater productivity increases in establishments with larger initial skill endowments. Technical change is also complementary with human capital, but the effects of OC is not simply due to its correlation with technological change but has an independent role.

Traditional growth theory emphasizes the incentives for capital accumulation rather than technological progress. Innovation is treated as an exogenous process or a by-product of investment in machinery and equipment. Grossman and Helpman develop a unique approach in which innovation is viewed as a deliberate outgrowth of investments in industrial research by forward-looking, profit-seeking agents.

This paper develops a search-theoretic model of technological change to explain why both patenting and the growth of productivity have remained roughly constant while research employment in the United States has increased by a factor of six over the past four decades. In the model, researchers sample from probability distributions determining the efficiency of potential new production techniques. Technological breakthroughs, resulting in patents, become increasingly hard to find as the level of technology advances. Given certain restrictions on the search distributions, the equilibrium of the model replicates the U.S. time-series pattern of research, patenting, and productivity.

Although innovation and the production of new goods and services have almost always been a part of economic activity, economic research on innovation has been to some extent

Long-run growth in many models is the product of two terms: the effective number of researchers and their research productivity. We present evidence from various industries, products, and firms showing that research effort is rising substantially while research productivity is declining sharply. A good example is Moore’s Law. The number of researchers required today to achieve the famous doubling of computer chip density is more than 18 times larger than the number required in the early 1970s. More generally, everywhere we look we find that ideas, and the exponential growth they imply, are getting harder to find. (JEL D24, E23, O31, O47)

This paper documents how plant-level wages, occupational mix, workforce education, and productivity vary with the adoption and use of new factory auto-mation technologies such as programmable controllers, computer-automated de-sign, and numerically controlled machines. Our cross-sectional results show that plants that use a large number of new technologies employ more educated work-ers, employ relatively more managers, professionals, and precision-craft workers, and pay higher wages. However, our longitudinal analysis shows little correlation between skill upgrading and the adoption of new technologies. It appears that plants that adopt new factory automation technologies have more skilled work-forces both pre- and postadoption. I.

This essay relates recent developments in growth theory to problems and ideas that first engaged R. F. Harrod, E. Domar, and their neoclassical successors. The body of ‘new growth theory’ began by finding special ways to assume that there are constant returns to capital. It is shown that this is a very nonrobust assumption, thus not a good basis for growth theory. More promising is the attempt to create a genuinely endogenous theory of the process of innovation. This notion has always been present in the literature or just beneath the surface. Current ideas, for all their ingenuity, may be too mechanical.

Abstract: This paper presents a simple R&amp;D-driven endogenous growth model to shed light on some puzzling economic trends. The model can account for why patent statistics have been roughly constant even though R&amp;D employment has risen sharply over the last 30 years. The model also illuminates why steadily increasing R&amp;D effort has not lead to any upward trend in economic growth rates, as is predicted by earlier R&amp;D-driven endogenous growth models with the “scale effect ” property.

Although surveys show that traditional orderings of average wage—i.e., higher earnings with higher schooling and concave age-wage profiles—have not changed during the past three decades, the actual size of the wage differentials measured by education or by work experience has varied from peak to trough by a factor of two-to-one. The patterns are not monotone, but there is a trend toward increased skill premiums. We first examine the structure of wages among white men distinguished by age and schooling for the period from 1963 to 1989. We then compare shifts in the distribution of wages and employment among the age x schooling categories to show in reference to a stable demand structure that employment alone cannot account for observed changes in relative wages. Finally, we describe the characteristics required of candidate demand shifters and offer examples using linear trend, business cycle shocks, and recent patterns of deficits in international trade.

We develop a methodology for testing Hicks's induced innovation hypothesis by estimating a product-characteristics model of energy-using consumer durables, augmenting the hypothesis to allow for the influence of government regulations. For the products we explored, the evidence suggests that (i) the <it>rate</it> of overall innovation was independent of energy prices and regulations; (ii) the <it>direction</it> of innovation was responsive to energy price changes for some products but not for others; (iii) energy price changes induced changes in the subset of technically feasible models that were offered for sale; (iv) this responsiveness increased substantially during the period after energy-efficiency product labeling was required; and (v) nonetheless, a sizable portion of efficiency improvements were autonomous.

The property that ideas are nonrivalrous leads to a tight link between idea-based growth models and increasing returns to scale. In particular, changes in the size of an economy’s population generally affect either the long-run growth rate or the long-run level of income in such models. This paper provides a partial review of the expanding literature on idea-based models and scale effects. It presents simple versions of various recent idea-based growth models and analyzes their implications for the relationship between scale and growth.

In this lecture, we use Schumpeterian growth theory, where growth comes from qualityimproving innovations, to elaborate a theory of growth policy and to explain the growth gap between Europe and the US. Our theoretical apparatus systematizes the case-by-case approach to growth policy design. The emphasis is on three policy areas that are potentially relevant for growth in Europe, namely: competition and entry, education, and macropolicy. We argue that higher entry and exit (higher firm turnover) and increased emphasis on higher education are more growth-enhancing in countries that are closer to the technological frontier. We also argue that countercyclical budgetary policies are more growth-enhancing in countries with lower financial development. The analysis thus points to important interaction effects between policies and state variables, such as distance to frontier or financial development, in growth regressions. Finally, we argue that the other endogenous growth models, namely the AK and product variety models, fail to account for the evidence on the relationship between competition, education, volatility, and growth, and consequently cannot deliver relevant policy prescriptions in the three areas we consider.

Rising educational attainment and research intensity in recent decades suggest that the U.S. economy is far from its steady state. This paper develops a model reconciling these facts with the stability of U.S. growth rates. In the model, long-run growth arises from the worldwide discovery of ideas, which depends on population growth. Nevertheless, constant growth can temporarily proceed at a faster rate, provided research intensity and educational attainment rise steadily over time. Growth accounting reveals that these factors explain 80 percent of recent U.S. growth, with less than 20 percent coming from world population growth.

Rapid advances in artificial intelligence (AI) and automation technologies have the potential to significantly disrupt labor markets. While AI and automation can augment the productivity of some workers, they can replace the work done by others and will likely transform almost all occupations at least to some degree. Rising automation is happening in a period of growing economic inequality, raising fears of mass technological unemployment and a renewed call for policy efforts to address the consequences of technological change. In this paper we discuss the barriers that inhibit scientists from measuring the effects of AI and automation on the future of work. These barriers include the lack of high-quality data about the nature of work (e.g., the dynamic requirements of occupations), lack of empirically informed models of key microlevel processes (e.g., skill substitution and human-machine complementarity), and insufficient understanding of how cognitive technologies interact with broader economic dynamics and institutional mechanisms (e.g., urban migration and international trade policy). Overcoming these barriers requires improvements in the longitudinal and spatial resolution of data, as well as refinements to data on workplace skills. These improvements will enable multidisciplinary research to quantitatively monitor and predict the complex evolution of work in tandem with technological progress. Finally, given the fundamental uncertainty in predicting technological change, we recommend developing a decision framework that focuses on resilience to unexpected scenarios in addition to general equilibrium behavior.

This article studies how portable skills accumulated in the labor market are. Using rich data on tasks performed in occupations, we propose the concept of task‐specific human capital to measure empirically the transferability of skills across occupations. Our results on occupational mobility and wages show that labor market skills are more portable than previously considered. We find that individuals move to occupations with similar task requirements and that the distance of moves declines with experience. We also show that task‐specific human capital is an important source of individual wage growth, accounting for up to 52% of overall wage growth.

This paper presents a Schumpeterian endogenous growth model in which a steady state exists with a constant growth rate even though population and the inputs to R. & D. are growing. The scale effect of rising population is nullified by product proliferation that fragments the growing demand for intermediate prodcuts, thus preventing the reward to any specific innovation from rising with population. All the ususal comparitive statics results of Schumpeterian growth theory are valid, including the positive effect of R. & D. subsidies on growth.

In 1960, 94 percent of doctors and lawyers were white men. By 2010, the fraction was just 62 percent. Similar changes in other highly‐skilled occupations have occurred throughout the U.S. economy during the last 50 years. Given that the innate talent for these professions is unlikely to have changed differently across groups, the change in the occupational distribution since 1960 suggests that a substantial pool of innately talented women and black men in 1960 were not pursuing their comparative advantage. We examine the effect on aggregate productivity of the convergence in the occupational distribution between 1960 and 2010 through the prism of a Roy model. Across our various specifications, between 20% and 40% of growth in aggregate market output per person can be explained by the improved allocation of talent.

We present a joint study of the U.S. structural transformation (the decline of agriculture as the dominating sector) and regional convergence (of southern to northern average wages). We find empirically that most of the regional convergence is attributable to the structural transformation: the nationwide convergence of agricultural wages to nonagricultural wages and the faster rate of transition of the southern labor force from agricultural to nonagricultural jobs. Similar results describe the Midwest's catch-up to the Northeast (but not the relative experience of the West). To explain these observations, we construct a model in which the South (Midwest) has a comparative advantage in producing unskilled laborintensive agricultural goods. Thus it starts with a disproportionate share of the unskilled labor force and lower per capita incomes. Over time, declining education/training costs induce an increasing proportion of the labor force to move out of the (unskilled) agricultural sector and into the (skilled) nonagricultural sector. The decline in the agricultural labor force leads to an increase in relative agricultural wages. Both effects benefit the South (Midwest) disproportionately since it has more agricultural workers. With the addition of a less than unit income elasticity of demand for farm goods and faster technological progress in farming than outside of farming, this model successfully matches the quantitative features of the U.S. structural transformation and regional convergence, as well as several other stylized facts on U.S. economic growth in the last century. The model does not rely on frictions on interregional labor and capital mobility, since in our empirical work we find this channel to be less important than the compositional effects the model emphasizes.

To explain why trade restrictions sometimes speed up worldwide growth and sometimes slow it down, we exploit an analogy with the theory or consumer behavior. Substitution effects make demand curves slope down, but income effects can increase or decrease the slope, and can sometimes overwhelm the substitution effect. We decompose changes in the worldwide growth rate into two effects (integration and redundancy) that unambiguously slow down growth, and a third effect (allocation) that can either speed it up or slow it down. We study two types of trade restrictions to illustrate the use of this decomposition. The First is across the board restrictions on traded goods in an otherwise perfect market. The second is selective protection of knowledge-intensive goods in a world with imperfect intellectual property rights. In both examples, we show that for trade between similar regions such as Europe and North America, the first two effects dominate; starting from free trade, restrictions unambiguously reduce worldwide growth. © 1991.

This paper uses a case study of Hong Kong and Singapore, two of the fastest growing economies in the postwar world, to evaluate competing theories of economic growth. Although broadly similar in historical background and economic structure, the two economies are strikingly different along three dimensions of interest to growth theorists: (1) Hong Kong began the postwar era with a considerably better educated labor force; (2) the subsequent accumulation of human and physical capital in Singapore far exceeds that in Hong Kong; and (3) Singapore has experienced considerably more rapid structural change, as government targeting policies have propelled the economy into technologically advanced sectors. A total factor productivity growth analysis of the two economies reveals that while TFP growth in Hong Kong accounts for over two-thirds of the increase in GDP per capita during the 1970s and 1980s, total factor productivity growth in Singapore during the same period is next to nil. These results constitute strong evidence against linear models of growth that emphasize contemporaneous externalities in the accumulation of factors of production. The poor TFP performance of the Singaporean economy, when associated with its astounding rate of structural transformation, supports models that emphasize the constraints imposed by learning by doing on the evolution of comparative advantage.

extends earlier work on the RID to patents relationship (Pakes-

In 1961, Nicholas Kaldor highlighted six “stylized” facts to summarize the patterns that economists had discovered in national income accounts and to shape the growth models being developed to explain them. Redoing this exercise today shows just how much progress we have made. In contrast to Kaldor's facts, which revolved around a single state variable, physical capital, our updated facts force consideration of four far more interesting variables: ideas, institutions, population, and human capital. Dynamic models have uncovered subtle interactions among these variables, generating important insights about such big questions as: Why has growth accelerated? Why are there gains from trade? (JEL D01, E01, E22, E23, E24, J11)

We show that skill requirements in job vacancy postings differentially increased in MSAs that were hit hard by the Great Recession, relative to less hard-hit areas. These increases persist through at least the end of 2015 and are correlated with increases in capital investments, both at the MSA and firm levels. We also find that effects are most pronounced in routine-cognitive occupations, which exhibit relative wage growth as well. We argue that this evidence is consistent with the restructuring of production toward routine-biased technologies and the more-skilled workers that complement them, and that the Great Recession accelerated this process. (JEL E24, E32, J24, J31, J63, L23, O33)

Beginning in the late 1980s, American corporations began increasingly linking the compensation of central research personnel to the economic objectives of the corporation. This paper examines the impact of the shifting compensation of the heads of corporate research and development. Among firms with centralized R&D organizations, a clear relationship emerges: more long-term incentives (such as stock options and restricted stock) are associated with more heavily cited patents. These incentives also appear to be associated with more patent awards and patents of greater originality. Short-term incentives appear to be unrelated to measures of innovation.

Several recent theories emphasize the negative effects of an aging population on economic growth, either because of the lower labor force participation and productivity of older workers or because aging will create an excess of savings over desired investment, leading to secular stagnation. We show that there is no such negative relationship in the data. If anything, countries experiencing more rapid aging have grown more in recent decades. We suggest that this counterintuitive finding might reflect the more rapid adoption of automation technologies in countries undergoing more pronounced demographic changes and provide evidence and theoretical underpinnings for this argument.

In this paper we study the dynamics of endogenous growth, both in the model of Lucas and in a generalization that incorporates a labor–leisure choice. We characterize the regions of the parameter space that give rise to unique equilibria as well as the regions that yield a continuum of equilibria with positive growth rates. We find that such multiple equilibria exist for empirically very plausible parameters, in particular when the Lucas model is modified to incorporate endogenous labor. Journal of Economic Literature Numbers: E00, E3, O40.

We study variation in skill demands for professionals across firms and labor markets. We categorize a wide range of keywords found in job ads into 10 general skills. There is substantial variation in these skill requirements, even within narrowly defined occupations. Focusing particularly on cognitive and social skills, we find positive correlations between each skill and external measures of pay and firm performance. We also find evidence of a cognitive social skill complementarity for both outcomes. As a whole, job skills have explanatory power in pay and firm performance regressions beyond what is available in widely used labor market data.

The college wage premium for new labor market entrants rose sharply during the 1980s. We ask how much of this change arose from changes in the skill level of the typical college graduate. We find that skills attained prior to college, as measured by standardized test scores and high school grades, had no effect on the change in the college wage premium for men. In contrast, the returns to math ability rose considerably for women; failing to account for math skills thus substantially overstates the growth in the female college wage premium. Skills acquired in college, as reflected in the distribution of students across majors, had important effects on the relative wages of men. The trend away from low-skill subjects such as education and toward high-skill subjects such as engineering accounts for one-fourth of the rise in the male college wage premium.

Abstract: In economics, interest has revived in economic growth, especially in long‐term convergence in per capita incomes and output between countries. This mainly empirical debate has promoted the development of endogenous growth theory, which seeks to move beyond conventional neoclassical theory by treating as endogenous those factors—particularly technological change and human capital— relegated as exogenous by neoclassical growth models. The economists at the forefront of the formulation of endogenous growth theory and the new growth empirics have begun to use long‐term regional growth patterns to test and develop their ideas. Their analyses suggest that regional convergence is a slow and discontinuous process. In this paper we consider whether endogenous growth theory can help to explain this finding. We argue that endogenous growth theory has important regional implications, but also major limitations when applied to a regional context. endogenous growth,

The transition from stagnation to growth and the associated phenomenon of the great divergence have been the subject of an intensive research in the growth literature in recent years. The discrepancy between the predictions of exogenous and endogenous growth models and the process of development over most of human history, induced growth theorists to advance an alternative theory that would capture in a single unified framework the contemporary era of sustained economic growth, the epoch of Malthusian stagnation that had characterized most of the process of development, and the fundamental driving forces of the recent transition between these distinct regimes. The advancement of unified growth theory was fueled by the conviction that the understanding of the contemporary growth process would be limited and distorted unless growth theory would be based on micro-foundations that would reflect the qualitative aspects of the growth process in its entirety. In particular, the hurdles faced by less developed economies in reaching a state of sustained economic growth would remain obscured unless the origin of the transition of the currently developed economies into a state of sustained economic growth would be identified, and its implications would be modified to account for the additional economic forces faced by less developed economies in an interdependent world. Unified growth theory suggests that the transition from stagnation to growth is an inevitable outcome of the process of development. The inherent Malthusian interaction between the level of technology and the size and the composition of the population accelerated the pace of technological progress, and ultimately raised the importance of human capital in the production process. The rise in the demand for human capital in the second phase of industrialization, and its impact on the formation of human capital as well as on the onset of the demographic transition, brought about significant technological advancements along with a reduction in fertility rates and population growth, enabling economies to convert a larger share of the fruits of factor accumulation and technological progress into growth of income per capita, and paving the way for the emergence of sustained economic growth. Variations in the timing of the transition from stagnation to growth and thus in economic performance across countries reflect initial differences in geographical factors and historical accidents and their manifestation in variations in institutional, social, cultural, and political factors. In particular, once a technologically driven demand for human capital emerged in the second phase of industrialization, the prevalence of human capital promoting institutions determined the extensiveness of human capital formation, the timing of the demographic transition, and the pace of the transition from stagnation to growth. © 2005 Elsevier B.V. All rights reserved.

When contracts are incomplete, dismissal laws prevent employers from arbitrarily discharging employees and thereby limit employers’ ability to hold up innovating employees after an innovation is successful. Therefore, dismissal laws can enhance employees’ innovative efforts and encourage firms to invest in risky but potentially groundbreaking projects. Other forms of labor laws that do not affect dismissal of employees do not have this bright side. We find support for these predictions in empirical tests that exploit country-level changes in dismissal laws in the United States, the United Kingdom, France, and Germany: more stringent dismissal laws foster innovation, particularly in innovation-intensive industries, but other labor laws do not.

Great achievements in knowledge are produced by older innovators today than they were a century ago. Nobel Prize winners and great inventors have become especially unproductive at younger ages. Meanwhile, the early life cycle decline is not offset by increased productivity beyond middle age. The early life cycle dynamics are closely related to age when the PhD was received, and I discuss a theory where knowledge accumulation across generations leads innovators to seek more education over time. More generally, the narrowing innovative life cycle reduces, other things equal, aggregate creative output. This productivity drop is particularly acute if innovators' raw ability is greatest when young. © 2010 The President and Fellows of Harvard College and the Massachusetts Institute of Technology.

This paper argues that long‐run trends in R&D and TFP are more supportive of fully endogenous “Schumpeterian” growth theory than they are of semi‐endogenous growth theory. The distinctive prediction of semi‐endogenous theory that sustained TFP growth requires sustained growth of R&D input is not supported by co‐integration tests and forecasting exercises, as TFP growth has been stationary even though the growth rate of R&D input has fallen three‐fold since the early 1950s. In contrast, the prediction of Schumpeterian theory that sustained TFP growth requires a sustained fraction of GDP to be spent on R&D is not contradicted by similar tests.

Abstract This paper presents a single unified framework that integrates the theoretical literature on Schumpeterian endogenous growth and major strands of the empirical literature on R&D, productivity growth and productivity convergence. Starting from a structural model of endogenous growth following Aghion and Howitt (1992, 1998) , we provide microeconomic foundations for the reduced‐form equations for total factor productivity (TFP) growth frequently estimated empirically using industry‐level data. R&D affects both innovation and the assimilation of others’ discoveries (“absorptive capacity”). Long‐run cross‐country differences in productivity emerge endogenously, and the analysis implies that many existing studies underestimate R&D's social rate of return by neglecting absorptive capacity.

We analyze a model economy with many agents, each with a different productivity level. Agents divide their time between two activities: producing goods with the production-related knowledge they already have and interacting with others in search of new, productivity-increasing ideas. These choices jointly determine the economy’s current production level and its rate of learning and real growth. We construct the balanced growth path for this economy. We also study the allocation chosen by an idealized planner who takes into account and internalizes the external benefits of search. Finally, we provide three examples of alternative learning technologies and show that the properties of equilibrium allocations are quite sensitive to two of these variations.

Using Data from the National Longitudinal Survey of Young Men, the author of this paper examines the relationship between predicted future earnings for five broad fields of study and college students' choice of major. Conditional logit models of major choice that incorporate alternative predicted earnings variables are specified and estimated. The results indicate that, holding family background characteristics constant, individuals are likely to choose majors offering greater streams of future earnings rather than, as some have argued, majors with higher beginning earnings at the time of the choice. It is also found that earnings profiles corrected for self-selection bias have flattened for more recent graduates in business, liberal arts, and education. The life-cycle earnings in these disciplines appear to be more severely depressed than those in science and engineering.

Human capital is supposed to be an important factor for innovation and economic development. However, the long-run impact of human capital on current innovation and economic development is still a black box, in particular at the regional level. Therefore, this paper makes the link between the past and the present. Using a large new dataset on regional human capital and other factors in the 19th and 20th century, we find that past regional human capital is a key factor explaining current regional disparities in innovation and economic development.

We present a task-based model in which high- and low-skill workers compete against machines in the production of tasks. Low-skill (high-skill) automation corresponds to tasks performed by low-skill (high-skill) labor being taken over by capital. Automation displaces the type of labor it directly affects, depressing its wage. Through ripple effects, automation also affects the real wage of other workers. Counteracting these forces, automation creates a positive productivity effect, pushing up the price of all factors. Because capital adjusts to keep the interest rate constant, the productivity effect dominates in the long run. Finally, low-skill (high-skill) automation increases (reduces) wage inequality.

An endogenous growth model is developed that produces convergence in per capita income and growth rates of output. Agents have identical preferences and access to identical technologies of production and investment, but differing levels of initial human capital. A spillover effect of human capital in the investment technology provides below-average human capital agents with a higher rate of return on investment than above-average human capital agents. Thus, below-average human capital agents grow faster than above-average human capital agents. This model explains income convergence of the developed world, regional income convergence within the United States, and intergenerational mobility. Copyright 1991 by University of Chicago Press.

The authors construct a model of the product cycle featuring endogenous innovation and technology transfer. Competitive entrepreneurs in the industrialized North introduce new products whenever the expected present value of oligopoly profits exceeds the cost of product development. In the middle-income South, entrepreneurs devote resources to learning the production processes that have been developed in the North. The authors study the determinants of the long-run rate of growth of the world economy and the long-run rate of imitation. They also study the effects of exogenous events and of public policy on relative wage rates in the two regions. Copyright 1991 by Royal Economic Society.

In this paper, we review the literature on declining business dynamism and its implications in the United States and propose a unifying theory to analyze the symptoms and the potential causes of this decline. We first highlight 10 pronounced stylized facts related to declining business dynamism documented in the literature and discuss some of the existing attempts to explain them. We then describe a theoretical framework of endogenous markups, innovation, and competition that can potentially speak to all of these facts jointly. We next explore some theoretical predictions of this framework, which are shaped by two interacting forces: a composition effect that determines the market concentration and an incentive effect that determines how firms respond to a given concentration in the economy. The results highlight that a decline in knowledge diffusion between frontier and laggard firms could be a significant driver of empirical trends observed in the data. This study emphasizes the potential of growth theory for the analysis of factors behind declining business dynamism and the need for further investigation in this direction. (JEL D33, E25, J24, L13, O33, O34)

In this lecture, it is argued that Schumpeterian Growth Theory, in which growth is driven by a sequence of quality-improving innovations, can shed light on two important puzzles raised by the recent evolution of wage inequality in developed economies. The first puzzle concerns wage inequality between educational groups, which has substantially risen in the US and the UK during the past two decades following a sharp increase in the supply of educated labor. The second puzzle concerns wage inequality within educational groups, which accounts for a large fraction of the observed increase in wage inequality, although in contrast to between-group wage inequality it has mainly affected the temporary component of income.

This paper constructs a dynamic general equilibrium model in which economic growth is explained by the evolution of the division of labor. The relationships among the accumulation of human capital, the evolution of the division of labor, endogenous comparative advantage, trade dependence, the market structure, and economic growth are investigated.

This paper examines the determinants of the choice of the college major when the length of studies and future earnings are uncertain. We estimate a three-stage schooling decision model, focusing on the effect of expected earnings on major choice. We control for dynamic selection through the use of mixture distributions. Exploiting variations across the French business cycle in the relative returns to the majors, our results yield a very low, though significant, elasticity of major choice to expected earnings. This suggests that at least for the French university context, nonpecuniary factors are a key determinant of schooling choices.

Is long-run economic growth exogenous? To address this question, we show that the empirical framework of Mankiw, Romer, and Weil (1992) can be extended to test any growth model that admits a balanced growth path, and we use that framework both to revisit variants of the Solow growth model and to evaluate simple alternative models of endogenous growth. To allow for the possibility that economies in our sample are not on their balanced growth paths, we also study the cross-sectional behavior of total-factor-productivity growth, which we estimate using alternative measures of labor's share. Our broad conclusion, based on both model estimation and growth accounting, is that long-run growth is significantly correlated with behavioral variables such as the savings rate, and that this correlation is not easily explained by models in which growth is treated as the exogenous variable. Hence, future empirical studies should focus on models that exhibit endogenous growth.

This article investigates patent protection for a long sequence of innovations where firms repeatedly supersede each other. Incentives for R&D can be insufficient if successful firms earn market profit only until competitors achieve something better. To correct this problem, patents must provide protection against future innovators. This article proposes using a patentability requirement -- a minimum innovation size required for patents. A patentability requirement can stimulate R&D investment and increase dynamic efficiency. Intuitively, requiring firms to pursue larger innovations prolongs market incumbency because larger innovations are harder to achieve, and longer market incumbency implies an increased reward to innovation.

We construct a Schumpeterian growth theory consistent with the divergence in per-capita income that has occurred between countries since the mid 19th Century, and with the convergence that occurred between the richest countries during the second half of the 20th Century. The theory assumes that technological change underwent a transformation late in the 19th Century, associated with modern R&D labs. Countries sort themselves into three groups. Those in the highest group converge to a steady state where they do leading edge R&D, while those in the intermediate group converge to a steady state where they implement technologies developed elsewhere. Countries in both of these groups grow at the same rate in the long run, as a result of technology transfer, but inequality between them increases during the transition. Countries in the lowest group grow at a slower rate, with relative incomes that fall asymptotically to zero. Once modern R&D has been introduced, a country may have only a finite window of opportunity in which to introduce the institutions that support it.

This paper presents empirical estimates of human-capital augmented growth equations for a panel of 21 OECD countries over the period 1971-98.It uses an improved dataset on human capital and a novel econometric technique that reconciles growth model assumptions with the needs of panel data regressions.Unlike several previous studies, our results point to a positive and significant impact of human capital accumulation to output per capita growth.The estimated long-run effect on output of one additional year of education (about 6 per cent) is also consistent with microeconomic evidence on the private returns to schooling.We also found a significant growth effect from the accumulation of physical capital and a speed of convergence to the steady state of around 15 per cent per year.Taken together these results are not consistent with the human capital augmented version of the Solow model, but rather they support an endogenous growth model à la Uzawa-Lucas, with constant returns to scale to "broad" (human and physical) capital.

Human capital investments are not independent of the aggregate state of labor markets: frictions and slackness of the labor market raise the returns to specific human capital investments relative to general investments. We build a macroeconomic model with two pure strategy regimes. In the pure G-regime, workers invest in general skills. This occurs when they face high turnover labor markets and in the absence of employment protection. The pure 5-regime in which workers invest in skills specific to their job appears when employment protection is high enough. Implications for a characterization of Europe-United States differences are provided in conclusion.

We show that technology spillovers shift the composition of corporate research and development by promoting innovation based on the exploitation of existing knowledge while disincentivizing innovation that explores new areas and breaks new ground. Accordingly, firms facing large technology spillovers attain fewer superstar inventors among their human capital, who are important drivers of breakthrough technology advancement. These findings complement the existing studies documenting the positive effect of technology spillovers in increasing firms’ overall innovation outputs; they highlight potential downsides of technology spillovers in reducing firm investment in technology breakthrough and valuable human capital. This paper was accepted by Gustavo Manso, finance.

Schumpeterian growth theory has operationalized Schumpeter's notion of creative destruction by developing models based on this concept. These models shed light on several aspects of the growth process that could not be properly addressed by alternative theories. In this survey, we focus on four important aspects, namely: (i) the role of competition and market structure; (ii) firm dynamics; (iii) the relationship between growth and development with the notion of appropriate growth institutions; and (iv) the emergence and impact of long-term technological waves. In each case, Schumpeterian growth theory delivers predictions that distinguish it from other growth models and which can be tested using micro data. © 2014 Elsevier B.V.

We build a model of firm-level innovation, productivity growth and reallocation featuring endogenous entry and exit. A new and central economic force is the selection between high-and low-type firms, which differ in terms of their innovative capacity. We estimate the parameters of the model using US Census micro data on firm-level output, R&D and patenting. The model provides a good fit to the dynamics of firm entry and exit, output and R&D. Taxing the continued operation of incumbents can lead to sizable gains (of the order of 1.4% improvement in welfare) by encouraging exit of less productive firms and freeing up skilled labor to be used for R&D by high-type incumbents. Subsidies to the R&D of incumbents do not achieve this objective because they encourage the survival and expansion of low-type firms.

This paper introduces into Schumpeterian growth theory an important element of heterogeneity in the structure of innovative activity-namely, the distinction between research and development. We construct a simple model of growth to investigate how the (steady-state) rate of growth affects and is affected by the relative mix between research and development. Although we assume for simplicity that the total supply of innovative activity is given it turns out that, with one important exception, the growth rate responds to most parameter changes in the same way as in previous models where growth was determined by the total amount of innovative activity. In particular, the level of research tends to covary positively with the rate of growth, even in the extreme case where the general knowledge that underlies long-run growth is created only by secondary innovations arising from the development process. The exception concerns the effects of competition on growth. Although simpler Schumpeterian growth models implied that increased competition would reduce growth by reducing the incentive to innovate, introducing the distinction between research and development implies that this effect is likely to be reversed.

We set up an endogenous growth model with physical capital, human capital and blueprints for intermediate goods. The model can generate steady-state growth or stagnation. Along the adjustment path for a developing economy we can distinguish different stages of development. The first stage is characterized by physical factor accumulation. At the second stage the economy follows a growth path which is mainly characterized by the accumulation of skills. Growth of the fully developed economy is identified by an increasing variety of goods originating from costly R and D efforts. Transition to a higher stage of development is explained endogenously. Thus, the model provides a high degree of generality by encompassing the standard neoclassical growth model and modern endogenous growth theory. (C) 2000 Elsevier Science B.V. All rights reserved.

Abstract. I use U.S. manufacturing industry data to estimate a system of three equations implied by a model of R&D‐induced growth in steady state. These equations relate R&D intensity to patenting, patenting to technological progress, and technological progress to economic growth. In each case, I find evidence of positive impact. Thus, I reject the null hypothesis that growth is not induced by R&D in favour of the Schumpeterian endogenous growth framework without scale effects. I also find strong support for technological spillovers from aggregate research intensity to industry‐level innovation success. JEL Classification: O40, O30

In this paper we provide an analysis of directed technical change in the sector of electricity generation. We rely on patent data in fossil-fuel (FF) and renewable energy (REN) technologies for 5471 European firms over the 1978-2006 period. The novelty of our approach is in the focus on firm's heterogeneity in driving technological change. We make a distinction between small specialized firms, which innovate in only one type of technology, and large mixed firms, which innovate in both technologies, to analyse how REN patents can replace FF ones at the sector level both through a shift in innovation activities within existing firms and through firms' entry and exit. We use zero-inflated count data estimation techniques to identify the factors that affect specialized versus mixed firms' patenting behaviour both at the intensive (i.e., levels of innovation) and extensive (i.e., technological entry) margins. We further investigate the implications of our firm-level estimations for reducing the gap between REN and FF innovation at the aggregate level. We establish two key findings: (1) a decrease in the FF-REN technology gap mainly comes about through technological entry of specialized REN firms following an increase in REN market size; (2) increases in FF prices, FF market size, and FF knowledge stocks all increase the technology gap by increasing mixed firms FF innovation rates. An important implication of our results is that policies aimed at increasing REN innovation should focus on helping small firms to start and sustain innovation in the long-run.

We develop a model of growth driven by successive improvements in 'General Purpose Technologies' (GPT's), such as the steam engine, electricity, or micro-electronics. Each new generation of GPT's prompts investments in complementary inputs, and impacts the economy after enough such compatible inputs become available. The long-run dynamics take the form of recurrent cycles: during the first phase of each cycle output and productivity grow slowly or even decline, and it is only in the second phase that growth starts in earnest. The historical record of productivity growth associated with electrification, and perhaps also of computerization lately, may offer supportive evidence for this pattern. In lieu of analytical comparative dynamics, we conduct simulations of the model over a wide range of parameters, and analyze the results statistically. We extend the model to allow for skilled and unskilled labor, and explore the implications for the behavior over time of their relative wages. We also explore diffusion in the context of a multi-sector economy.

We examine differences in minority science graduation rates among University of California campuses when racial preferences were in place. Less prepared minorities at higher ranked campuses had lower persistence rates in science and took longer to graduate. We estimate a model of students' college major choice where net returns of a science major differ across campuses and student preparation. We find less prepared minority students at top ranked campuses would have higher science graduation rates had they attended lower ranked campuses. Better matching of science students to universities by preparation and providing information about students' prospects in different major-university combinations could increase minority science graduation.

In this article we estimate the dynamic relationship between employment in R&D and generation of knowledge as measured by patent applications across OECD countries. In several recently developed models, known as 'idea-based' models of growth, the 'idea-generating' process is the engine of productivity growth. Moreover, in real business cycle models technological shocks are an important source of fluctuations. Our empirical strategy is able to test whether knowledge spillovers are strong enough to generate sustained endogenous growth and to estimate the quantitative impact of international knowledge on technological innovation of a country in the short and in the long run. Copyright 2007 The Author(s). Journal compilation Royal Economic Society 2007.

This paper develops a tractable incomplete-markets model of economic growth in which households invest in risk-free physical capital and risky human capital. The paper shows that a reduction in uninsurable idiosyncratic labor income risk decreases physical capital investment, but increases human capital investment, growth, and welfare. A quantitative analysis based on a calibrated version of the model reveals that these effects are substantial and of the same order of magnitude as the effects of distortionary income taxation. The analysis further suggests that government-sponsored severance payments to displaced workers increase growth and welfare even if these payments have to be financed through distortionary income taxation. I.

This paper defines and estimates general equilibrium treatment effects. The conventional approach in the literature on treatment effects ignores interactions among individuals induced by the policy interventions being studied. Focusing on the impact of tuition policy, and using estimates from our dynamic overlapping generations general equilibrium model of capital and human capital formation, we find that general equilibrium impacts of tuition on college enrollment are an order of magnitude smaller than those reported in the literature on microeconomic treatment effects. The assumptions used to justify the LATE parameter in a partial equilibrium setting do not hold in a general equilibrium setting. Policy changes induce two way flows. We extend the LATE concept to a general equilibrium setting. We present a more comprehensive evaluation to program evaluation by considering both the tax and benefit consequences of the program being evaluated and placing the analysis in a market setting.

These sixteen essays are drawn from a body of work strongly influenced by the thought of Joseph A. Schumpeter. They are particularly appropriate in a time when low rates of growth have become the norm in the Western world and much of the economic debate focuses on prescriptions for industrial regeneration. Each essay tests hypotheses derived from the Schumpeterian propositions that technological innovation gives capitalist economies their peculiar dynamics through a process of "creative destruction, " that technological progress has radically increased real income per capita in Western industrialized nations, and that monopoly market structures and their pursuit are a powerful engine of technological progress.

We construct a structural model of on-the-job search in which workers differ in skills along several dimensions and sort themselves into jobs with heterogeneous skill requirements along those same dimensions. Skills are accumulated when used, and depreciate when not used. We estimate the model combining data from O*NET with the NLSY79. We use the model to shed light on the origins and costs of mismatch along heterogeneous skill dimensions. We highlight the deficiencies of relying on a unidimensional model of skill when decomposing the sources of variation in the value of lifetime output between initial conditions and career shocks. (JEL J24, J41, J64)

We study asset-pricing implications of innovation in a general-equilibrium overlapping-generations economy. Innovation increases the competitive pressure on existing firms and workers, reducing the profits of existing firms and eroding the human capital of older workers. Due to the lack of inter-generational risk sharing, innovation creates a systematic risk factor, which we call "displacement risk." This risk helps explain several empirical patterns, including the existence of the growth-value factor in returns, the value premium, and the high equity premium. We assess the magnitude of displacement risk using estimates of inter-cohort consumption differences across households and find support for the model. © 2012 Elsevier B.V.

Modern Schumpeterian growth theory focuses on the product line as the main locus of innovation and exploits endogenous product proliferation to sterilize the scale effect. The empirical core of this theory consists of two claims: (i) growth depends on average employment (i.e., employment per product line); (ii) average employment is scale invariant. We show that data on employment, RandD personnel, and the number of establishments in the US for the period 1964-2001 provide strong support for these claims. While employment and the total number of R&D workers increase with no apparent matching change in the long-run trend of productivity growth, employment and RandD employment per establishment exhibit no long-run trend. We also document that the number of establishments, employment and population exhibit a positive trend, while the ratio employment/establishment does not. Finally, we provide results of time series tests consistent with the predictions of these models. © Springer Science+Business Media, LLC 2006.

This paper proposes a model of endogenous growth where innovating requires both researchers, who produce inventions, and entrepreneurs who implement them. As research and entrepreneurship compete in the allocation of aggregate resources, the relation between growth and research effort is hump‐shaped. When entrepreneurs appropriate too little rents from innovation, too few resources are allocated to entrepreneurship and returns to R&D are low because of this lack of entrepreneurial skills. When so, innovation should be promoted by encouraging entrepreneurship rather than research.

Modern growth theory suggests that more than three-quarters of growth since 1950 reflects rising educational attainment and research intensity. As these transition dynamics fade, US economic growth is likely to slow at some point. However, the rise of China, India, and other emerging economies may allow another few decades of rapid growth in world researchers. Finally, and more speculatively, the shape of the idea production function introduces a fundamental uncertainty into the future of growth. For example, the possibility that artificial intelligence will allow machines to replace workers to some extent could lead to higher growth in the future.

Using the text from job ads, we introduce a new dataset to describe the evolution of work from 1950 to 2000. We show that the transformation of the US labor market away from routine cognitive and manual tasks and toward nonroutine interactive and analytic tasks has been larger than prior research has found, with a substantial fraction of total changes occurring within narrowly defined job titles. We provide narrative and systematic evidence on changes in task content within job titles and on the emergence and disappearance of individual job titles. (JEL E24, J21, J24, J31, N32)

The intellectual breakthrough contributed by the new growth theory was the recognition that investments in knowledge and human capital endogenously generate economic growth through the spillover of knowledge. Endogenous growth theory does not explain how or why spillovers occur. The missing link is the mechanism converting knowledge into economically relevant knowledge. This Paper develops a model that introduces a filter between knowledge and economic knowledge and identifies entrepreneurship as a mechanism that reduces the knowledge filter. A cross-country regression analysis over the period 1981-2001 provides empirical support for the model. We conclude that public policies facilitating knowledge spillovers through entrepreneurship may be an important new approach to promoting economic growth.

Abstract This article introduces a general equilibrium model of endogenous technical change through basic and applied research. Basic research differs from applied research in the nature and the magnitude of the generated spillovers. We propose a novel way of empirically identifying these spillovers and embed them in a framework with private firms and a public research sector. After characterizing the equilibrium, we estimate our model using micro-level data on research expenditures by French firms. Our key finding is that uniform research subsidies can accentuate the dynamic misallocation in the economy by oversubsidizing applied research. Policies geared towards public basic research and its interaction with the private sector are significantly welfare-improving.

We study the expected welfare effects of targeted R&D subsidies using project-level data from Finland. We model the application and R&D investment decisions of firms and the subsidy-granting decision of the public agency in charge of the program. Our model and institutional environment allow us to identify different benefits and costs of the R&D subsidy program. We find that expected effects of subsidies are very heterogeneous and estimated application costs low on average. The social rate of return on targeted subsidies is 30% to 50%, but spillover effects of subsidies are smaller than effects on firm profits.

ABSTRACT In this paper, a formal model for the relationship between innovation and growth in European Union regions is developed drawing upon the theoretical contribution of the systems of innovation approach. The model combines the analytical approach of the regional growth models with the insights of the systemic approach. The cross-sectional analysis, covering all the Enlarged Europe (EU-25) regions (for which data are available), shows that regional innovative activities (for which a specific measure is developed) play a significant role in determining differential regional growth patterns. Furthermore, the model sheds light on how geographical accessibility and human capital accumulation, by shaping the regional system of innovation, interact (in a statistically significant way) with local innovative activities, thus allowing them to be more (or less) effectively translated into economic growth. The paper shows that an increase in innovative effort is not necessarily likely to produce the same effect in all EU-25 regions. Indeed, the empirical analysis suggests that in order to allow innovative efforts in peripheral regions to be as productive as in core areas, they need to be complemented by huge investments in human capital.

We develop a quantitative macroeconomic theory of input-saving technical change to analyze how markets economize on scarce natural resources, with an application to fossil fuel. We find that aggregate US data call for a very low short-run substitution elasticity between energy and the capital/labor inputs. Our estimates imply that energy-saving technical change took off when the oil shocks hit in the 1970s. This response implies significant substitutability with the other inputs in the long run: even under ever-rising energy prices, long-run consumption growth is still possible, along with a modest factor share of energy.

In this paper, the authors analyze the speed of convergence to a balanced path in a class of endogenous growth models with physical and human capital. They show that such rate depends locally on the technological parameters of the model but does not depend on preferences parameters. This result stands in sharp contrast with that of the one-sector neoclassical growth model, where both preferences and technologies determine the speed of convergence to a steady-state growth path. Copyright 1997 by American Economic Association.

In this review, we argue that the Schumpeterian growth paradigm, which models growth as resulting from innovations involving creative destruction, sheds light on several aspects of the growth process that cannot be properly addressed by alternative theories. We focus on three important aspects for which Schumpeterian growth theory delivers predictions that distinguish it from other growth models, namely, (a) the role of competition and market structure, (b) firm dynamics, and (c) the relationship between growth and development.

One of the most well‐known empirical regularities in the R&D‐productivity literature is the existence of substantial under‐investment in R&D. This strongly suggests that government should actively promote research activities. However, the so‐called ‘quality‐ladders’ models of endogenous technological progress are inconsistent with this observation. In an extreme case, Grossman and Helpman (1991, Ch.4) suggest that R&D should always be taxed irrespective of the size of quality improvement. This paper attempts to reconcile these empirical and theoretical findings by showing that the normative results of Grossman and Helpman are not robust.

Development is associated with systematic changes in the firm size distribution. I document that the mean and dispersion of firm size are larger in rich countries, and increased over time for US firms. To analyze the firm size-development link, I construct a frictionless general equilibrium model of occupational choice with skill-biased change in entrepreneurial technology (i.e., technical progress favors better entrepreneurs). The model accounts for key aspects of the US experience with only changes in aggregate technology. It attributes half the variation in mean and dispersion of firm size across countries to technical change. Distortions also affect the size distribution. (JEL J24, L11, L25, L26, O33)

Many view the period after the Second Industrial Revolution as a paradigm of a transition to a new economy following a technological revolution, including the Information Technology Revolution. We build a quantitative model of diffusion and growth during transitions to evaluate that view. With a learning process quantified by data on the life cycle of US manufacturing plants, the model accounts for the key features of the transition after the Second Industrial Revolution. But we find that features like those will occur in other transitions only if a large amount of knowledge about old technologies exists before the transition begins. (JEL L60, N61, N62, N71, N72, O33)

Conventional growth accounting exercises are extended in this paper to allow for endogeneity of capital, demographic transitions, age dependency, and employment rates, among other factors. Using data for the OECD countries in the period 1870-2006 it is shown that growth has been predominantly driven by demographics and TFP growth. TFP has, in turn, been driven by R&D, knowledge spillovers through the channel of imports, educational attainment, and the interaction between educational attainment and the distance to the technology frontier. The estimates suggest permanent growth effects of R&D and human capital. © 2010 Elsevier B.V.

Schumpeterian growth theory has "operationalized" Schumpeter''s notion of creative destruction by developing models based on this concept. These models shed light on several aspects of the growth process which could not be properly addressed by alternative theories. In this survey, we focus on four important aspects, namely: (i) the role of competition and market structure; (ii) firm dynamics; (iii) the relationship between growth and development with the notion of appropriate growth institutions; (iv) the emergence and impact of long-term technological waves. In each case Schumpeterian growth theory delivers predictions that distinguish it from other growth models and which can be tested using micro data.

We introduce firm and worker heterogeneity into a model of innovation†driven endogenous growth. Individuals who differ in ability sort into either a research activity or a manufacturing sector. Research projects generate new varieties of a differentiated product. Projects differ in quality and the resulting technologies differ in productivity. In both sectors, there is a complementarity between firm quality and worker ability. We study the co†determination of growth and income inequality in both the closed and open economy, as well as the spillover effects of policy in one country to outcomes in others.

Missing from recent discussions of tax reform is any systematic analysis of the effects of various tax proposals on skill formation. This gap in the literature in empirical public finance is due to the absence of any empirically based general equilibrium models with both human capital formation and physical capital formation that are consistent with observations on modern labor markets. This paper is a progress report on our ongoing research on formulating and estimating dynamic general equilibrium models with endogenous heterogeneous human capital accumulation. Our model explains many features of rising wage inequality in the U.S. economy (James Heckman, Lance Lochner and Christopher Taber, 1998). In this paper, we use our model to study the impacts on skill formation of proposals to switch from progressive taxes to flat income and consumption taxes. For the sake of brevity, we focus on steady states in this paper, although we study both transitions and steady states in our research.

As the workforce has become more educated, educational decisions are about what type of education to pursue as well as how much to pursue. In college, individuals somewhat specialize through their choice of college major. Further specialization occurs in graduate school. This chapter investigates how majors and graduate school affect labor market outcomes, as well as how individuals make these potentially important decisions. To do so, we develop a dynamic model of educational decision-making. In light of the model, we examine the estimation issues associated with obtaining causal effects of educational choices on earnings. We then examine ways that authors have overcome the selection problem, as well as the approaches authors have taken to estimate the process by which these educational decisions are made. © 2016 Elsevier B.V.

We test whether predictions of the Aghion et al. (Aghion, P., Bloom, N., Blundell, R., Griffith, R. and Howitt, P. (2004) Competition and Innovation: An Inverted U Relationship. NBER Working Paper series, No. 9269.) model are supported by firm-level data. In particular, we analyze if there is an inverted U-shaped relation between competition and R&D. Results show that the inverted U-shaped relation is supported by the Herfindahl index but not by the price cost margin. Using the Herfindahl index, results suggest that breaking up monopolies increases R&D, whereas further increases in competition most likely lead to reduced R&D. Comparing different estimators, we find that time series-based estimators typically result in less clear-cut results, probably driven by a lack of time series variation in measures of competition.

In this paper, we explore the dynamic properties of an endogenous growth model with finite patent length. We show that there exists a unique equilibrium growth path and that this path exhibits damped oscillations in contrast to the equilibrium path of an endogenous growth model with infinite patent length. We also examine the effects of patent policy on social welfare and show that infinite patent length does not maximize social welfare. Furthermore, we show that, in a growth model that does not exhibit scale effects, a finite patent length maximizes social welfare on the balanced growth path. © 2005 Elsevier Inc. All rights reserved.

The study investigates the drivers of total factor productivity (TFP) growth, covering 99 European regions from 31 countries over the period 2000–13. It shows that human capital endowment had a positive effect upon TFP growth, particularly in advanced regions, but the effect from regions’ own research and development (R&D) expenditures was largely absent. The effects of human capital and R&D on TFP growth varied with the productivity gap. Further, there was a threshold effect in convergence, where stronger TFP growth was associated with both a larger productivity gap and a higher initial level of productivity. Spatial spillover effects had a positive impact upon TFP growth.

This paper investigates the impact of rank-and-file employees on corporate innovation. We show that paying higher relative wages to rank-and-file employees promotes better innovation outcomes in terms of patent quantity and quality. This effect is more significant among firms with large proportions of skilled employees, industries with high levels of R&D intensity, provinces with competitive local labor markets, and non-SOEs. Further analyses reveal that efficiency wages can serve as an underlying economic channel that fosters innovation by retaining and attracting valuable human capital and stimulating their working enthusiasm. Finally, we show that technological innovation is a mechanism through which rank-and-file employees affect productivity growth and thereby affect the economy.

This paper studies interactions between innovation, public capital, and human capital in an OLG model of endogenous growth. Public capital affects growth not only through productivity, but also through innovation capacity and human capital accumulation. Numerical simulations, based on a calibrated version of the model, are used to illustrate these channels. Panel data regressions are presented next; they show that higher innovation performance promotes growth directly, whereas public capital has both direct and indirect growth effects by promoting human capital accumulation and innovation capacity. Elasticity estimates derived from simultaneous equation techniques show that the general equilibrium effects of public capital on steady-state output per capita (which account for indirect effects) are significantly higher than those derived from single equation methods.

We show empirically that high‐risk sectors, which contribute strongly to aggregate productivity growth, are relatively small and have relatively low productivity growth in countries with strict employment protection legislation (EPL). To understand these findings, we develop a two‐sector matching model where firms endogenously choose between a safe technology and a risky technology. For firms that have chosen the risky technology, EPL raises the costs of shedding workers in case they receive a low productivity draw. According to our calibrated model, high‐EPL countries benefit less from the arrival of new risky technologies than low‐EPL countries. Parameters estimated through reduced‐form regressions of employment and productivity on exit costs, riskiness, and in particular their interaction are qualitatively similar for actual cross‐country data and simulated model data. Our model is consistent with the slowdown in productivity in the European Union relative to the United States since the mid‐1990s.

College graduates do not always pursue careers related to their major. Science majors working in jobs unrelated to their field of study earn approximately 30% lower wages than those working in related jobs. We develop a structural model of major choice and labor market outcomes that allows for skill uncertainty and differential accumulation of human capital across major. Our findings confirm that the average return to obtaining a science degree and working in a related job remains close to 30%. We also find that individuals are uncertain about their future productivity at the time of the college major decision.

We formulate a simple multiagent evolutionary scheme as a model of collective learning, i.e. a situation in which firms experiment, interact, and learn from each other. This scheme is then applied to a stylized endogenous growth economy in which firms have to determine how much to invest in R&D, where innovations are the stochastic product of their R&D activity, spillovers occur, but technological advantages are only relative and temporary and innovations actually diffuse, both at the intra and interfirm levels. The model demonstrates both the existence of a unique long-run growth attractor (in the linear case) and distinct growth phases on the road to that attractor. We also compare the long-run growth patterns for a linear and a logistic innovation function, and produce some evidence for a bifurcation in the latter case. © 1994 Springer-Verlag.

This paper introduces a new internationally comparable data set that permits an empirical investigation of the effects of patent law on innovation. The data have been constructed from the catalogues of two 19th century world fairs: the Crystal Palace Exhibition in London, 1851, and the Centennial Exhibition in Philadelphia, 1876. They include innovations that were not patented, as well as those that were, and innovations from countries both with and without patent laws. I find no evidence that patent laws increased levels of innovative activity but strong evidence that patent systems influenced the distribution of innovative activity across industries. Inventors in countries without patent laws concentrated in industries where secrecy was effective relative to patents, e.g., food processing and scientific instruments. These results suggest that introducing strong and effective patent laws in countries without patents may have stronger effects on changing the direction of innovative activity than on raising the number of innovations.

Policy can eliminate technology-induced joblessness. Labor can gain from labor-saving and capitalsaving technologies if its supply is less elastic than capital's. Skill-biased technical change could raise the relative demand for skilled workers faster than the supply of skilled workers increases. Workers can earn more of their income from capital than from working-by owning part of the robots that replace them.

We explore the relationship between human capital and firms’ innovation in emerging economies. Most papers consider the formal knowledge developed in R&D laboratories as a major source of innovation. However, a critical portion of knowledge required for innovation resides in human resources and is created outside any formalised R&D activity. We consider that, to improve their technological capabilities, firms should invest in different forms of human capital, namely highly educated workforce and experienced managers, but also in strategic human resource (HR) practices aimed at developing human capital by increasing employees’ firm-specific technical skills and competences. Besides looking at the type of innovation outcomes, we place greater emphasis on the strategies of innovation development, as these should signal an improved firms’ ability, not just to innovate, but to put their own creative effort in the development of innovation. Our results contrast with the traditional view of firms in emerging economies as mainly relying on the external acquisition of innovations, by showing their actual ability to develop new technologies. In this respect, HR practices aimed at fostering employees’ learning and autonomy at work appear more important than the educational attainment of workers, whilst the experience of managers does not seem effective.

Despite the significance of limited labor mobility across sectors, few attempts have been made to produce dynamic models of sectoral adjustment that are consistent with perfect foresight and, yet, flexible enough to allow for a variety of dynamic experiments. This paper proposes a simple perfect-foresight model of two-sector economies in which aggregate sectoral movement of labor takes place through the process of demographic change. The model is tractable enough that one can easily examine the effects of intertemporally complicated relative price shocks (both exogenous and endogenous) under a variety of assumptions on technology. Copyright 1992 by The Review of Economic Studies Limited.

This paper examines the productivity growth effects of educational attainment and its interaction with the distance to the world technology frontier, which is the percentage distance to the country with the highest total factor productivity (TFP) (the United Kingdom or United States), while allowing for the endogeneity of educational attainment in some of the estimates. For this purpose, a new annual data set for educational attainment is constructed for 21 industrialized countries over the period from 1870 to 2009. The results show that changes in educational attainment and the interaction between education and the distance to the frontier, as predicted by Schumpeterian growth theory, have been influential for productivity growth over the past 140 years.

We present an endogenous growth model in which some firms devote resources to developing higher-quality products (innovative R&D) and other firms devote resources to copying these products (imitative R&D). Although consumers benefit from the knowledge created by both types of R&D activities, only innovative R&D subsidies lead to faster economic growth; imitative R&D subsidies actually lead to slower economic growth. A key assumption driving these conclusions is that R&D activities are subject to decreasing returns. When R&D activities are subject to constant returns, as is commonly assumed, the only equilibrium with both innovation and imitation is unstable.

This paper incorporates uncertainty in two distinct models of endogenous growth. In both models the representative agent is uncertain about the productivity of knowledge creation, as represented by a probability measure over the relevant parameter. The main purpose of this paper is to analyze the effects of risk or volatility in productivity of knowledge creation on the decision variables and the expected long‐run growth rate. Both the first and the second models may explain part of the observed negative link between volatility and growth.

We study how exploration versus exploitation innovations impact economic growth through a tractable endogenous growth framework that contains multiple innovation sizes, multi-product firms, and entry/exit. Firms invest in exploration R&D to acquire new product lines and exploitation R&D to improve their existing product lines. We model and show empirically that exploration R&D does not scale as strongly with firm size as exploitation R&D. The resulting framework conforms to many regularities regarding innovation and growth differences across the firm size distribution. We also incorporate patent citations into our theoretical framework. The framework generates a simple test using patent citations that indicates that entrants and small firms have relatively higher growth spillover effects.

This paper uses a case study of Hong Kong and Singapore, two of the fastest growing economies in the postwar world, to evaluate competing theories of economic growth. Although broadly similar in historical background and economic structure, the two economies are strikingly different along three dimensions of interest to growth theorists: (1) Hong Kong began the postwar era with a considerably better educated labor force; (2) the subsequent accumulation of human and physical capital in Singapore far exceeds that in Hong Kong; and (3) Singapore has experienced considerably more rapid structural change, as government targeting policies have propelled the economy into technologically advanced sectors. A total factor productivity growth analysis of the two economies reveals that while TFP growth in Hong Kong accounts for over two-thirds of the increase in GDP per capita during the 1970s and 1980s, total factor productivity growth in Singapore during the same period is next to nil. These results constitute strong evidence against linear models of growth that emphasize contemporaneous externalities in the accumulation of factors of production. The poor TFP performance of the Singaporean economy, when associated with its astounding rate of structural transformation, supports models that emphasize the constraints imposed by learning by doing on the evolution of comparative advantage.

We construct an endogenous growth model with random interactions where firms are subject to distortions. The TFP distribution evolves endogenously as firms seek to upgrade their technology over time either by innovating or by imitating other firms. We use the model to quantify the effects of misallocation on TFP growth in emerging economies. We structurally estimate the stationary state of the dynamic model targeting moments of the empirical distribution of R&D and TFP growth in China during the period 2007–2012. The estimated model fits the Chinese data well. We compare the estimates with those obtained using data for Taiwan and perform counterfactuals to study the effect of alternative policies. R&D misallocation has a large effect on TFP growth.

We provide a unifying framework to quantify the impact of several determinants of changes in US between-group inequality. We use an assignment framework with many labor groups, equipment types, and occupations in which changes in inequality are driven by changes in workforce composition, occupation demand, computerization, and labor productivity. We parameterize the model using direct measures of computer usage within labor group-occupation pairs and quantify the impact of each shock for various dimensions of between-group inequality between 1984 and 2003. We find, for example, that computerization and shifts in occupation demand jointly account for roughly 80 percent of the rise in the skill premium, with computerization alone accounting for roughly 60 percent. In an open-economy extension of the model, we show how computerization and changes in occupation demand can be caused by changes in the extent of international trade and perform counterfactual exercises to quantify these effects. (JEL D63, J16, J22, J23, J24, J31)

In this chapter we argue that the endogenous growth model with quality-improving innovations provides a framework for analyzing the determinants of long-run growth and convergence that is versatile, simple and empirically useful. Versatile, as the same framework can be used to analyze how growth interacts with development and cross-country convergence and divergence, how it interacts with industrial organization and in particular market structure, and how it interacts with organizations and institutional change. Simple, since all these aspects can be analyzed using the same elementary model. Empirically useful, as the framework generates a whole range of new microeconomic and macroeconomic predictions while it addresses empirical criticisms raised by other endogenous growth models in the literature.

We quantitatively characterize the optimal mix of progressive income taxes and education subsidies in a model with endogenous human capital formation, borrowing constraints, income risk and incomplete financial markets. In addition to the distortions of labor supply, progressive taxes weaken the incentives to acquire education. The latter distortion can potentially be mitigated by an education subsidy. We find that the welfare-maximizing fiscal policy is indeed characterized by a substantially progressive labor income tax code and a positive subsidy for college education. Both the degree of tax progressivity and the education subsidy are larger than in the current US status quo.

During World War II, the US government’s Office of Scientific Research and Development (OSRD) supported one of the largest public investments in applied R&D in US history. Using data on all OSRD-funded invention, we show this shock had a formative impact on the US innovation system, catalyzing technology clusters across the country, with accompanying increases in high-tech entrepreneur-ship and employment. These effects persist until at least the 1970s and appear to be driven by agglomerative forces and endogenous growth. In addition to creating technology clusters, wartime R&D permanently changed the trajectory of overall US innovation in the direction of OSRD-funded technologies. (JEL H56, N42, N72, O31, O33, O38, R11)

We assess the long-run growth effects of public policies to business R&D using data for US manufacturing industries and taking Schumpeterian growth theory as guideline. Our analysis indicates that R&D policy in the form of R&D tax credits fosters the rate of productivity growth over the long-term horizon. This effect is quantitatively important: increasing R&D tax credits by 10% raises the growth rate of labour productivity by 0.4% per year. We show that our findings are robust to controlling for several policy instruments, growth determinants and econometric issues. Moreover, the overall evidence is consistent with the predictions of second-generation fully-endogenous growth models.

An inventor's own knowledge is a key input in the innovation process. This knowledge can be built by interacting with and learning from others. This paper uses a new large-scale panel dataset on European inventors matched to their employers and patents. We document key empirical facts on inventors' productivity over the life cycle, inventors' research teams, and interactions with other inventors. Among others, most patents are the result of collaborative work. Interactions with better inventors are very strongly correlated with higher subsequent productivity. These facts motivate the main ingredients of our new innovation-led endogenous growth model, in which innovations are produced by heterogeneous research teams of inventors using inventor knowledge. The evolution of an inventor's knowledge is explained through the lens of a diffusion model in which inventors can learn in two ways: By interacting with others at an endogenously chosen rate; and from an external, age-dependent source that captures alternative learning channels, such as learning-by-doing. Thus, our knowledge diffusion model nests inside the innovation-based endogenous growth model. We estimate the model, which fits the data very closely, and use it to perform several policy exercises, such as quantifying the large importance of interactions for growth, studying the effects of reducing interaction costs (e.g., through IT or infrastructure), and comparing the learning and innovation processes of different countries.

How important is mastering information and communication technology (ICT) on modern labor markets? We answer this question with unique data on ICT skills tested in 19 countries. Our two instrumental-variable models exploit technologically induced variation in broadband Internet availability that gives rise to variation in ICT skills across countries and German municipalities. We find statistically and economically significant returns to ICT skills. For instance, an increase in ICT skills similar to the gap between an average-performing and a top-performing country raises earnings by about 8 percent. One mechanism driving positive returns is selection into occupations with high abstract task content.

We discuss a unified theory of directed technological change and technology adoption that can shed light on the causes of persistent productivity differences across countries. In our model, new technologies are designed in advanced countries and diffuse endogenously to less developed countries. Our framework is rich enough to highlight three broad reasons for productivity differences: inappropriate technologies, policy-induced barriers to technology adoption, and within-country misallocations across sectors due to policy distortions. We also discuss the effects of two aspects of globalization, trade in goods and migration, on the wealth of nations through their impact on the direction of technical progress. By doing so, we illustrate some of the equalizing and unequalizing forces of globalization.

The authors explore the educational response to fracking—a recent technological breakthrough in the oil and gas industry—by taking advantage of the timing of its diffusion and spatial variation in shale reserves. They show that fracking has significantly increased relative demand for less-educated male labor and increased high school dropout rates of male teens, both overall and relative to females. Estimates imply that, absent fracking, the teen male dropout rate would have been 1 percentage point lower over the period 2011–15 in the average labor market with shale reserves, implying an elasticity of school enrollment with respect to earnings below historical estimates. Fracking increased earnings and job opportunities more among young men than male teenagers, suggesting that educational decisions respond to improved earnings prospects, not just opportunity costs. Other explanations for the findings, such as changes in school quality, migration, or demographics, receive less empirical support.

Abstract Should governments direct research and development (R&D) away from “dirty” technologies towards “clean” ones? How important is this compared to carbon pricing? We address these questions with the introduction of two model features to the literature on directed technological change and the environment. We introduce decreasing returns to R&D, and allow future carbon taxes to influence current R&D decisions. Our results suggest that governments should prioritize clean R&D. Dealing with major environmental problems requires an R&D shift towards clean technology. However, in the case where most researchers are working with clean technology, both productivity spillovers and the risks of future replacement increase. Consequently, the gap between the private and social values of an innovation is greatest for clean technologies.

The introduction of Medicare in 1965 was the single largest change in health insurance coverage in U.S. history.Many economists and commentators have conjectured that the introduction of Medicare may have also been an important impetus for the development of new drugs that are now commonly used by the elderly and have substantially extended their life expectancy.In this paper, we investigate whether Medicare induced pharmaceutical innovations directed towards the elderly.Medicare could have played such a role only if two conditions were met.First, Medicare would have to increase drug spending by the elderly.Second, the pharmaceutical companies would have to respond to the change in market size for drugs caused by Medicare by changing the direction of their research.Our empirical work finds no evidence of a "first-stage" effect of Medicare on prescription drug expenditure by the elderly.Correspondingly, we also find no evidence of a shift in pharmaceutical innovation towards therapeutic categories most used by the elderly.On the whole, therefore, our evidence does not provide support for the hypothesis that Medicare had a major effect on the direction of pharmaceutical innovation.

Since 1991 more than two dozen states have adopted merit-based student financial aid programs, intended at least in part to increase the stock of human capital by improving the knowledge and skills of the state’s workforce. At the same time, there has been growing concern that the United States is producing too few college graduates in science, technology, engineering, and mathematics (STEM) fields. Using microdata from the American Community Survey, this paper examines whether recently adopted state merit aid programs have affected college major decisions, with a focus on STEM fields. We find consistent evidence that state merit programs did in fact reduce the likelihood that a young person in the state will earn a STEM degree.

We explore the educational response to fracking, a recent technological breakthrough in the oil and gas industry, taking advantage of the timing of its diffusion and spatial variation in shale reserves. We show that fracking has significantly increased relative demand for less-educated male labor and high school dropout rates of male teens, both overall and relative to females. Our estimates imply that, absent fracking, the teen male dropout rate would have been 1 percentage point lower over 2011-15 in the average labor market with shale reserves, implying an elasticity of school enrollment with respect to earnings below historical estimates. Fracking increased earnings more among young men than teenage boys, suggesting that educational decisions respond to improved earnings prospects, not just opportunity costs. Other explanations for our findings, like changes in school quality, migration, or demographics, receive less empirical support.

We analyze the growth and welfare effects of governmental basic research investments in an R&D-based growth model with endogenous fertility and endogenous education. In line with the empirical evidence, our model accounts for (i) the negative effect of population growth on economic growth, (ii) the positive effect of education on economic growth, (iii) the positive association between the level of per capita GDP and expenditures for basic research, and (iv) the gestation lag of basic research investments. Our results indicate that there exists an interior long-run welfare-maximizing investment rate in basic research that is much higher than the rates observed in OECD countries. The model-based explanation that we provide for this discrepancy is that raising public investments in basic research toward the optimal level reduces the growth rate of GDP and welfare in the short run because taxes have to increase and resources have to be drawn away from other productive sectors of the economy. These adverse short-run welfare effects are one potential explanation for the reluctance of governments and their currently living voters to increase public R&D expenditures despite the long-run benefits of such a policy.

This article examines career choices using a dynamic structural model that nests a job search model within a human capital model of occupational and educational choices. Wage growth occurs in the model because workers move between firms and occupations as they search for suitable job matches and because workers endogenously accumulate firm and occupation specific human capital. Simulations performed using the estimated model reveal that both self‐selection in occupational choices and mobility between firms account for a much larger share of total earnings and utility than the combined effects of firm and occupation specific human capital.

We document and discuss a dramatic change in the cyclical behavior of aggregate skilled hours since the mid‐1980s. Using CPS data for 1979:1–2003:4, we find that the volatility of skilled hours relative to the volatility of GDP has nearly tripled since 1984. In contrast, the cyclical properties of unskilled hours have remained essentially unchanged. We evaluate whether a simple supply/demand model for skilled and unskilled labor with capital‐skill complementarity in production can help explain this stylized fact. Our model accounts for about 60% of the observed increase in the relative volatility of skilled labor.

Economic theory suggests that market economies are likely to underprovide innovation because of the public good nature of knowledge. Empirical evidence from the United States and other advanced economies supports this idea. We summarize the pros and cons of different policy instruments for promoting innovation and provide a basic “toolkit” describing which policies are most effective according to our reading of the evidence. In the short run, R&D tax credits and direct public funding seem the most productive, but in the longer run, increasing the supply of human capital (for example, relaxing immigration rules or expanding university STEM admissions) is likely more effective.

Abstract This paper studies how the labor market frictions of skilled workers affect corporate valuation. The analysis features immigrant workers’ mobility constraints imposed by the U.S. green card application process and exploits exogenous variations caused by imperfections in the current immigration system. The study finds that relaxing mobility constraints negatively influences firm value. This effect is stronger for firms with higher labor adjustment costs. Reductions in investments and increases in labor costs are channels through which labor mobility adversely affects firm value. The findings suggest that monopoly rent over skilled workers is an important economic determinant of corporate valuation.

Do resources flow to patenting firms? Cross-country evidence from firm level dataThis paper exploits longitudinal data on firm performance and patenting activity for 23 OECD countries over the period 2003-2010 to explore the extent to which changes in the patent stock are associated with flows of capital and labour to patenting firms.While the finding that patenting is associated with real changes in economic activity at the firm level is in line with recent literature, new empirical evidence presented suggests that the impact of patenting on firm size is likely to be causal.Moreover, these data reveal important differences across OECD countries in the extent to which innovative firms can attract the complementary tangible resources that are required to implement and commercialise new ideas.In turn, the contribution of framework policies to explaining the observed cross-country differences in the magnitude of these flows is explored.While further research is required to establish causality, the results are consistent with the idea that well-functioning product, labour and capital markets; efficient judicial systems and bankruptcy laws that do not overly penalise failure can raise the returns to innovative activity.The paper also investigates the heterogeneous impacts of policies and finds that young firms -which are more likely to experiment with disruptive technologies and rely on external financing to implement and commercialise their ideas -disproportionately benefit from reforms to labour markets and more developed markets for credit and seed and early stage finance.

This article presents an endogenous growth model that is designed to be roughly consistent with the experience of high‐tech firms like Intel. In the model, industry leaders invest in R&D to improve their products, small firms invest in R&D to become industry leaders, and innovating becomes progressively more difficult over time. Consistent with the empirical evidence, the model implies that economic growth is independent of economy size and R&D intensity is independent of firm size. For plausible parameter values, it is optimal to heavily subsidize R&D activities.

In this paper we merge three datasets -individual income data, patenting data, and IQ data -to analyze the determinants of an individual's probability of inventing. We find that: (i) parental income matters even after controlling for other background variables and for IQ, yet the estimated impact of parental income is greatly diminished once parental education and the individual's IQ are controlled for; (ii) IQ has both a direct effect on the probability of inventing an indirect impact through education. The effect of IQ is larger for inventors than for medical doctors or lawyers. The impact of IQ is robust to controlling for unobserved family characteristics by focusing on potential inventors with brothers close in age. We also provide evidence on the importance of social family interactions, by looking at biological versus non-biological parents. Finally, we find a positive and significant interaction effect between IQ and father income, which suggests a misallocation of talents to innovation.

A model of endogenous growth, based on Schumpeter's notion of trustified capitalism, is developed and applied to firm-level data for the period 1973-1991. The model relates the market value of a firm to its current profits and to its R&D expenditures. The relationship depends upon the expected rate of knowledge growth, the expected value of an innovation and the elasticity of the R&D production function. Over the sample period, investors expected knowledge to grow at an average rate of 5 percent, a measure which reflects both process innovations and new product discoveries. Elasticities of the R&D production functions are estimated for thirteen industry groups and interpreted as measures of technological opportunity. There is no evidence of secular decline in technological opportunity over the sample period, but there is some evidence of diminishing returns to R&D intensity. Variations in technological opportunity over time are not correlated across industries. In contrast, the expected rates of knowledge growth at the industry level are highly correlated with the aggregate expected rate.

This paper develops a new approach to measuring human capital specificity, in the context of college majors, and estimates its labor market return over a worker's life cycle. To measure specificity, we propose a novel method grounded in human capital theory: a Gini coefficient of earnings premia for a major across occupations. Our measure captures the notion of skill transferability across jobs. Education and nursing are the most specific majors, while philosophy and psychology are among the most general. Using data from the American Community Survey, we find that the most specific majors typically pay off the most, with an early-career earnings premium of about 5–6% over average majors (15-20% over the most general majors), driven by higher hourly wages. General majors lag far behind at every age. Despite their earnings advantage, graduates from specific majors are the least likely to hold managerial positions, with graduates from majors of average specificity being the most likely to do so. It may be that managerial positions require a mix of specific knowledge and broadly applicable skills.

This article examines the effects of market structure on the variety of research projects undertaken and the amount of duplication of research. A characterization of the equilibrium market portfolio of R&D projects and the socially optimal portfolio is provided. It is shown that a merger decreases the variety of developed projects and decreases the amount of duplication of research. An increase in the intensity of competition among firms leads to an increase in the variety of developed projects and a decrease in the amount of duplication of research.

Prior work argues that the “market for ideas” supports an open system of innovation, allowing for efficient development of technology across firms. Although this literature has described important features of this market, how it influences the rate and direction of innovation remains an open question. We exploit an exogenous shock to a subset of U.S. medical device firms to study this question. We first document the breakdown in the market for ideas after a federal investigation made it more difficult for the leading orthopedic firms to work with physician‐inventors. We then present evidence of a dramatic decline in the rate of innovation for these firms. Further, a marked shift in direction occurs toward lower‐quality inventions and away from product categories where physician knowledge is critical . Copyright © 2014 John Wiley & Sons, Ltd.

This article is concerned with explaining the observed positive relationship in a cross section of industries between technical advance and R&D intensity. It presents a set of models to explore how technological opportunity and appropriability affect both of these variables. The models suggest that the key factors explaining the relationship are differences across industries in technological opportunity. Differences in appropriability make the relationship noisy.

Does regulation affect the pace and nature of innovation and if so, by how much? We build a tractable and quantifiable endogenous growth model with size-contingent regulations. We apply this to population administrative firm panel data from France, where many labor regulations apply to firms with 50 or more employees. Nonparametrically, we find that there is a sharp fall in the fraction of innovating firms just to the left of the regulatory threshold. Further, a dynamic analysis shows a sharp reduction in the firm's innovation response to exogenous demand shocks for firms just below the regulatory threshold. We then quantitatively fit the parameters of the model to the data, finding that innovation at the macro level is about 5.4% lower due to the regulation, a 2.2% consumption equivalent welfare loss. Four-fifths of this loss is due to lower innovation intensity per firm rather than just a misallocation towards smaller firms and lower entry. We generalize the theory to allow for changes in the direction of R&D, and find that regulation's negative effects only matter for incremental innovation (as measured by citations and text-based measures of novelty). A more regulated economy may have less innovation, but when firms do innovate they tend to "swing for the fence" with more radical (and labor saving) breakthroughs.

Between 1940 and 2000 there was a substantial increase in educational attainment in the United States. What caused this trend? We develop a model of human capital accumulation that features a nondegenerate distribution of educational attainment in the population. We use this framework to assess the quantitative contribution of technological progress and changes in life expectancy in explaining the evolution of educational attainment. The model implies an increase in average years of schooling of 24%, which is the increase observed in the data. We find that technological variables and in particular skill‐biased technical change represent the most important factors in accounting for the increase in educational attainment. The strong response of schooling to changes in income is informative about the potential role of educational policy and the impact of other trends affecting lifetime income.

We propose a simple model to assess the evolution of the US labor share and how automation affects employment. In our model, heterogeneous firms may choose a manual technology and hire a worker subject to matching frictions. Alternatively, they may choose an automated technology and produce using only machines (robots). Our model suggests that automation reduces the labor share but increases employment and wages. Furthermore, our model suggests that labor market institutions are unlikely to have played a major role in the fall of the US labor share after 1987. Instead, technological factors are a more promising candidate.

The transition from stagnation to growth and the associated phenomenon of the great divergence have been the subject of an intensive research in the growth literature in recent years. The discrepancy between the predictions of exogenous and endogenous growth models and the process of development over most of human history, induced growth theorists to advance an alternative theory that would capture in a single unified framework the contemporary era of sustained economic growth, the epoch of Malthusian stagnation that had characterized most of the process of development, and the fundamental driving forces of the recent transition between these distinct regimes.The advancement of unified growth theory was fueled by the conviction that the understanding of the contemporary growth process would be limited and distorted unless growth theory would be based on micro-foundations that would reflect the qualitative aspects of the growth process in its entirety. In particular, the hurdles faced by less developed economies in reaching a state of sustained economic growth would remain obscured unless the origin of the transition of the currently developed economies into a state of sustained economic growth would be identified, and its implications would be modified to account for the additional economic forces faced by less developed economies in an interdependent world.Unified growth theory suggests that the transition from stagnation to growth is an inevitable outcome of the process of development. The inherent Malthusian interaction between the level of technology and the size and the composition of the population accelerated the pace of technological progress, and ultimately raised the importance of human capital in the production process. The rise in the demand for human capital in the second phase of industrialization, and its impact on the formation of human capital as well as on the onset of the demographic transition, brought about significant technological advancements along with a reduction in fertility rates and population growth, enabling economies to convert a larger share of the fruits of factor accumulation and technological progress into growth of income per capita, and paving the way for the emergence of sustained economic growth.Variations in the timing of the transition from stagnation to growth and thus in economic performance across countries reflect initial differences in geographical factors and historical accidents and their manifestation in variations in institutional, social, cultural, and political factors. In particular, once a technologically driven demand for human capital emerged in the second phase of industrialization, the prevalence of human capital promoting institutions determined the extensiveness of human capital formation, the timing of the demographic transition, and the pace of the transition from stagnation to growth.

This paper develops a new empirical framework for analyzing occupational choice and career progression. I merge the NLSYs with O*Net and find that pre-market skills (primarily ASVAB test scores) predict the task content of the workers’ occupations. These measures account for 71 percent of the gender gap in science and engineering occupations. Career trajectories are similar across workers, so that initial differences in occupation persist over time. I then quantify the effect of layoffs on career trajectory and find that a layoff erases one-fourth of a worker’s total career increase in task content but this effect only lasts two years.

We examine how workers invest in human capital following unanticipated local labor market downturns. We find that, on average, two-year college enrollment increases by three students within three years for every one hundred workers laid off. This rise in enrollment accounts for half the observed increase in labor force nonparticipation following mass layoffs. Completions in career-technical programs also increase, especially in short-term certificates, but vary by field of study. We find the effect on completions is strongest in fields of study with larger earnings returns.

In this lecture we argue that important aspects of the growth process cannot easily be accounted for using models where capital accumulation is the main source of growth. The four aspects we emphasize in this lecture are: the transition trap, secular stagnation, the recent rise in top income inequality, and firm dynamics. The lecture argues that by contrast these aspects can be addressed by the Schumpeterian growth paradigm in which: (i) growth results primarily from innovation; (ii) innovation responds to incentives shaped by economic policies and institutions; (iii) new innovations replace old technologies (creative destruction).

This is an invited chapter for the forthcoming Volume 4 of the Handbook of Industrial Organization. We summarize the state of the literature on the economics of innovation and highlight open policy questions. We first articulate the key market failures in markets for innovation, and then discuss how both scientific norms and market-oriented policies help overcome those market failures. We close by discussing recent work on the diffusion of inventions as well as on the links between innovation and inequality.

We examine the quantitative impact of policy-induced changes in innovative investment by firms on growth in aggregate productivity and output in a model that nests several of the canonical models in the literature. We isolate two statistics, the impact elasticity of aggregate productivity growth with respect to an increase in aggregate innovative investment and the degree of intertemporal knowledge spillovers in research, that play a key role in shaping the model's predicted dynamic response of aggregate productivity, output, and welfare to a policy-induced change in the innovation intensity of the economy. Given estimates of these statistics, we find that there is only modest scope for increasing aggregate productivity and output over a 20-year horizon with uniform subsidies to firms' investments in innovation of a reasonable magnitude, but the welfare gains from such a subsidy may be substantial.

Abstract This paper examines the allocation of inventive effort in complex product systems. I argue that complex product systems, e.g., personal computers (PCs), are distinguished by functional interaction among several components, each guided by a relatively autonomous bundle of technical and economic characteristics. I try to explore whether the dynamics of such interactions between components of complex product systems can help us understand changes in the relative allocation of inventive effort. I advance and empirically test three hypotheses: (1) emergence of component constraints (bottlenecks) in product systems will trigger research and development (R&D) investment to resolve the constraints; (2) slack component firms have a strong incentive to invest in resolving component constraints; and (3) the incentive of slack component firms to invest in resolving component constraints is increasing in their prior sunk R&D investments in slack components. In sum, I argue that interactions between components in a product system conditions the R&D incentives of firms and also that the incentives are increasing in their prior investments or capabilities. Using product reviews from technical journals, I trace the constraint components in the PC from 1981 to 1998 and attempt to predict shifts in the allocation of inventive effort in the subsequent period. The empirical results strongly support all three hypotheses. This study highlights the paradoxical effect of modularity in complex product systems. Modular design architectures, while contributing to accelerating the pace of technical change, also tend to limit the economic benefits of firms' component R&D efforts, especially when different components technologies are progressing at different rates. This often creates an impetus to enlarge the scope of firm R&D activities beyond the component product markets that firms operate in. Other implications for R&D decision making are discussed. Copyright © 2007 John Wiley & Sons, Ltd.

POWER AND PROGRESS: Our Thousand-Year Struggle over Technology and Prosperity by Daron Acemoglu and Simon Johnson. PublicAffairs, 2024. 560 pages. Paperback; $21.99. ISBN: 9781541702547. *In this book, two highly acclaimed MIT economists, and Nobel prize winners, make the bold claim that technological progress does not automatically result in prosperity for all. This is contrary to the claims of what they call the "technology bandwagon," founded on the economic dogma arising from the rise in productivity and wages that occurred over the 20th century. Put simply, this dogma states that "when businesses become more productive they expand their output" which results in "a need for more workers" so they "get busy with hiring" and "collectively bid up wages" (p. 15). *To make its case, the book examines the relationship between technology, wages, and inequality over a thousand years with a view to determining what needs to be done to ensure that all parts of society share in the prosperity arising from innovation. From the opening chapter, it is clear that the authors are concerned about the current direction of digital technology, especially AI and its control by an elite few in Big Tech, what they term "a vision oligarchy" (p. 33) that needs to be "reigned in" (p. 34). Anyone interested in the ethics around technological development and its consequences on society, particularly recent developments in AI, will be interested in these perspectives. *Interpreting the economic and social data over a thousand years through to the present, the authors show how the economic prosperity of the post-World War II years was an outcome of a long struggle over the direction of technological progress and a balancing of power between employer and employee. Various examples are cited by the authors to justify their view that to create an economic elite involves a compelling vision and a social standing that affords opportunity to frame and set the agenda for debates on innovation, prosperity, human flourishing, and how to solve the world's big problems. The influence of the powerful becomes self-perpetuating if they have access to influence policy makers and if their ideas and arguments are persuasive and have broad appeal. *Many illuminating economic facts are employed throughout the book. Typical is that, apart from famine years or other disturbances such as war, food production remained roughly in line with population growth until the early 19th century, and that, despite the innovation of the middle ages, the quality of life of a European peasant changed little over several millennia. Productivity improvements benefited a very small elite of kings and their retinue, nobles, and the clergy. *Turning to the Industrial Revolution, the authors claim the poor did not share the wealth generated through technology innovation because of the bias in automation which favored those wealthy enough to purchase machinery and because of the lack of worker representation in setting wages. They also argue that the "aspirant" class in this period focused on accumulating wealth for themselves and did nothing to alleviate the appalling conditions in the first half of the 19th century. In making this claim, a glaring omission in the authors' analysis of the 18th and 19th century in Britain is the influence of evangelicals in the reform movement, such as the Clapham Sect, and businessmen, such as Cadbury, who conducted his business differently to most, providing homes for his workers and education for their children. This omission is surprising given that these evangelicals shaped institutions and public opinion in ways that the authors view as crucial to bringing about a change of vision in business leaders and institutions, as well as in the public. *The change in direction of technology in the second half of the 19th century plus and institutional changes up to the post-World War II period, ground the authors' conclusion that "the productivity bandwagon depends on new tasks and opportunities for workers and an institutional framework that enables them to share the productivity gains" (p. 218). A key 19th-century transition point was that the direction of technology shifted away from automation and people began to benefit more from the progress of technology. Key examples involve steam and electricity, which created new tasks and job opportunities in transport infrastructure and associated industries, such as steel and coal. Later, as electricity transformed factories by allowing distributed power rather than centralized steam power, there was a significant increase in the demand for engineers and white collar workers, pushing up wages. Contributing to this trend were institutional changes such as trade unions that gave greater bargaining power to workers, creating improved rent sharing between employers and employees. Political representation resulted in regulation with attendant improvements in conditions and public health. After World War II, there was a significant year-on-year increase in the "Total Factor Growth" measure of technological progress, and there was more inclusive economic growth with inequality declining rapidly as wages rose. *The closing chapters of the book focus on digital technology and AI, and detail how the 1,000-year struggle that finally resulted in a more inclusive prosperity began to unravel in the 1980s. Economic growth slowed and labor's share of national income has been on a protracted downward trend in most industrialized economies. The share of wealth in the richest 1% of the population increased from 10% in 1980 to 19% in 2019. Several factors that brought about these changes are reviewed, including the advent of the digital age and the automation of manual labor that it afforded, along with a change in economic doctrine, the erosion of union power, and deregulation that has favored cutting labor costs. All of this, it is argued, has led to a change of vision, often expressed as, "the social responsibility of business is to increase profits" and to generate "high returns for their shareholders" (p. 271), views now taught in most business schools. *The authors also argue that the "move fast and break things" mentality is symptomatic of a shift in the direction of digital technology and that the current AI vision of technology leaders is an illusion. This vision claims that AI will benefit humankind, yet in reality, it sidelines humans while generating huge wealth by reshaping our view of digital and AI technology away from creating new tasks and opportunities toward automating work and cutting labor costs, re-creating the old two-tier society of the previous millennia. Nevertheless, while some data is provided to justify this assertion of the authors in the use of robotics, there is much debate about the real impact of AI among white collar workers, a topic about which the authors offer no projections of their own. *Central to the book's thesis is the claim that a deterministic view of technology is a fallacy. Different choices could have been made in developing AI, away from automation and in directions more beneficial to society. However, what these directions might be are not really examined in any detail. A Christian redemptive approach to culture, while resonating with this nondeterministic view, would want to frame the argument in terms of responsible design choices involving stewardship, love for neighbor, and avoiding technological design that dumbs down humanity or leads to addiction or results in idolatry. *The final chapter outlines how Progressive movement activists, reformers, and journalists changed the views of the public, organized politically, and challenged institutions and government in America in the late 19th and early 20th century, leading to a redistribution of power and a change in direction for technological progress. A three-pronged formula is proposed as a way out of our current predicament: (1) "altering the narrative" and "changing the norms," (2) "cultivating countervailing powers," and (3) providing "policy solutions." How this would work is then sketched out using examples, such as how the environmental movement worked to redirect technologies. The authors' proposals for "Remaking Digital Technologies" were rather weak. Their suggestion that "improving productivity in workers' current jobs" (p. 394) is precisely what companies such as Microsoft would argue they are offering through their "co-pilot." I was also not convinced by the longer section on policy solutions that missed any reflection on proposed standards for responsible AI or policy proposals, such as the EU AI Act, details of which have been under discussion for the last few years. *In the complex world of social history and economics, it is often hard to prove a causal link between one factor and another, let alone when there are several variables in play. No doubt other economists and social historians will have a different take on the role of power and technological progress in shaping our world, and Christians will want to provide an interpretation through the lens of biblical truth. This book does, however, provide a helpful counterpoint to the prevailing AI vision that innovation is essential for growth and prosperity and that regulation stifles progress. *Reviewed by Jeremy Peckham, AI entrepreneur, ethicist, and former CEO, Bewdley, UK.

This paper compares changes in the structure of wages in France, Great Britain, Japan. and the United States over the last twenty years. Wage differentials by education and occupation (skill differentials) narrowed substantially in all four countries in the 1970s. Overall wage inequality and skill differentials expanded dramatically in Great Britain and the United States and moderately in Japan during the 1980s. In contrast, wage inequality did not increase much in France through the mid-1980s. Industrial and occupational shifts favored more-educated workers in all four countries throughout the last twenty years. Reductions in the rate of the growth of the relative supply of college-educated workers in the face of persistent increases in the relative demand for more-skilled labor can explain a substantial portion of the increase in educational wage differentials in the United States, Britain, and Japan in the 1980s. Sharp increases in the national minimum wage (the SM1C) and the ability of French unions to extend contracts even in the face of declining membership helped prevent wage differentials from expanding in France through the mid-1980s.

Abstract Research Summary In this paper, we suggest that staffing decisions in R&D alliances can reduce the inherent tension between value creation and value protection faced by participating firms. By considering R&D workers a primary source of knowledge leakage, we analyze the role of their intellectual property (IP) protection in shaping the misappropriation threat posed by the partner. We rely on patent ownership and inventorship data to analyze the selection of individuals for R&D collaborations in the pharmaceutical industry between 1991 and 2010. Our results suggest that an inventor's strength of IP protection is an important determinant in allocation decisions since it contributes to offsetting leakage risks in the alliance. The effect is especially strong in alliances that anticipate higher hazards. Managerial Summary The literature argues that firms can reap many benefits from R&D collaborations. However, such activities are challenging to manage because they require firms to put valuable knowledge at risk of misappropriation by the partner. We draw attention to the role that inventors play in generating and channeling knowledge during collaborative work and posit that the strength of the IP protection covering their innovations can safeguard against the consequences of knowledge leakage. We analyze pharmaceutical alliances and find that managers are more prone to allocate to collaborations inventors whose knowledge is better protected, particularly when the alliance anticipates considerable misappropriation risks. Our study has implications for how firms allocate inventors across projects, which may be an important factor for the overall success of firms' R&D strategies.

Abstract While the firm-level distributional consequences of market liberalization are well understood, previous studies have paid only limited attention to how variations in domestic institutions across countries affect the winners and losers from opening up to trade. We argue that the presence of coordinated wage-bargaining institutions, which impose a ceiling on wage increases, and state-subsidized vocational training, which creates a large supply of highly skilled workers, generate labor market frictions. Upward wage rigidity, in particular, helps smaller firms weather the rising competition and increasing labor costs triggered by trade liberalization. We test this hypothesis using a firm-level data set of European Union countries, which includes more than 800,000 manufacturing firms between 2003 and 2014. We find that, for productive firms, gains from trade are 20 percent larger in countries with liberal market economies than they are in coordinated market economies. Symmetrically, less productive firms in coordinated market economies experience significantly smaller revenue losses compared to liberal market economies. We show that both the presence of an institutionalized wage ceiling and the availability of subsidized vocational training are key mechanisms for reducing the reallocation of revenue from unproductive to productive firms in coordinated market economies compared to liberal market economies. In line with our theory, we find that wages and employment in liberalized industries increase differentially across both types of labor markets. Finally, we provide suggestive evidence that trade liberalization triggers a differential demand for redistribution at the individual level across different labor markets, which is in line with our firm-level analysis.

With quantity-based innovation targets and subsidy programs launched since the mid-2000s, China has seen a patent surge, accounting for 46% of the world's total patent applications in 2020; however, the overall patent quality has been declining after 2008. This paper develops a Schumpeterian growth model featuring innovating firms’ quantity–quality trade-off between radical and incremental innovations, and decomposes subsidies’ aggregate impact into quantity and quality channels. We calibrate the model to Chinese firm-level data in the early 2010s. Our quantitative analysis shows that the quality channel effects are negative and dominant, and quantity-based subsidies in that period reduce the TFP growth rate and welfare by 0.19 percentage points and 3.31%, respectively. We evaluate welfare gains under a constrained planner's problem, and propose skill subsidies which are quality-biased and effectively recover the optimal allocation.

In this paper, we estimate a rich model of college major choice using a panel of experimentallyderived data. Our estimation strategy combines two types of data: data on self-reported beliefs about future earnings from potential human capital decisions and survey-based measures of risk and time preferences. We show how to use these data to identify a general life-cycle model, allowing for rich patterns of heterogeneous beliefs and preferences. Our data allow us to separate perceptions about the degree of risk or perceptions about the current versus future payoffs for a choice from the individual's preference for risk and patience. Comparing our estimates of the general model to estimates of models which ignore heterogeneity in risk and time preferences, we find that these restricted models are likely to overstate the importance of earnings to major choice. Additionally, we show that while men are less risk averse and patient than women, gender differences in expectations about own-earnings, risk aversion, and patience cannot explain gender gaps in major choice.

In the 1980s and 1990s, the U.S. labor market experienced a remarkable polarization along with fast technological catch-up as Europe and Japan improved their global innovation performance. Is foreign technological convergence an important source of wage polarization? To answer this question, we build a multicountry Schumpeterian growth model with heterogeneous workers, endogenous skill formation, and occupational choice. We show that convergence produces polarization through business stealing and increasing competition in global innovation races. Quantitative analysis shows that these channels can be important sources of U.S. polarization. Moreover, the model delivers predictions on the U.S. wealth-income ratio consistent with empirical evidence.

This paper develops a dynamic model of innovation and international trade in which agents can direct their research efforts to specific goods in the economy. Trade affects the direction of innovation through its impact on the expected market size for an invention, leading to a two-way relationship between trade and technology absent in standard quantitative Ricardian models. Following a theory-consistent strategy to estimate the extent of endogenous adjustments in technology, I find that they can account for about half of the observed variance in comparative advantage in production in a sample of 29 countries and 18 manufacturing industries. In addition, the model suggests that standard Ricardian models overestimate the reductions in real income from increases in trade costs and underestimate the rise in real income due to trade liberalizations. (JEL F11, F14, L60, O31, O32)

In this paper, we propose an analytically tractable overlapping-generations model of human capital accumulation and study its implications for the evolution of the US wage distribution from 1970 to 2000. The key feature of the model, and the only source of heterogeneity, is that individuals differ in their ability to accumulate human capital. Therefore, wage inequality results only from differences in human capital accumulation. We examine the response of this model to skill-biased technical change (SBTC) theoretically. We show that in response to SBTC, the model generates behavior consistent with some prominent trends observed in the US data including (i) a rise in overall wage inequality both in the short run and long run, (ii) an initial fall in the education premium followed by a strong recovery, leading to a higher premium in the long run, (iii) the fact that most of this fall and rise takes place among younger workers, (iv) a rise in within-group inequality, (v) stagnation in median wage growth (and a slowdown in aggregate labor productivity), and (vi) a rise in consumption inequality that is much smaller than the rise in wage inequality. These results suggest that the heterogeneity in the ability to accumulate human capital is an important feature for understanding the effects of SBTC and interpreting the transformation of the US labor markets since the 1970s.

We study the effect of the competitive selection process on the incentive to innovate and the economy’s rate of growth by extending standard quality-ladder models of endogenous growth to allow for the possibility that in each period several asymmetric firms (i.e., an endogenously determined number of past innovators) may be simultaneously active in an industry. Stronger competitive\npressure has conflicting effects on the incentive to innovate, lowering prices but also selecting the more efficient firms. We show that the market selection effect of competition always increases the incentive to innovate and find circumstances in which it can outweigh the traditional negative Schumpeterian effect on growth

This paper introduces endogenous and directed technical change in a growth model with environmental constraints and limited resources. A unique final good is produced by combining inputs from two sectors. One of these sectors uses "dirty" machines and thus creates environmental degradation. Research can be directed to improving the technology of machines in either sector. We characterize dynamic tax policies that achieve sustainable growth or maximize intertemporal welfare, as a function of the degree of substitutability between clean and dirty inputs, environmental and resource stocks, and cross-country technological spillovers. We show that: (i) in the case where the inputs are sufficiently substitutable, sustainable long-run growth can be achieved with temporary taxation of dirty innovation and production; (ii) optimal policy involves both "carbon taxes" and research subsidies, so that excessive use of carbon taxes is avoided; (iii) delay in intervention is costly: the sooner and the stronger is the policy response, the shorter is the slow growth transition phase; (iv) the use of an exhaustible resource in dirty input production helps the switch to clean innovation under laissez-faire when the two inputs are substitutes. Under reasonable parameter values (corresponding to those used in existing models with exogenous technology) and with sufficient substitutability between inputs, it is optimal to redirect technical change towards clean technologies immediately and optimal environmental regulation need not reduce long-run growth. We also show that in a two-country extension, even though optimal environmental policy involves global policy coordination, when the two inputs are sufficiently substitutable environmental regulation only in the North may be sufficient to avoid a global disaster.

This article reviews the recent literature on the determinants of college major choices. We first highlight long-term trends and persistent differences in college major choices by gender, race, and family background. We then review the existing research in six key areas: expected earnings and ability sorting, learning, subjective expectations, non-pecuniary considerations, peer and family effects, and supply side factors. We examine and compare the various approaches employed by previous research and highlight key areas for future research.

We use a large and novel administrative dataset to investigate returns to different university ‘degrees’ (subject-institution combinations) in the United Kingdom. Conditioning on a rich set of background characteristics, we find substantial variation in returns across degrees with similar selectivity levels, suggesting students’ degree choices matter a lot for later-life earnings. Returns increase with university selectivity much more at the top of the selectivity distribution than further down, and much more for some subjects than others. Returns are poorly correlated with observable degree characteristics other than selectivity, which could have important implications for student choices and the incentives of universities.

Between 2000 and 2006, the Gini coefficient declined in 12 of the 17 Latin American countries for which data are available. Why has inequality declined? Have the changes in inequality been driven by market forces such as the demand and supply for labor with different skills? Or have governments become more redistributive than they used to be, and if so, why? This paper attempts to answer these questions by focusing on the determinants of inequality in four countries: Argentina, Brazil, Mexico and Peru. The analysis suggests that the decline in inequality is accounted for by two main factors: (i) a fall in the earnings gap between skilled and low-skilled workers (through both quantity and price effects); and (ii) more progressive government transfers (monetary and in-kind transfers). Demographic factors, such as a change in the proportion of adults (and working adults) per household, have been equalizing but the magnitude of their contribution has been small by comparison. In Brazil, Mexico and Peru, the fall in earnings gap, in turn, is mainly the result of the expansion of basic education over the last couple of decades, which reduced inequality in attainment and made the returns to education curve less steep. It also results from the petering out of the unequalizing effect of skill-biased technical change in the 1990s associated with the opening up of trade and investment. In Argentina, the decline in earnings inequality seems to be associated with government policies that without the windfall of high commodity prices will be hard to sustain.

Growth is closely related to structural transformation, the reallocation of economic activity among sectors. A well-functioning labor market plays an important role in this process by enabling workers to find employment in the growing, more productive sectors. We review the literature on labor market frictions that limit worker flows, slow structural transformation, and trap workers in poverty. The three main areas of focus are the extent of sectoral wage gaps, labor market dynamics, and evidence on specific frictions. Evidence in each area points to the presence of frictions that hinder worker reallocation. The literature also suggests policies that may help remediate frictions and improve worker mobility. We conclude by noting several open questions that provide promising avenues for future work.

Generalizing models of directed technical change, I show that complementarities between innovations and factors of production (here, energy resources) can drive transitions away from a dominant sector. In a calibrated numerical implementation, the economy gradually transitions energy supply from coal to gas and then to renewable energy, even in the absence of policy. The welfare-maximizing tax on carbon emissions is J-shaped, immediately redirects most research to renewables, and rapidly transitions energy supply directly to renewables. The emission tax is twice as valuable as either the welfare-maximizing research subsidy or the welfare-maximizing mandate to use renewable resources. (JEL H23, O31, O33, Q35, Q41, Q54)

How do college students choose their majors, and what role does the family play in their choices? I use data from two major longitudinal surveys to develop and estimate a model in which students learn about earning opportunities associated with different majors through the wages of older siblings and parents. The probability of a student choosing a major that corresponds to the occupation of a family member is strongly correlated with the family member's wage at the time the major choice is made. This correlation remains strong after controlling for family-correlated abilities or preferences, and additional empirical evidence suggests that the observed correlation arises through a family-based wage information channel.

We examine the quantitative impact of policy-induced changes in innovative investment by firms on growth in aggregate productivity and output in a model that nests several of the canonical models in the literature. We isolate two statistics, the impact elasticity of aggregate productivity growth with respect to an increase in aggregate innovative investment and the degree of intertemporal knowledge spillovers in research, that play a key role in shaping the model's predicted dynamic response of aggregate productivity, output, and welfare to a policy-induced change in the innovation intensity of the economy. Given estimates of these statistics, we find that there is only modest scope for increasing aggregate productivity and output over a 20-year horizon with uniform subsidies to firms' investments in innovation of a reasonable magnitude, but the welfare gains from such a subsidy may be substantial.

Abstract There is growing concern that the U.S. is producing too few college graduates in science, technology, engineering, and mathematics (STEM) fields, and there is a desire to understand how various policies affect college major decisions. This paper uses student administrative records from the University System of Georgia to examine whether and how Georgia’s HOPE Scholarship has affected students’ college major decisions, with a focus on STEM. We find that HOPE reduced the likelihood of earning a STEM degree. The research is complementary to a forthcoming paper by the authors, but using USG administrative records allows us to address several additional issues beyond the effect of merit aid on the likelihood of earning a STEM degree, including: the effect on initial major, earned major, and the transition between them; the roles of student ability, student performance, and institutional choice; and other possible mechanisms through which merit aid affects STEM education. JEL codes I23, J24

We study the origins of labor productivity growth and its differences across sectors. In our model, sectors employ workers of different occupations and various forms of capital, none of which are perfect substitutes, and technology evolves at the sector-factor cell level. Using the model we infer technologies from US data over 1960-2017. We find that sectoral differences in labor productivity growth are largely due to sectoral differences in the growth rate of routine labor augmenting technologies. Neither capital accumulation nor the occupational employment structure within sectors explains much of the sectoral differences in labor productivity growth.

During the last 15 years--especially in the 1980s--wage inequality rose in the United States. It appears that this can be explained by a secular shift in production functions favoring workers with intellectual rather than manual skills, together with slower growth in the supply of skilled labor than in the previous decade.

Income and wage inequality increased rapidly in a number of OECD economies. This report surveys the literature on the determinants of wage and income inequality and presents a framework for analyzing policy. The focus is on human capital policies, but other policies that could also reduce income inequality are considered. The report concludes that increased income inequality in OECD economies reflects greater wage inequality and higher skill premia and that the most likely cause of the rise in skill premia is technical change that has increased the demand for skills and education, though changes in labor market institutions, such as minimum wage laws and the importance of union bargaining, are also likely to have played some role. Although increasing the supply of skills may have some beneficial effects, the most useful policies to reduce inequality would be those that can close the gap of skills between the top and the bottom of the income distribution, such as policies to improve the quality of secondary schooling and to encourage on-the-job training. Disclaimer: The views expressed are those of the author(s) and do not necessarily reflect the

This chapter reviews the economic literature on subjective expectations in the domain of education with a focus on high income countries. It begins with highlighting the motivations that prompted systematic survey elicitation and statistical analysis of youth's expectations of the returns to schooling and with tracing key milestones in the development of this research program. It then proceeds to reviewing the relevant body of research by organizing the discussion around four topics: (i) the analysis of the perceived monetary returns, risks, and costs of schooling; (ii) the analysis of the perceived nonmonetary returns, risks, and costs of schooling; (iii) the analysis of schooling decisions; (iv) the analysis of expectation formation and learning. Possible avenues for future research are discussed in the conclusion.

This study aims to examine the differential effect of technological diversification on research and development (R&D) productivity based on the qualitative properties of technology portfolios (i.e., the direction of technological diversification). Using the U.S. patent database from 1980 to 2010, we divided overall technological diversification into related and unrelated technological diversification; furthermore, the two potential moderating factors of technology portfolio centrality and R&D consistency were tested across the different types of technological diversification. The notable findings are as follows: First, technology portfolio centrality has a positive moderating effect that is more pronounced as the degree of technological diversification increases. Second, R&D consistency has a positive and linear moderating effect. Third, the positive (negative) effect of technological diversification is more pronounced under related (unrelated) technological diversification. Consequently, firms can better utilize the R&D-productivity-enhancing effect of technological diversification by considering both the current degree of technological diversification and the properties of their technology portfolios.

Since the early-1990s the UK experienced an unprecedented increase in university graduates. The proportion of people with a university degree by age 30 more than doubled from 16% for born in 1965-69 to 33% for those born ten years later. At the same time the age profile of the graduate premium remained largely unchanged across cohorts. This paper first establishes the facts using a detailed analysis of micro-data on wage and employment patterns over the last two decades, benchmarked against the US economy. We then show that the stability of the age profile in the premium across different birth cohorts is unlikely to be explained by either composition changes or selection on unobservables. We also argue that it is inconsistent with skill-biased technical change affecting all advanced economies in the same way. We further rule out explanations based on factor price equalisation. Our resolution of the puzzle is a model in which increases in level of education induce firms to transit toward a decentralised technology in which decision-making is spread more widely through the workforce. We provide empirical support for this view.

The use of computer technology at work has increased dramatically over the past decades from about 20 per cent in the early 1980s to more than 70 per cent at the beginning of the new millennium. This increase in the adoption and use of new technology is likely to have changed the labour market in many dimensions.1 With respect to wages it has been found that computer users earn substantially higher wages than non-users, with wage premiums up to levels as high as 20 per cent. It is however not clear whether this observed premium is a reflection of the returns to (computer) skills, the result of unobserved heterogeneity between computer users and non-users, or whether there are other sources underlying these wage differentials.2 Computer technology is particularly used by the more highly educated workers, suggesting skill advantages play a crucial role in adjusting to and using new technologies. Hence, adoption of computer technology is easily connected to changes in the wage structure. On the other hand, looking at the present use of computer technology, it is hard to understand why more highly educated workers have an advantage in using for example a PC compared with less highly educated workers. Related to this observation is the

The U.S. college wage premium doubles over the life cycle, from 27 percent at age 25 to 60 percent at age 55. Using a panel survey of workers followed through age 60, I show that growth in the college wage premium is primarily explained by occupational sorting. Shortly after graduating, workers with college degrees shift into professional, nonroutine occupations with much greater returns to tenure. Nearly 90 percent of life cycle wage growth occurs within rather than between jobs. To understand these patterns, I develop a model of human capital investment where workers differ in learning ability and jobs vary in complexity. Faster learners complete more education and sort into complex jobs with greater returns to investment. College acts as a gateway to professional occupations, which offer more opportunity for wage growth through on-the-job learning.

The paper develops a two-country endogenous growth model to investigate possible causes for the existence and persistence of productivity growth differentials between nations, even though these countries show a common technology, constant returns to scale and perfect international capital mobility. Private consumption is derived from a three-period overlapping generations specification. The source of productivity (growth) differentials in our model is the existence of a non-traded capital good (`human capital') whose augmentation requires a non-traded current input (time spent by the young in education rather than leisure). We consider the influence on productivity growth differentials of private thrift, public debt, the taxation of capital and savings and of policy towards human capital formation.

This paper reviews and synthesizes the literature on the macroeconomic implications of human capital theory. I begin with a review of the canonical model of education and the wage structure pioneered by Tinbergen (1975) and developed more fully by Goldin and Katz (2007). I also review innovations such as the task framework developed by Acemoglu and Autor (2011). The canonical model does a surprisingly good job of predicting changes in the wage structure in the U.S. and other developed countries over the last half-century. Relative to the canonical model, the task framework adopts a more flexible view of technology and does a better job of fitting non-monotonic changes in the wage structure. Yet the task framework does not fully explain why educated workers have done so well since 1980, nor does it explain other recent facts such as flattening returns to cognitive skills and growing returns to non-cognitive, “higher-order” skills such as teamwork. To understand these recent trends, we must move beyond a single index view of human capital, toward richer, multi-dimensional frameworks. I conclude with a discussion of the nascent literature on the implications of multi-dimensional human capital for the wage structure, which raises more questions than it answers.

This study examines the extent to which changing the composition of college majors among working-age population may affect the supply of human capital or effective labor supply. We use the South Korean setting, in which the population is rapidly aging, but where, despite their high educational attainment, women and young adults are still weakly attached to the labor market. We find that engineering majors have an advantage in various outcomes such as likelihood of being in the labor force, being employed, obtaining long-term position, and earnings, while Humanities and Arts/Athletics majors show the worst outcomes. We then conduct a back-of-the-envelope calculation of the impact of the recently proposed policy change to increase the share of engineering majors by 10% starting in 2017. Our calculation suggests that the policy change may have a positive but small impact on labor market outcomes.

Abstract Empirical analysis of firm‐level investment in research and development (R&D) and its effect on innovation patterns and productivity has advanced as a result of innovation surveys in many countries. The weak link in the analysis of these surveys is the empirical model of firm R&D choice. In this paper we summarize how a dynamic, structural model of firm investment can be used to estimate firm demand for R&D with the data collected in innovation surveys. The estimates provide a natural measure of the expected benefit to the firm of investing in R&D. They also allow the researcher to simulate how the firm's R&D investment will respond to factors that shift cost or demand such as a policy change designed to subsidize R&D expenditures or provide financial support to firms with less favorable access to capital markets.

Taxation affects the allocation of talented individuals across industries by blunting material incentives and thus relatively magnifying the non-pecuniary benefits of pursuing a "calling". If higher-paying industries (e.g. finance and management) generate less positive net externalities than lower-paying professions (e.g. public service and education) this may enhance efficiency. We develop a theory of income taxation as implicit Pigouvian taxation of these externalities and calibrate it using data on the distribution of income and talent across industries. Even without any redistributive motive, tax rates are highly sensitive to the externalities assumed within a spectrum many would consider reasonable: they range from extremely regressive to highly progressive at high incomes. Our theory thus offers an alternative, pure efficiency rationale for non-linear income taxation, challenging the connection between high long-run labor supply elasticities and low optimal tax rates and motivating further study of the externalities generated by professions.

This paper presents new evidence on the practice of industrial Research and Development (R&D), especially the allocation between learning and internal research, and the role of outside knowledge, as represented by R&D spillovers, in reshaping this allocation. The evidence describes the sources of outside knowledge, portrays the flow of that knowledge into firms, and interprets the channels by which outside knowledge influences R&D. The empirical work is based on a sample of 220 R&D laboratories owned by 115 firms in the U.S. chemicals, machinery, electrical equipment, and motor vehicles industries. The findings are consistent with the view that universities and firms generate technological opportunities in R&D laboratories. In addition to partnerships that define rather strict channels of opportunity, the paper uncovers broader effects of R&D spillovers. The results also suggest that academic spillovers drive learning about universities, and that industrial spillovers drive learning about industry. In this way externally derived opportunities reshape the rate and direction of R&D. Overall the findings paint an image of practitioners of industrial R&D reaching aggressively for opportunities, rather than waiting for opportunities to come to them.

This article analyses how firms allocate their resources when they compete for multiple patents in heterogeneous research projects simultaneously. A simple model shows that firms’ resource allocation is biased away from risky and basic research, even when imitation is not possible and firms are fully rational. Therefore a market may lack major innovations despite large aggregate research expenditure and strong patent protection. This article also shows that as a market becomes more competitive, firms invest relatively less in basic research but more in risky research. These results provide a novel explanation for an ambiguous empirical relationship between innovation and market concentration.

We study how strategic considerations which pertain to the microeconomic process of innovation affect the macroeconomic process of growth and its efficiency. To a lesser extent, we also study how they may cause fluctuations to occur. We do this by the means of two models.;The first model pictures a one-good economy where long-run growth and output fluctuations are endogenous consequences of the decisions taken by entrepreneurs on the allocation of their resources between production and innovation in a Markovian sequence of one-period games. We show, first, that the log of output follows a process with random walk characteristics; second, that a recession is the consequence, not of a Kydland and Prescott (1982) negative shock on technology, but of the reallocation of factors in the face of an "increased opportunity" ex ante which the entrepreneurs fail to exploit ex post; third, and finally, that, although any generation could make itself unilaterally better off by reducing its level of research, the one-good economy is intertemporally efficient in that such a move would imply a reduction in the level of welfare of future generations.;The second model pictures an infinite-horizon multisector economy where endogenous growth in the aggregate results from the innovation races which go on in each and every sector. We study the coordination failures that may arise and uncover five externalities which have strategic implications. We show that, at least for some parametrisations of our model, there exist multiple stationary equilibria associated with different rates of growth. One noteworthy source of that multiplicity is the "real income effect", which relates the resources spent on research to the real value of nominal incomes through the lower real prices innovations eventually entail. Finally, we show that, because agents are infinitely-lived and many externalities occur, the multisector economy is intertemporally inefficient. Furthermore, we show that, depending upon the particular values taken by the parameters, the competitive process may lead to either overinvestment or underinvestment in research; i.e. the rate of growth may either be higher or lower than efficiency would warrant.

Abstract Research Summary Strategic human capital scholars suggest that firms' human capital rents are greater when labor market frictions are more prevalent. Taking this argument further, we suggest that when frictions for one type of human capital decrease, firms are motivated to place greater emphasis on human capital that is interdependent in production where frictions are unchanged. Empirically, we exploit an exogenous institutional change in the National Football League to demonstrate that coaching (managerial) dismissal and replacement is more likely to occur (and is influenced by a wider variety of information) after frictions for player (production worker) human capital decrease. Our findings suggest adding a new dimension—tradeoffs between related labor market segments—to the scholarly conversation about how firms manage their human capital (and other resources) strategically. Managerial Summary We often point to the existence of labor market frictions for different types of human capital as reasons why firms would emphasize one type of human capital over another. But changes in frictions in markets for related human capital, the present focus, can influence this decision as well. In particular, our study demonstrates that when the NFL implemented free agency and a salary cap in the market for player talent in 1993, limiting the ability of teams to stockpile talented players, NFL teams responded by increasing their emphasis on coaching talent in the production of wins. This increased emphasis manifested in more frequent coaching dismissals and a greater influence of different types of information on the desirability of replacing a team's head coach after 1993.

This paper offers a thesis as to why the US overtook the UK and other European countries in the 20th century in both aggregate and per-capita GDP, as a case study of recent models of endogenous growth where human capital is the engine of growth. The conjecture is that the ascendancy of the US as an economic superpower owes in large measure to its relatively faster human capital formation. Whether the thesis has legs to stand on is assessed through stylized facts indicating that the US led other OECD countries in schooling attainments per adult population over the 20 century, especially at the secondary and tertiary levels. While human capital is viewed as the direct facilitator of growth, the underlying factors driving the US ascendancy are linked to the superior returns the political-economic system in the US has so far offered individual human capital attainments, both home-produced and imported.

Recent research has been able to measure two forms of technical change---one (fossil) energy-saving and one saving on capital/labor. The results first show strong evidence for "directed technical change" in the sense that the total resources devoted to saving on the inputs responds endogenously to incentives and that that the two aggregate technology series display a negative medium-run correlation. Second, the elasticity of substitution between these inputs is close to zero (in contrast to the standard assumption of 1). Against this background we set up a simple and almost-tractable model of directed technical change where a final good is produced with capital and finite fossil energy. Specifically, the model features log utility, Leontief production, full depreciation and zero-cost fossil-fuel extraction. We calibrate the model and show that it can simultaneously capture important features of the U.S. post-war period. First, if the capital-augmenting technology starts below its steady state value, the model features peak oil and a relatively fast growth in the capital-augmenting technology and a low growth rate of the energy-saving technology as in the U.S. up to the first oil-price shock. Second, monopoly power (modeled as a reduction in the natural resource) causes an increase in the rate of energy-saving technical change and slow growth in the technology that saves on capital (i.e., a productivity slowdown) as in the period after the first oil shock.

What determines the comovements of aggregate employment and wages? This classic question in macroeconomics has received renewed attention since the Great Recession, when real wages did not fall despite a crash in employment. This paper proposes a microfoundation for the short-run dynamics of aggregate labor markets which relies on worker heterogeneity. I develop a model in which workers differ in their skills for various occupations, sectors employ occupations with different weights in production, and skills are imperfectly transferable. When shocks are concentrated in particular sectors, the extent to which workers can reallocate across the economy determines aggregate labor market dynamics. I apply the model to study the recession of 2008-09. I estimate the distribution of worker skills using two-period panel data prior to the recessions. Shocking the estimated model with sector-level TFP series replicates the increase in aggregate wages in 2008-09, and decline in 1990-91. The model implies that if either the composition of sector shocks or the distribution of skills in the economy had been the same in the 2008-09 recession as in the 1990-91 recession, real wages would have fallen, while employment would have declined less. This is because skills became less transferable between the 1980s and 2000s. In addition, the declining sectors during 2008-09 all employed a similar mix of skills, which induced many low-skill workers to leave the labor force and limited downward wage pressure on the rest of the economy. Finally, the model suggests a reduced form method to correct aggregate wages for selection in the human capital of workers, which accounts for cyclical job downgrading by focusing on the wage movements of occupation-stayers and recovers wage declines during the Great Recession.

August 1997 The United States has recently seen a dramatic rise in income inequality, all the more surprising because the long term trend had been toward equality. This paper examines one of the leading explanations; computerization in the workplace. I offer a theory of computers’ impact on white-collar work which goes far toward explaining the timing, form, and locus of recent labor market changes. The theory looks at the bureaucratic and organizational applications of computers that have been first, largest, and most influential. They have two effects on firms’ demand for labor at different skill levels. Computer decisionmaking has been a substitute for human decisionmaking over a limited range of tasks. Low- and middle-skill white collar work has been the most affected. Substitution of computers for high-skill workers has been quite limited. The rising demand for more highly skilled workers is driven by broad changes in the economics of the firm with many causes including computerization. While this theory is in agreement with many recent analyses pointing to computers, the specifics are quite different. Complementarities between computer use and individual workers’ skills are not an important component of change. This very different view of the mechanisms of skill biased technical change has new implications for understanding labor markets over the last 25 years, for the policy debate, and for predicting the future evolution of labor demand.

This chapter reviews recent advances in the task model and shows how this framework can be put to work to understand trends in the labor market in recent decades.Production in each industry requires the completion of various tasks that can be assigned to workers with different skills or to capital.Factors of production have well-defined comparative advantage across tasks, which governs substitution patterns.Technological change can: (1) augment a specific labor type-e.g., increase the productivity of labor in tasks it is already performing; (2) augment capital; (3) automate work by enabling capital to perform tasks previously allocated to labor; (4) create new tasks.The task model clarifies that these different technologies have distinct effects on labor demand, factor shares, and productivity and their full impact depends on the substitution patterns between workers that arise endogenously in the task framework.We explore the implications of the task framework using reduced-form evidence, highlighting the central role of automation and new tasks in recent labor market trends.We also explain how the general equilibrium effects ignored in these reduced-form approaches can be estimated structurally.

In light of skilled-labor shortage, the effect of a change in the wage of nurses on their labor supply is intensely discussed in recent literature. Using extensive data of German 14- to 15-year-olds, I analyze the role of the beliefs about a nurse's wage in the decision to become one. To estimate a partial effect, I select controls and their functional form using post-double-selection, which is a data-driven selection method based on regression shrinkage. Highlighting the importance of wages at the extensive margin of labor supply, the wage beliefs play a positive and statistically significant role. Although information is publicly available, educational choices knowingly suffer from misinformation. I find that especially those who do not become a nurse understate the wage. The results lead to two important policy implications. First, increasing the wage may help to overcome the shortage observed in many countries. Second, providing more information on the (relative) wage may be a successful strategy to attract more individuals into this profession. To assess the sensitivity of the results regarding omitted variable bias, I apply a novel approach. It turns out that potential unobserved confounders would have to be strong to overrule the conclusions.

Abstract How should the government respond to automation? We study this question in a heterogeneous agent model that takes worker displacement seriously. We recognize that displaced workers face two frictions in practice: reallocation is slow and borrowing is limited. We analyze a second best problem where the government can tax automation but lacks redistributive tools to fully alleviate borrowing frictions. The equilibrium is (constrained) inefficient and automation is excessive. Firms do not internalize that automation depresses the income of automated workers early on during the transition, precisely when they become borrowing constrained. The government finds it optimal to slow down automation on efficiency grounds, even when it does not value equity. Quantitatively, the optimal speed of automation is considerably lower than at the laissez-faire. The optimal policy improves efficiency and delivers meaningful welfare gains.

We estimate long-run interrelated demand functions for skilled labour, unskilled labour, clerical labour and capital services within dynamic framework using a panel of data on Irish manufacturing sectors during the 1980s. We group the sectors into three production ?types? ? high-growth sectors, medium-growth sectors and declining sectors. The results indicate very important differences in the demand for skilled labour compared to the demand for unskilled labour and in the underlying production technologies for the three groups of sectors. The medium-growth group of sectors are characterised by a stable production technology where skilled labour, unskilled labour and capital are all limited substitutes in production and there is little evidence of skill-biased technical change or trade effects. Most of the relatively minor shifts in factor shares in this group are accounted for by movements in relative factor prices. This group numbered over half of all manufacturing employment throughout the period under study. The high-growth group of sectors has all the features of the production technology described in modern growth theory: skilled labour and capital are complements in production, technical progress is biased against unskilled labour and the skill-intensity of production is increasing over time. This favours the 'skill-biased technical change' hypothesis. And the declining group of sectors are in secular decline with no stable long-run demand for labour. This would favour the ?trade effect? hypothesis where low-skill technologies are relocating form Ireland to low-large countries because of import penetration.

Abstract. Using the Canadian Workplace and Employee Survey and controlling for individual and establishment fixed‐effects, we find that within a year of adopting a computer, the average worker earns a 3.6% higher wage than a worker who did not use a computer. Returns are even larger for managers and professionals, highly educated workers, and those with significant prior computer experience. Employees who adopt computers for use with applications that require high cognitive skills earn the highest returns.

Most of the time, we assess an economy's performance using broad aggregate measures of output and wealth. In this regard, the United States is doing quite well. It is the richest country in the world. U.S. gross domestic product exceeded $11 trillion last year-roughly $38,000 per capita. And despite the slowdown associated with the 2001 recession, the economy has expanded at an average annual rate of more than 3 percent over the past 10 years. The way people actually feel about the economy's performance is shaped by their individual experiences, however, and here there is always great diversity. Indeed, there remains substantial anxiety about the direction the economy is heading, especially in regard to the growing disparity in income. The gap in real wage rates between those at the higher end of the distribution and those at the lower end has been widening for some time. In addition, the real wages of workers at the lowest part of the distribution were stagnant or falling during much of this extended period of growing wage inequality. This essay will explain why wage inequality has been increasing in the United States; in doing so, we will draw upon the scholarly literature, including work done by Richmond Fed economist Andreas Hornstein with Per Krusell of Princeton University and Giovanni Violante of New York University. We also will discuss the associated policy implications-that is, what can be done to better assure that all Americans have the opportunity to secure well-paying jobs, as well as which policies may hinder that goal. Overall, we will argue that technical innovation has significantly affected the wage distribution in the United States. But the direction of that effect has not been uniform. In the early part of the twentieth century, various technical innovations had the effect of compressing the wage structure. Since the 1970s, however, technical innovation-particularly the introduction and widespread use of information technology-has produced wage dispersion. Another force to which many have attributed recent labor market developments is globalization. We conclude that international trade and immigration, while significant trends, are not by themselves the primary force behind growing wage inequality. To some extent, globalization is itself a result of advances in information technology, which allow the production of goods and services to take place over a broader geographic area. As for public policy, research suggests that increased emphasis on education is a sound response to recent trends in wage inequality, particularly education early in life and programs focusing on general, broadly applicable skills. Early skill acquisition yields rewards over a relatively long period of time because individuals can recoup their investment in human capital throughout their working lives. In addition, such training tends to build on itself: acquiring skills early in life makes it easier to acquire additional skills later in life. In contrast, policies that would aim to slow the growth in wage inequality by imposing barriers to globalization, such as trade restrictions, would likely do little to achieve their intended goal, while lowering aggregate income and overall social welfare. Before discussing why wage inequality has been growing and the steps policymakers may wish to consider in response, it is necessary to look at the facts. In the next section, we present data on wage inequality from the early twentieth century to the present. 1. THE FACTS Most economists agree that wage inequality has been increasing in the United States recently.1 But this has not always been so. Wage inequality was large during the first part of the twentieth century, decreased during the middle part of the century, and accelerated again toward the end of the century. During the early part of the twentieth century, several factors contributed to a decline in the demand for less-skilled workers. …

Generative artificial intelligence (gen AI) holds immense potential to boost productivity growth and advance public service delivery, but it also raises profound concerns about massive labor disruptions and rising inequality. This note discusses how fiscal policies can be employed to steer the technology and its deployment in ways that serve humanity best while cushioning the negative labor market and distributional effects to broaden the gains. Given the vast uncertainty about the nature, impact, and speed of developments in gen AI, governments should take an agile approach that prepares them for both business as usual and highly disruptive scenarios.

This paper studies the complementarity between domestic and foreign skilled workers. It develops a simple model where employers seek to recruit a foreign worker when finding domestic workers takes more time. This paper confirms the predictions of the model. I rely on a within-firm within-occupation identification strategy to compare recruitment decisions made by a given employer for similar positions that differ in job posting duration. To identify this relationship, I have collected and assembled a new and original dataset at the job level. I match online job postings to administrative data on labor condition applications (LCAs) submitted as the first step in applying for H-1B temporary skilled worker visas. I find that employers are 28 percent more likely to submit an LCA when the job posting duration is one standard deviation longer. I provide evidence suggesting that this phenomenon is due to insufficient domestic labor supply in these occupations.

The positive correlations found between computer use and human capital are often interpreted as evidence that the adoption of computers have raised the relative demand for skilled labor, the widely touted skill-biased technological change hypothesis. However, several models argue the skill-intensity of technology is endogenously determined by the relative supply of skilled labor. We use instruments for the supply of human capital coupled with a rich dataset on computer usage by businesses to show that the supply of human capital is an important determinant of the adoption of personal computers. Our results suggest that great caution must be exercised in placing economic interpretations on the correlations often found between technology and human capital.

This paper surveys the experience of economic growth in the 20th century with a focus on technological change at the frontier together with issues related to success and failure in catch-up growth. A detailed account of growth performance based on historical national accounts data is given and is accompanied by a review of growth accounting evidence on the sources of economic growth. The key features of our analysis of divergence in growth outcomes are an emphasis on the importance of “directed” technical change, of institutional quality, and of geography. We provide brief case studies of the experience of individual countries to illustrate these points.

We use a controlled experiment to show that ability and belief calibration jointly determine the benefits of working with Artificial Intelligence (AI). AI improves performance more for people with low baseline ability. However, holding ability constant, AI assistance is more valuable for people who are calibrated, meaning they have accurate beliefs about their own ability. People who know they have low ability gain the most from working with AI. In a counterfactual analysis, we show that eliminating miscalibration would cause AI to reduce performance inequality nearly twice as much as it already does.

To understand students’ motives in obtaining postgraduate qualifications, we elicit intentions to pursue postgraduate education and beliefs about its returns in a sample of 1002 university students. We find large gaps in perceptions about the immediate and later-life benefits of postgraduate education, both between first- and continuing-generation students and within the latter group. Differences in student beliefs about returns can account for 70% of the socioeconomic gaps in intentions to pursue postgraduate studies. We document large differences in students’ current undergraduate experiences by socioeconomic background and find these to be predictive of perceived returns to postgraduate education.

This study examines the impact of university shareholding on corporate innovation. We find that university shareholding significantly promotes corporate innovation performance. University–firm connections improve corporate innovation through attracting more highly educated talent and promoting university–firm cooperation. Additional tests suggest that the positive effect is more pronounced for firms in high-tech industries or those facing more-intense labor market competition. Moreover, the combined effects of university shareholding and corporate innovation significantly enhance firm value. Our findings provide insights for understanding the role of universities in firms' innovation practices in emerging markets. • University shareholding significantly promotes corporate innovation performance. • University–firm connections improve corporate innovation by attracting highly educated talent and promoting cooperation. • The effect is more pronounced for firms in high-tech industries or those facing more-intense labor market competition.

The positive correlations found between computer use and human capital are often interpreted as evidence that the adoption of computers has raised the relative demand for skilled labor, the widely touted skill-biased technological change hypothesis.However, several models argue that the skill intensity of technology is endogenously determined by the relative supply of skilled labor.We use instruments for the supply of human capital coupled with a rich data set on computer usage by businesses to show that the supply of human capital is an important determinant of the adoption of personal computers.Our results suggest that great caution must be exercised in placing economic interpretations on the correlations often found between technology and human capital.

This paper studies a non-degenerate price distribution for the homogeneous good within a model of endogenous directed technical change. A probabil-ity density function is analytically derived and shown to be related to the technology and innovation parameters of the model.

Abstract In the 1970s and 1980s the US position as the global technological leader was increasingly challenged by J apan and E urope. In those years the US skill premium and residual wage inequality increased substantially. This paper presents a two‐region, quality‐ladder growth model where the lagging economy progressively catches up with the leader. As the innovation gap closes, the advanced country experiences fiercer foreign technological competition that forces its firms to innovate more. Faster technical change increases the skill premium and residual inequality. Offshoring production and innovation plays a key role in shaping the link between international competition and inequality.

The divergence of unemployment rates between the U.S. and Europe coincided with a substantial acceleration in capital-embodied technical change in the late 70’s. Furthermore, evidence suggests that European economies have been lagging behind the U.S. in the adoption and usage of new technologies. This paper argues that the pace of technology adoption plays a fundamental role for how an economy’s labor market reacts to an acceleration in capital-embodied growth. The framework proposed offers a novel explanation for the divergence of unemployment rates across economies that are hit by the very same shock (i.e. the acceleration in embodied technical change) but differ in their technology adoption. Moreover, the results of the paper challenge the popular- but controversial- view that high European unemployment is the result of institutional rigidities by claiming that institutions are not the principal cause per se but they rather amplify certain forces that promote the emergence of high unemployment.

What quantitative lessons can we learn from models of endogenous technical change through innovative investments by firms for the impact of changes in the economic environment on the dynamics of aggregate productivity in the short, medium, and long run? We present a unifying model that nests a number of canonical models in the literature and characterize their positive implications for the transitional dynamics of aggregate productivity and their welfare implications in terms of two sufficient statistics. We review the current state of measurement of these two sufficient statistics and discuss the range of positive and normative quantitative implications of our model for a wide array of counterfactual experiments, including the link between a decline in the entry rate of new firms and a slowdown in the growth of aggregate productivity given that measurement. We conclude with a summary of the lessons learned from our analysis to help direct future research aimed at building models of endogenous productivity growth useful for quantitative analysis.

How does personal exposure to economic conditions affect individual human capital investment choices? Focusing on bachelor’s degree recipients, we find that cohorts exposed to higher unemployment rates during typical schooling years select majors that earn higher wages, have better employment prospects, and lead to work in a related field. Conditional on expected earnings, recessions also encourage women to enter male-dominated fields, and students of both genders pursue more difficult majors. We conclude that economic environments change how students select majors, and we find evidence that students who respond to the business cycle enjoy earnings typical of their new majors.

Using matched worker/plant level data for Norway, trade and technology explanations for the change in skill composition are assessed using direct evidence on the job creation and destruction for workers with high, medium and low education level. In order to disentangle the supply and demand effects, we fix the skill level by analysing the job creation process for skills within cohorts of workers. The econometric analysis support skill-biased technical change via positive gross and net job creation effects in high-tech sectors. A opposed to most studies we also find a negative impact on job displacement for all types of workers is found in sectors exposed to strong import competition. Skill-biased technical change is also supported by the descriptive statistics in that most net employment changes takes place within both the manufacturing sector and a service sector, and most job turnover takes place between plants.

This paper innovatively defines the structure of human capital based on existing theories and use the heterogeneous stochastic frontier model to test the relationship between various types of human capital and innovation. It is found that with the progress of technology, the structure of human capital and the mechanism of promoting innovation have undergone profound changes. Together with commercial leasing human capital, financial human capital, and entrepreneurs, information human capital and transportation human capital participate in innovation activities and create a new collaborative space, which in turn helps to spread the value of innovation and influence the effectiveness of the entire innovation system. This study enriches the theoretical research on the mechanism of human capital promoting innovation in the information age and provides new insight to explain the imbalance between human capital input and innovation output. Based on the theory of innovation chain value and flow space, this study redefines the connotation of innovative human capital structure by distinguishing the nature of innovation activities. This study also reveals the difference in the influence of human capital structure on different types of innovation activities. This study partly explains the problem of increasing human capital input but decreasing innovation efficiency in developing countries. Therefore, from a macro perspective, our research provides guidance for understanding the transformation of innovation under the background of technological progress and formulating effective urban innovation strategies for managers.

We construct an endogenous growth model with random interactions where firms are subject to distortions. The TFP distribution evolves endogenously as firms seek to upgrade their technology over time either by innovating or by imitating other firms. We use the model to quantify the effects of misallocation on TFP growth in emerging economies. We structurally estimate the stationary state of the dynamic model targeting moments of the empirical distribution of R&D and TFP growth in China during the period 2007-12. The estimated model fits the Chinese data well. We compare the estimates with those obtained using data for Taiwan and perform counterfactuals to study the effect of alternative policies. R&D misallocation has a large effect on TFP growth.

Abstract The majority of innovations are developed by multi-sector firms. The knowledge needed to invent new products is more easily adapted from some sectors than from others. We study this network of knowledge linkages between sectors and its impact on firm innovation and aggregate growth. We first document a set of sectoral-level and firm-level observations on knowledge applicability and firms’ multi-sector patenting behaviour. We then develop a general equilibrium model of firm innovation in which inter-sectoral knowledge linkages determine the set of sectors a firm chooses to innovate in and how much R&D to invest in each sector. It captures how firms evolve in the technology space, accounts for cross-sector differences in R&D intensity, and describes an aggregate model of technological change. The model matches new observations as demonstrated by simulation. It also yields new insights regarding the mechanism through which sectoral fixed costs of R&D affect growth.

What quantitative lessons can we learn from models of endogenous technical change through innovative investments by firms for the impact of changes in the economic environment on the dynamics of aggregate productivity in the short, medium, and long run? We present a unifying model that nests several canonical models in the literature and characterize both their positive implications for the transitional dynamics of aggregate productivity and their welfare implications in terms of two sufficient statistics. We review the current state of measurement of these two sufficient statistics and discuss the range of positive and normative quantitative implications of our model for a wide array of counterfactual experiments, including the link between a decline in the entry rate of new firms and a slowdown in the growth of aggregate productivity. We conclude with a summary of the lessons learned from our analysis to help direct future research aimed at building models of endogenous productivity growth that are useful for quantitative analysis.

We document for a broad panel of advanced economies that increases in GDP per capita are associated with a shift in the composition of value added to sectors that are intensive in high-skill labor. It follows that further development in these economies leads to an increase in the relative demand for skilled labor. We develop a two-sector model of this process and use it to assess the contribution of this process of skill-biased structural change to the rise of the skill premium in the US over the period 1977 to 2005. We find that these compositional demands account for roughly 30% of the overall increase of the skill premium due to technical change.

Innovation is a key driver of long run economic growth.Studying innovation requires a clear view of the characteristics and behavior of the individuals that create new ideas.A general lack of rich, large-scale data has constrained such analyses.We address this by introducing a new dataset linking patent inventors to survey, census, and administrative microdata at the U.S. Census Bureau.We use this data to provide a first look at the demographic characteristics, employer characteristics, earnings, and employment dynamics of inventors.These linkages, which will be available to researchers with approved access, dramatically increases the scope of what can be learned about inventors and innovative activity.

In this paper we present an analytically tractable overlapping generations model of human capital accumulation, and study its implications for the evolution of the U.S. wage distribution from 1970 to 2000. The key feature of the model, and the only source of heterogeneity, is that individuals differ in their ability to accumulate human capital. Therefore, wage inequality results only from differences in human capital accumulation. We examine the response of this model to skill-biased technical change (SBTC) theoretically. We show that in response to SBTC, the model generates behavior consistent with several features of the U.S. data including (i) a rise in overall wage inequality both in the short run and long run, (ii) an initial fall in the education premium followed by a strong recovery, leading to a higher premium in the long run, (iii) the fact that most of this fall and rise takes place among younger workers, (iv) a rise in within-group inequality, (v) stagnation in median wage growth (and a slowdown in aggregate labor productivity), and (vi) a rise in consumption inequality that is much smaller than the rise in wage inequality. These results suggest that the heterogeneity in the ability to accumulate human capital is an important feature for understanding the effects of SBTC, and interpreting the transformation that the U.S.

This paper explores how talent flow network and the firm life cycle affect the innovative performances of firms. We first established an interorganizational talent flow network with the occupational mobility data available from the public resumes on LinkedIn China. Thereafter, this information was combined with the financial data of China's listed companies to develop a unique dataset for the time period between 2000 and 2015. The empirical results indicate the following: (1) The breadth and depth of firms' embedding in the talent flow network positively impact their innovative performances; (2) Younger firms' innovations are mostly promoted by the breadth of network embedding, but this positive effect weakens as firms increase in age; (3) Mature firms' innovations are primarily driven by the depth of network embedding, and this positive effect strengthens as firms increase in age. This paper enriches and deepens the studies of talent flow networks, and it provides practical implications for innovation management based on talent flow for various types of firms at different development stages.

SUMMARY This paper examines recent proposals and research suggesting that artificial intelligence (AI) adoption should be delayed until its potential harms are fully understood. Conclusions on the social optimality of delayed AI adoption are shown to be sensitive to assumptions about the process by which regulators learn about the salience of particular harms. When such learning is by doing – based on the real-world adoption of AI – this generally favours acceleration of AI adoption to surface and react to potential harms more quickly. This case is strengthened when AI adoption is potentially reversible. This paper examines how different conclusions regarding the optimality of accelerated or delayed AI adoption influence and are influenced by other policies that may moderate AI harm.

Firms frequently provide general skill training for workers. Theories propose that labor market frictions entail wage compression, generate larger productivity gains than wage growth to skill acquisition, and motivate a firm to offer general skill training, but few studies directly test them. We use unusually rich data from a temporary help service firm that records both workers’ wages and their productivity as measured by the fees charged to client firms. We find that skill acquired through training and learning by doing increases productivity more than wages, with such wage compression accounting for half of the average 40% productivity growth over 5 years of tenure.

Abstract I model and quantify the impact of a new dimension of globalization: offshore R&D. In the model, firms employ researchers across the globe to develop new product blueprints and then engage in offshore production and exporting. Frictions impeding trade and the separation of production from R&D lead to a “market-access” motive for offshore R&D, while cross-country differences in the distributions of firm knowhow and worker ability generate a “talent-acquisition” motive. I discipline the model using empirical facts derived from a new firm-level dataset. Counterfactual experiments show that the two motives can account for a significant portion of the observed offshore R&D. Incorporating offshore R&D amplifies the gains from globalization by a factor of 1.3 and generates new implications for the impacts of traditional forms of global integration, namely trade and multinational production.

We explore the determinants of research specialization and examine the e¤ects of technology di¤usion and trade on the incentives for innovation. To some extent, di¤usion substitutes for trade. If trade costs are low relative to di¤erences in comparative advantage in doing research, rapid di¤usion implies specialization in research activity. Country size has an ambiguous e¤ect on specialization in research. 1

How does firm heterogeneity affect the long-run consequences of financial crises? To answer this question, I introduce aggregate shocks and differences in the innovative potential of firms into a Schumpeterian endogenous growth model. Firm heterogeneity amplifies the effects of a financial crisis on aggregate innovation, because small firms are both relatively more innovative than large firms and hit harder by the crisis. A calibration using manufacturing data from Spain shows that a representative-firm model which ignores these mechanisms underestimates the long-run output losses due to the 2008-2013 crisis by around 40%.

This paper introduces multi-quality firms within a Schumpeterian framework. Featuring non-homothetic preferences and income disparities in an otherwise standard quality-ladder model, it shows that the resulting differences in the willingness to pay for quality among consumers generate both positive investments in R&D by industry leaders and positive market shares for more than one quality, hence allowing for the emergence of multi-product firms within a vertical innovation framework. This positive investment in R&D by incumbents is obtained with complete equal treatment in the R&D field between the incumbent patent holder and the challengers: in this framework, the incentive for a leader to invest in R&D stems from a “surplus appropriation effect” specific to vertically-differentiated markets, i.e. the perspective of more efficient price discrimination when expanding the product portfolio. Such a framework makes it possible to analyse the impact of income distribution, as well as that of several possible R&D policies, both on long-term growth and on the allocation of R&D activities between challengers and incumbents.

The United States has become reliant on workers from abroad to meet its demand for the knowledge-based economy. However, some migrants may face an earnings deficit relative to similar US-born workers. This paper examines the sources of the deficit and asks whether we should expect the initial deficit to disappear with education attainment and work experience in the US. They are challenging to answer as few data sources measure and track market experiences and educational trajectories of migrants over time. Migration and educational trajectories which reflect the country's source of formal educational credentials as well as other forms of capital may explain the deficit, this study applies sequence analysis to the National Science Foundation's 'National College Graduate Survey' to examine earnings differences between China- and US-born STEM workers. After identifying the dominant migration-education profiles for these STEM workers, I show that the wages of migrants with exclusively China-based education are 5–25% lower than those of workers with at least some US-based education, even among workers who are otherwise similar in terms of experience, legal status, employer type, occupation, degree level and time since migration. These findings point to significant and lasting penalties due to non-US education.

This paper introduces technological progress into an efficiency wage model, and argues that changes in the rate of technical change affect not only the demand for but also the effective supply of labour. This creates a new mechanism through which technological progress impacts on the wage of skilled workers relative to that of the unskilled. Previous work has argued that an increase in the relative wage would only come about if there were an acceleration in the rate of skill-biased technological change. In contrast, we find that technical change affects the skill premium even when it is ‘neutral’. Moreover, the paper shows that slower technical change may also increase the relative wage, allowing us to reconcile the change in the skill premium with the productivity slowdown experienced by OECD countries. The main problem of demand-based explanations of the increase in the skill premium is that they cannot account for the simultaneous increase in the unemployment rates for both skilled and unskilled workers. Our framework emphasises the joint determination of wages and employment, and generates wage and employment patterns that are consistent with the evidence.

Economic growth requires factor reallocation across firms and continuous replacement of technologies. Labor market institutions influence economic dynamism by their impact on the supply of a key factor, skilled workers to new and expanding firms, and the shedding of workers from declining and failing firms. Growth-favoring labor market institutions include portable pension plans and other job tenure rights, health insurance untied to the current employer, individualized wage-setting, and public income insurance systems that encourage mobility and risk-taking in the labor market.

The rate of change in the share of skilled labor has increased steadily over the past 35 years in Swedish manufacturing. A closer inspection of the period after 1970 indicates that while relative supply changes of skilled labor seem to have been the main driving force behind the growing skill shares in manufacturing industries over the period 1970-85, an acceleration in the relative demand for skills appears to have propelled higher skill shares during the late 1980s and in the beginning of the 1990s. Consistent with such a development is the finding of an increasing degree of complementarity between knowledge capital and skilled labor and that Swedish manufacturing firms, in recent years, have invested heavily in R&D. There is also some support for the belief that intensified competition from the South has increased the relative demand for skilled labor. However, the impact appears to be small and concentrated to the 1970s and the beginning of the 1980s.

We analyze a multi-sector growth model with directed technical change where man-made capital and exhaustible resources are essen- tial for production. The relative profitability of factor-specific inno- vations endogenously determines whether technical progress will be capital- or resource-augmenting. We show that convergence to bal- anced growth implies zero capital-augmenting innovations: in the long run, the economy exhibits purely resource-augmenting technical change. This result provides sound microfoundations for the broad class of models of exogenous/endogenous growth where resource-aug- menting progress is required to sustain consumption in the long run, contradicting the view that these models are conceptually biased in favor of sustainability.

We examine whether adverse selection has unraveled private markets for equity and state-contingent debt contracts for financing higher education. Using survey data on beliefs, we show a typical college-goer would have to repay $1.64 in present value for every $1 of financing to overcome adverse selection in an equity market. We find that risk-averse college-goers are not willing to accept these terms, so markets unravel. We discuss why moral hazard, biased beliefs, and outside credit options are less likely to explain the absence of these markets. We quantify the welfare gains for subsidizing equity-like contracts that mitigate college-going risks. (JEL D82, D83, G51, I22, I23, I26, J24)

Abstract With growing demand for workers in science, technology, engineering, mathematics (STEM) and health care, it is important to assess not only whether education interventions impact educational attainment, but also students’ majors. This study examines the impact of Early College High Schools (ECHSs) on bachelor's degree attainment by field of study using data on four hundred thousand students from North Carolina (7,300 in an ECHS). Using propensity score weighting, I find that ECHSs increase bachelor's degree attainment within ten years of high school entry by 4.7 percentage points (19 percent over baseline), with STEM degree attainment increasing by 1.3 to 2.4 points (18 to 34 percent). However, within STEM and STEM-related fields, ECHSs increase degrees in the natural sciences (1.3 points or 45 percent), math/computer science (0.6 points or 60 percent), and psychology (1.2 points or 57 percent), but have null and directionally negative effects on engineering (−0.1 points or −7 percent) and health care (−0.3 points or −17 percent). Patterns are generally similar across student subgroups, though male students drive increases in computer science/mathematics, whereas female and White students drive decreases in health care. Thus, ECHSs increase STEM degree attainment overall, but more research is needed to examine whether intensive dual-enrollment experiences like the ECHS may create barriers or disincentives to pursuing certain STEM fields.

Abstract Motivated by discussions of skill mismatches on local German vocational educational and training (VET) markets, this paper analyses how occupational segments of VET entry of individuals with lower and intermediate secondary school degree relate to local labor market characteristics. The econometric analysis applies data from a survey conducted with 9th graders within the German National Educational Panel Study (NEPS). Considering opportunity structures and the local competition for training positions, we find that the match between occupations' skill demands and individuals' abilities tends to be specifically close in diverse and competitive urban labor markets. In non-competitive peripheral labor markets, in contrast, graduates with lower school certificates seem to have a higher likelihood of entering VET in segments that are specifically attractive for graduates with upper secondary school degree. The results on the allocation of abilities and the weight of preferences under different labor market conditions have different welfare implications from an individual, regional and general economic perspective.

This paper studies the innovation dynamics of an oligopolistic industry. The firms compete not only in the output market but also by engaging in productivity enhancing innovations to reduce labor costs. Rent sharing may generate productivity dependent wage differentials. Productivity growth creates intertemporal spillover effects, which affect the incentives for innovation at subsequent dates. Over time the industry equilibrium approaches a steady state. The paper characterizes the evolution of the industry's innovation behavior and its market structure on the adjustment path. © 2012 Elsevier B.V.

High-tech firms hold substantial cash reserves. We build a parsimonious industry equilibrium model with endogenous productivity to study high-tech firms' cash-saving policy and explore its role in innovation. The model incorporates multiple well-cited explanations in the literature and examines which factors are the main determinants of high-tech firms' saving behavior. We find that innovation uncertainty is the major driver, followed by knowledge spillover and financial frictions. Market competition has non-monotonic effects, while R&D adjustment costs play a relatively minor role. We also find that even if the productivity process is transitory, firms manage to save enough to reduce R&D distortions from financial frictions.

Abstract This chapter discusses how the strategic interaction between incumbents and innovators in the market for ideas shapes (and is shaped by) the potential for product market competition. On the one hand, if the market for ideas is efficient (e.g., there can be perfect, low-cost transfer of both new designs and process innovations), then incumbents and entrants will have an incentive to cooperate (rather than compete) in the commercialization process. However, when technology transfer of either product designs or processes is imperfect, then innovators will have an incentive to enter the product market and so start-up innovation will be associated with increased competition. An overarching lesson of the analysis is that the incentives for entry are higher when the underlying technologies are more (horizontally) differentiated from each other. Because the gains from cooperation are higher when the degree of differentiation is lower, the likelihood of entrepreneurial entry is higher under conditions of high product differentiation and imperfect technology transfer.

This article develops and estimates an industry equilibrium model of the Korean electric motor industry from 1991 to 1996. Plant‐level decisions on R&D, physical capital investment, entry, and exit are integrated in a dynamic setting with knowledge spillovers. We apply the novel approximation of oblivious equilibrium to estimate the R&D cost, magnitude of knowledge spillovers, adjustment costs of physical investment, and plant scrap value distribution. Knowledge spillovers are essential to explaining the firm‐level productivity evolution and the equilibrium market configuration. A R&D subsidy maximizes industry output and is broadly consistent with a past policy initiative of the Korean government.

We focus on human capital measured by education outcomes (skills) and establish the relationship between human capital, R&D investments, and productivity across 12 OECD economies and 17 manufacturing and service industries. Much of the recent literature has relied on school attainment rather than on skills. By making use of data on adult cognitive skills from the Programme for the International Assessment of Adult Competences (PIAAC), we compute a measure of sectoral human capital defined as the average cognitive skills in the workforce of each country-sector combination. Our results show a strong positive relationship between those cognitive skills and the labour productivity in a country-sector combination. The part of the cross-country cross-sector variation in labour productivity that can be explained by human capital is remarkably large when it is measured by the average sectoral skills whereas it appears statistically insignificant in all our specifications when it is measured by the mere sectoral average school attainment. Our results corroborate the positive link between R&D investments and labour productivity, finding elasticities similar to those of previous studies. This evidence calls for a focus on educational outcomes (rather than on mere school attainment) and it suggests that using a measure of average sectoral cognitive skills can represent a major step forward in any kind of future sectoral growth accounting exercise.

Using population-wide data on a vector of cognitive abilities and productive non-cognitive traits among Swedish male workers, we show that occupational employment growth has been monotonically skill-biased in terms of these intellectual skills, despite a simultaneous (polarizing) decline in middle-wage jobs. Employees in growing low-wage occupations have more of these skills than employees in other low-wage occupations. Conversely, employees in declining, routine-task intensive, mid-wage occupations have comparably little of these skills. Employees in occupations that have grown relative to other occupations with similar wages have more intellectual skills overall but are particularly well-endowed with the non-cognitive trait “Social Maturity” and cognitive abilities in the “Technical” and “Verbal” domains. Projections from the US Bureau of Labor Statistics about future occupational labor demand do not indicate that the relationship between employment growth and skills is about to change in the near future.

This paper examines the importance of knowledge-skill complementarity in the process of contemporary economic growth. By analyzing Dutch manufacturing and carrying out an extensive spillover and wage inequality analysis, it is shown that knowledge-intensive sectors pay their high-skilled workers a relatively higher wage in the form of a wage premium, which is defined as the sector bias of technical change.

We estimate the effect of imported machines on the wages of machine operators utilizing Hungarian linked employer-employee data. We infer exposure to imported machines from detailed trade statistics of the firm and the occupation description of the worker. We find that workers exposed to imported machines earn about 8 percent higher wages than other machine operators at the same firm. When we proxy for unobserved worker characteristics, we find a significant 3 percent wage premium, suggesting that the relationship is causal. The return to schooling is also higher on imported machines. We build a simple matching model consistent with these findings. Our findings suggest that machine imports can be an important channel through which skill-biased technical change reaches less developed and emerging economies.

The objective of the paper is to study how the tax burden arising from an exogenous stream of public expenditures and transfers should be distributed between labor and capital in a scale-less endogenous growth model, where the engine of growth are successful innovations. Our laboratory is a prototypical quality ladder model with a labor/leisure choice where research and development productivity is decreasing in the size of the economy. Our contribution is to show that even when labor supply has no effects on growth in the long run, it will still be optimal to tax capital for reasonable parametrizations of the model.

What quantitative lessons can we learn from models of endogenous technical change through innovative investments by firms for the impact of changes in the economic environment on the dynamics of aggregate productivity in the short, medium, and long run? We present a unifying model that nests a number of canonical models in the literature and characterize their positive implications for the transitional dynamics of aggregate productivity and their welfare implications in terms of two sufficient statistics. We review the current state of measurement of these two sufficient statistics and discuss the range of positive and normative quantitative implications of our model for a wide array of counterfactual experiments, including the link between a decline in the entry rate of new firms and a slowdown in the growth of aggregate productivity given that measurement. We conclude with a summary of the lessons learned from our analysis to help direct future research aimed at building models of endogenous productivity growth useful for quantitative analysis.

Purpose The main purpose of this paper is to explore whether the nature of innovation is accumulative or radical and to what extent past year accumulation of technology stock can predict future innovation. More importantly, the authors are concerned with whether a change of policy regime or a variance in the quality of technology will moderate the nature of innovation. Design/methodology/approach The authors examined a dataset of 3.6 million Chinese patents during 1985–2015 and constructed more than 5 million citation pairs across 8 sections and 128 classes to track knowledge spillover across technology fields. The authors used this citation dataset to calculate the technology innovation network. The authors constructed a measure of upstream invention, interacting the pre-existing technology innovation network with historical patent growth in each technology field, and estimated measure's impact on future innovation since 2005. The authors also constructed three sets of metrics – technology dependence, centrality and scientific value – to identify innovation quality and a policy dummy to consider the impact of policy on innovation. Findings Innovation growth is built upon past year accumulation and technology spillover. Innovation grows faster for technologies that are more central and grows more slowly for more valuable technologies. A pro-innovation and pro-intellectual property right (IPR) policy plays a positive and significant role in driving technical progress. The authors also found that for technologies that have faster access to new information or larger power to control knowledge flow, the upstream and downstream innovation linkage is stronger. However, this linkage is weaker for technologies that are more novel or general. On most occasions, the nature of innovation was less responsive to policy shock. Originality/value This paper contributes to the debate on the nature of innovation by determining whether upstream innovation has strong predictive power on future innovation. The authors develop the assumption used in the technology spillover literature by considering a time-variant, directional and asymmetric matrix to model technology diffusion. For the first time, the authors answer how the nature of innovation will vary depending on the technology network configurations and policy environment. In addition to contributing to the academic debate, the authors' study has important implications for economic growth and industrial or innovation management policies.

Akcigit's research focuses on economic growth, productivity, firm dynamics, and the economics of innovation

This paper provides quasi-experimental evidence on the long-term causal effect of increases in human capital on participation in agriculture. We use variation in male educational attainment generated by Indonesia’s Sekolah Dasar INPRES program, one of the largest ever school building programs. Consistent with the first evaluation [Duflo, 2001], we find that males exposed to a higher program intensity have improved measures of human capital as adults. We then show that treated cohorts are more likely to be employed outside of agriculture–particularly in industry–and less likely to be agricultural workers. Then, exploiting variation in exposure across adjacent districts, we demonstrate that higher INPRES intensity in neighboring districts decreases non-agricultural employment and earnings, consistent with cross-district spillovers mediating the total impacts. Together, the results suggest that government investment in human capital can have profound effects on the rural economy and may help to accelerate shifts away from agriculture.

In this article, we present a unified treatment of and explanation for the evolution of wages and employment in the US over the last 30 years. Specifically, we account for the pattern of changes in wage inequality, for the increased relative wage and employment of women, for the emergence of the college wage premium and for the shift in employment from the goods to the service-producing sector. The underlying theory we adopt is neoclassical, a two-sector competitive labor market economy in which the supply of and demand for labor of heterogeneous skill determines spot market skill rental prices. The empirical approach is structural. The model embeds many of the features that have been posited in the literature to have contributed to the changing US wage and employment structure including skill-biased technical change, capital-skill complementarity, changes in relative product-market prices, changes in the productivity of labor in home production and demographics such as changing cohort size and fertility.

Existing skill-biased technical change theory predicts that differences in factor endowments will affect technology adoption and the return to skill. We document regional variation in endowments in the American past. We then estimate the returns to education using a new data source: a report from the Commissioner of Education in 1909. We find significant variation in the returns to schooling aligned with differences in resource endowments, with large (within-occupation) returns in the Midwest and Southwest but much lower returns in the South and West. Our results appear generalizable to broader returns to education in the United States.

This paper analyzes whether computerization can be associated with a shift in the value of skills, using skill scores from the 1999 Higher Education and Graduate Employment in Europe Survey and IT use by region. We develop a GLS estimator for random coecient models to deal with unbalanced panels and a mixture of fixed parameters and parameters systematically varying between regions to investigate whether or not the coecients of the skill scores in the wage function depend on the degree of computerization in a region. We oer an analysis of the impact of IT on the value of skills and obtain that the complementarity between IT and skills is not straightforward, but that IT generally substitutes for tasks requiring skills such as cooperation and team working and complements tasks requiring analytical skills.

Les étudiants qui décident de poursuivre des études supérieures sont confrontés à des choix d’orientation complexes qui ont des conséquences majeures en termes de salaires et de trajectoire professionnelle. Il est crucial de comprendre comment ces décisions sont formées, afin notamment d’aboutir à des politiques publiques permettant d’infléchir les inégalités d’accès aux différents cursus. La recherche en économie sur cette question met l’accent sur le rôle essentiel joué par les préférences et les aptitudes pour les différents types de formations, sur l’importance des imperfections d’information ainsi que sur l’effet du revenu familial sur l’orientation des étudiants.

Using administrative data for college graduates, we model earnings and employment probabilities as functions of a credit-weighted index of the occupational specificity of college coursework, decomposed into within-major, within-discipline (but outside the major), and nondisciplinary components. We define the occupational specificity of each college field as the likelihood that a student majoring in that field subsequently works in an occupation requiring specific skills acquired in the field. We find that occupationally-specific, non-disciplinary courses are strongly associated with earnings; e.g., a five percentage-point shift among English majors from their least occupationally-specific courses outside the humanities to computer science is associated with a 0.055 increase in log-earnings.

This paper develops a multi-sector endogenous growth model that includes an innovation network, which captures intrasectoral as well as heterogeneous intersectoral knowledge flows. We analyze the importance of sectors (nodes) and directed knowledge linkages (edges) in the innovation network by their contribution to the growth of knowledge in this economy. We show that the growth rate of knowledge is equal to the spectral radius of the innovation network. We also demonstrate that a sector's importance to growth (``key sectors'') is related to its positions in both the downstream and upstream technology network. Finally, the importance of a knowledge linkage is characterized by both the upstream centrality of its source sector, the downstream centrality of its target sector and the strength of knowledge flows from the source sector to the target sector.

In the standard Schumpeterian‐growth models only follower firms invest in R&D activities and larger economies grow faster. Since these results are counterfactual, this paper reveals that leader firms often support R&D activities and economic growth can be independent of the market size. In particular, the maintenance of R&D leadership increases with: (i) the technological‐knowledge gap between leader and followers, since a firm‐specific learning effect of accumulated technological knowledge from past R&D is considered, (ii) the leaders' strategies that delay the next successful R&D supported by some follower firm, (iii) the market size, and (iv) the up‐grade of each innovation.

Abstract The rate of innovation in Information Technology (IT) has slowed down over time. The slowdown is evident both in the data on quality-adjusted prices of computers, and performance of microprocessors used in computers. The model in this paper shows that an IT–labor elasticity of substitution that is greater than 1 can explain the slowdown. With an elasticity of substitution greater than 1, however, slowing innovation can result in sustained labor productivity and output growth. Sustained growth is possible because an IT–labor elasticity of substitution greater than 1 results in a continuously increasing share of IT in production costs, which counteracts the effect of slowing innovation on labor productivity and output growth. In this environment of slowing innovation, increasing IT share and sustained growth, employment can increase or decrease, depending on the values of the IT–labor elasticity of substitution and the price elasticity of demand for IT-enabled consumption goods.

Abstract The paper discusses the knowledge cost approach as a comprehensive framework to account for endogenous technological change and test it to explain productivity differences across European regions. The assessment of the limited transferability of knowledge and the appreciation of the intentional efforts required to use knowledge spillovers question the assumptions of automatic, spontaneous, homogenous, symmetric and universal effects of knowledge spillovers conjectured by the New Growth Theory. The knowledge cost approach, instead, stresses the localized, idiosyncratic and contextual effects of knowledge spillovers that are strong -only- in high-quality innovation systems. If the access and absorption of knowledge in high-quality innovation systems is cheaper, the cost of knowledge falls below equilibrium levels and its use in the technology production function contributes to higher total factor productivity growth rates. Using a sample of 192 European regions for which we estimate productivity growth for the period from 2005 to 2020, we confirm that regions with lower knowledge costs exhibit higher Total Factor Productivity growth rates.

Current estimates suggest that over the coming decades, slower population growth and lower labor force participation will constrain the supply of labor in the U.S. The U.S. labor force will also become more diverse as immigration and fertility trends increase the size of minority populations. New forms of automation will likely require workers to adapt to keep their old jobs, while many will be displaced or face less demand for their work (while others benefit). Firms will continue to implement alternative staffing arrangements, like turning workers into independent contractors or outsourcing their human resource management to other firms; and many will adopt "low-road" employment practices to keep labor costs low. Exactly whom these changes will benefit or harm remains unclear, though non-college workers will likely fare the worst; higher productivity from new technologies and reduced labor supply could raise average wages, but many workers will clearly be worse off. Policy makers should provide incentives for firms to train current employees, rather than replace them, and should encourage schools and colleges to teach flexible, transferable skills, as the future workforce will likely need to adapt quickly to new and changing job requirements. Lifelong learning accounts for workers could help. Expanding wage insurance and improving unemployment insurance and workforce services could help workers adapt after suffering job displacement. Policies that make work pay, like the EITC, and others designed to increase labor force attachment, like paid family leave, could help mitigate declines in the labor force. Reforms in immigration and retirement policy will help as well, as would policy experimentation at the state and local level (with federal support).

This study investigates the non-monetary rewards associated with artificial intelligence (AI) skills in the U.S. labour market. Using a dataset of approximately ten million online job vacancies from 2018 to 2024, we identify AI roles-positions requiring at least one AI-related skill-and examine the extent to which these roles offer non-monetary benefits such as tuition assistance, paid leave, health and well-being perks, parental leave, workplace culture enhancements, and remote work options. While previous research has documented substantial wage premiums for AI-related roles due to growing demand and limited talent supply, our study asks whether this demand also translates into enhanced non-monetary compensation. We find that AI roles are significantly more likely to offer such perks, even after controlling for education requirements, industry, and occupation type. It is twice as likely for an AI role to offer parental leave and almost three times more likely to provide remote working options. Moreover, the highest-paying AI roles tend to bundle these benefits, suggesting a compound premium where salary increases coincide with expanded non-monetary rewards. AI roles offering parental leave or health benefits show salaries that are, on average, 12% to 20% higher than AI roles without this benefit. This pattern is particularly pronounced in years and occupations experiencing the highest AI-related demand, pointing to a demand-driven dynamic. Our findings underscore the strong pull of AI talent in the labor market and challenge narratives of technological displacement, highlighting instead how employers compete for scarce talent through both financial and non-financial incentives.

Career progression is important for people's lives and economic decisions.We develop an empirical measure of an occupation's local "career value"-the long-run value of the earnings that will result from working in that job and following the career ladder associated with it.We then document that career values have been stagnating over the 2000-2016 period, in spite of growing wages, due to a deterioration in career mobility.We estimate the effect of robot automation on career values over the same time period and find that one additional industrial robot per 1,000 workers lowered local career values by about 1.5 percent.The reason is that robotization reduces transitions into better-paid occupations and redirects workers toward similar-or lower-paid jobs.The impact is largest in high-manufacturing areas, for mid-experience workers, and for males.Demotions from management jobs that result from robotization are more likely for less-educated workers and for women, who are more likely to respond by upskilling.Declines in career values led to a reduction in housing construction and college enrollment, and an increase in Republican vote shares in 2016, which highlights how the career effects of automation shape forward-looking household decisions.

An issue with estimating the impact of industrial support is that the firms that receive support may be politically connected, introducing omitted variable bias. Applying fixed-effects regressions on Vietnamese panel data containing several proxies for political connectedness to correct this bias, we find that firms that receive industrial support in the form of tax holidays experience more rapid productivity growth, particularly in R&D-intensive industries, and less so among politically connected firms. These findings do not appear to be due to the presence of financing constraints. We then develop a second-generation Schumpeterian growth model with many industries, and show that tax holidays disproportionately raise productivity growth in R&D-intensive industries. These results are significant and important for governments, especially those in transition and developing countries, in better targeting their industrial policy to facilitate higher productivity growth.

University programs differ in the subsequent earnings processes of their enrollees, including many features that students might care about to differing degrees such as the level of average earnings, earnings growth, and volatility. Do the earnings features of a university program’s enrollees reflect the causal effect of enrolling in that program or the self-selection of students into that program? Would students experience a different earnings process if they enrolled in a different program of study? To estimate the causal impact of enrolling in a program of study on the enrollees’ future earnings process, we exploit a discontinuity built into the Danish national university admissions system, which provides quasi-random assignment of similar applicants to different programs. We leverage the rich cross-program variation in the enrollees’ future earnings processes to measure the impact of entering a program whose enrollees experience high earnings levels, growth, and volatility on their own subsequent earnings level, growth, and volatility. We find that a student’s subsequent earnings levels and volatility – but not their earnings growth – are caused by entering programs of study whose enrollees have those features.

A considerable amount of research at the micro level has shown that that carbon tax combined with research subsidy may be regarded as an optimal policy in view of diffusing low carbon technologies for the benefit of the society. The paper exploits the macro economic approach of the endogenous growth models with technological change for a comparative assessment of these policy measures on the economic growth in the US and Japan in the medium and the long run. The results of our micro estimates reveal several important differences across the Japanese and US energy firms: lower elasticity of innovation production function in R&amp;D expenditure, lower probability of a radical innovation, and larger advances of dirty technologies in Japan. This may explain our quantitative findings of stronger reliance on carbon tax than on research subsidies in Japan relative to the US.

We examine the effect of board size on the adoption of artificial intelligence (AI)-skilled employees, a critical determinant of firms’ ability to leverage AI technologies. Board size, a crucial dimension of board governance, plays a pivotal role in shaping strategic decision-making and a firm's adaptability to technological change. Using a novel dataset by Babina et al. (2024) that identifies AI-related roles through advanced textual analysis of resumes, we utilize the share of AI workers in firms over time. Our findings reveal that smaller boards significantly enhance the integration of AI-skilled employees. Specifically, a reduction in board size by one standard deviation increases the share of AI workers by 8.4 %. Interaction analyses show that smaller boards are particularly advantageous in R&D-intensive firms and those with substantial cash reserves. Smaller boards also result in greater variability in AI workforce integration, reflecting their capacity to foster flexibility, and adaptive learning in dynamic environments.

Artificial Intelligence (AI) is being hailed as the latest example of a General Purpose Technology that could transform productivity and help tackle important societal challenges. This outcome is however not guaranteed: a myopic focus on short-term benefits could lock AI into technologies that turn out to be sub-optimal in the longer-run. Recent controversies about the dominance of deep learning methods and private labs in AI research suggest that the field may be getting narrower, but the evidence base is lacking. We seek to address this gap with an analysis of the thematic diversity of AI research in arXiv, a widely used pre-prints site. Having identified 110,000 AI papers in this corpus, we use hierarchical topic modelling to estimate the thematic composition of AI research, and this composition to calculate various metrics of research diversity. Our analysis suggests that diversity in AI research has stagnated in recent years, and that AI research involving private sector organisations tends to be less diverse than research in academia. This appears to be driven by a small number of prolific and narrowly-focused technology companies. Diversity in academia is bolstered by smaller institutions and research groups that may have less incentives to race and lower levels of collaboration with the private sector. We also find that private sector AI researchers tend to specialise in data and computationally intensive deep learning methods at the expense of research involving other (symbolic and statistical) AI methods, and of research that considers the societal and ethical implications of AI or applies it in domains like health. Our results suggest that there may be a rationale for policy action to prevent a premature narrowing of AI research that could reduce its societal benefits, but we note the incentive, information and scale hurdles standing in the way of such interventions.

Green innovation is a pivotal way to achieve both environmental and economic bene- fits. This study constructs an evolutionary game model to validate the Porter Hypothesis from the perspective of guiding innovation resources toward green innovation. The findings indicate that (1) Environmental regulation can reset the expected returns of different innovation directions. When the value gap between different innovation directions exceeds a certain threshold, innovation direction selection strategies will eventually evolve into a dispersed innovation mode. This not only directs in- novation resources towards green but also avoids the problem of innovation direction congestion. (2) Environmental regulation should be positively proportional to the economic value gap of innovation directions and the extent of environmental tax reduction, and inversely proportional to the pollution emission gap. (3) Weaker environmental regulations should be implemented to maximize social benefits when the economic value and pollution emission gap are both small. This reduces compliance costs for innovators and achieves higher social benefits. Conversely, when the economic value gap is large, stronger environmental regulations should be implemented to alleviate innovation direction congestion, and ensure resource allocation for green innovation. Therefore, it is essential to reasonably adjust the environmental regulations to achieve a positive cycle of sustainable development. Additionally, hetero- geneous environmental regulations should be designed and implemented for different industries and types of innovation.

There is widespread and growing concern about the availability of good jobs in the U.S. economy. Inequality has been growing for thirty years and is now at levels not seen since the 1920s. Stable and remunerative employment has become harder for U.S. workers to find. With the widespread plant closings of the 1980s, the loss of these middle-class employment opportunities was confined largely to blue-collar workers with high-school educations. As a group, members of the U.S. labor force with college educations always do better than those with high-school educations, but over the course of the 1980s the wage premium to having a college education expanded significantly. During the 1990s and 2000s, however, older and experienced college-educated white-collar workers began to find their earnings under pressure as the career away from the norm of a white-collar career with one company that had prevailed since the 1940s. Then in the 2000s U.S. white- collar workers faced the incessant offshoring of jobs to be filled by college-educated workers in lower-wage developing economies, with India and China in the forefront. The Great Recession of 2007 to 2009 and its aftermath have only heightened fundamentally from the dominant paradigm among economists known as skill-biased technical changes (SBTC). Like all economists who adhere to the neoclassical theory of the market economy, SBTC assumes that wages are determined through the forces of supply and demand in the labor market. In contrast, our study of the development of the U.S. economy over the past half century views the primary determinant of wages on a sustainable basis as the employment practices of major business enterprises. We contend that, except when labor is an interchangeable commodity, wages are determined in business organizations where the promise of wage increases over time are both an inducement to supply more and better work effort when engaged in productive activities, and a reward for having done so in ways that add value over time. For employees in high-tech fields – known collectively as STEM (science, technology, engineering, and mathematics) workers – wages are determined not by supply and demand in the labor market but rather by the employment relations that prevail within leading business enterprises. The reason: Productivity in high-tech fields depends on learning that is both collective and cumulative. Focusing on STEM employment, we explore the hypothesis that the productivity and earnings of high-tech workers depend on collective and cumulative careers.

This paper analyzes the pattern of occupational change in four Western European countries over the last two decades: what kind of jobs have been expanding -- high-paid jobs, low-paid jobs or both? By addressing this issue, we also examine what theoretical account is consistent with the observed pattern of change: skill-biased technical change, skill supply evolution or wage-setting institutions? Our empirical findings show a picture of massive occupational upgrading that closely matches educational expansion. In all four countries, by far the strongest employment growth occurred at the top of the occupational hierarchy, among managers and professionals. Yet in parallel, in Britain and Switzerland, as well as in Germany and Spain after 1996 and 2002 respectively, relative employment declined more strongly in the middling occupations (among clerks and production workers) than at the bottom (among interpersonal service workers). This slightly polarized pattern of occupational upgrading is consistent with the "routinization" hypothesis that technology is a better substitute for average-paid jobs in production and the office that for low-paid jobs in interpersonal services. However, we find large cross-country differences in the employment evolution at the bottom of the occupational hierarchy, among low-paid services workers: sizeable growth in Britain and Spain, but stagnation in Germany and Switzerland. This results points towards the possibility that wage-setting institutions filter the pattern of occupational change.

W opracowaniu przedstawiono społeczne i biznesowe uwarunkowania automatyzacji procesów pracy oraz bariery i zakres stosowania tego typu rozwiązań z perspektywy treści pracy. W ramach tak zdefiniowanego zamierzenia wyjaśniono istotę pojęcia automatyzacji, odnosząc się w tym względzie do terminów korespondujących z pojęciem automatyzacji. Na bazie przeprowadzonego badania bibliometrycznego ukazano wzrost liczby publikacji poświęconych automatyzacji, robotyzacji i mechanizacji. Na bazie przeglądu literatury przedmiotu dokonano identyfikacji zjawisk opisujących wpływ postępu technicznego na treść pracy, w szczególności wymagania stawiane pracownikom w procesach pracy. W ramach tego wątku dokonano zaprezentowano trzy efekty, w tym tzw. unskill-biased technical change, skill-biased technical change oraz talent-biased technical change. W dalszej kolejności wskazano podejścia umożliwiające szacowanie potencjalnego zakresu automatyzacji procesów pracy oraz ryzyka automatyzacji poszczególnych zawodów. Podejścia te mogą stanowić punkt wyjścia do opracowania narzędzia diagnostycznego umożliwiającego szacowanie potencjalnego zakresu prac podlegających automatyzacji oraz identyfikację czynników i zmiennych stanowiskowych stanowiących bariery automatyzacji procesów pracy.

This paper studies the impact of human capital on the adoption and diffusion of Information and Communications Technologies (ICT) in the Pakistani firms using the World Bank Enterprise Survey 2002-07. The paper considers various indicators of human capital and measures of ICT adoption and diffusion. On-the-job training, manager's level of qualification and production workers' level of education are found to positively determine the use of emails, website and other means of communication in a firm. The results are robust to the inclusion of geographical, sectoral and structural control variables. Firm size, sales and workers' compensation are also positively associated with the use of ICT. The findings show the importance of accumulation and development of human capital in the productivity growth in the era of skill-biased technical change. A concerted national effort for the enhancement of the workforce's computing skills is therefore a must if a developing economy such as Pakistan is to improve its competitiveness.

Avec l'émergence de l'IA générative (ChatGPT), la révolution de l'intelligence artificielle (IA) connaît une accélération sans précédent : simplicité d'utilisation des outils, très large éventail d'applications, rapidité de génération de contenus riches et complexes, réalisme des textes, images et sons générés...Cette révolution inouïe suscite craintes et espoirs : craintes que soient massivement détruits des emplois, craintes d'une dégradation accrue de notre environnement, craintes d'utilisations dévoyées de l'IA avec notamment une remise en cause des droits d'auteur ; espoir que l'IA permette enfin de sortir de la croissance atone que nous subissons depuis des années, espoir qu'elle améliore notre qualité de vie dans ses différents aspects. Réunie par Philippe Aghion et Anne Bouverot, la Commission de l'intelligence artificielle se défend de tout excès de pessimisme ou d'optimisme. Elle souligne qu'il nous faudra investir massivement dans la formation, la puissance de calcul, l'accès aux données collectives et la transformation des entreprises pour que l'IA devienne un véritable facteur de progrès. Avec cette conviction partagée : l'Europe et la France ont des atouts pour être des acteurs de cette révolution, mais elles doivent rapidement se mettre en ordre de marche, et sur le long terme. C'est avec cette ambition que la Commission propose un plan d'action au service de nos besoins, de nos valeurs et de nos principes

Defence date: 7 June 2007

The main goal of this thesis is to investigate the reasons why firms do often adopt inefficient technologies even when superior ones are widely available, and to assess their consequences. From a macroeconomic perspective it has been emphasized that differences in the adoption and diffusion rates of technology have a significant impact on economic growth and development, affecting output and productivity differentials among countries. The importance of these issues explains why the understanding of the determinants of technical change has attracted a great deal of attention by both theorists and applied economists.\nA firm's technology choice and its timing rests on the expected costs and benefits of adoption, which in turn are a function of a number of microeconomic and macroeconomic factors. Several such factors have been investigated in the literature, giving rise to patterns of technology adoption more or less successful when brought to the data. One single feature of technology adoption that is widely emphasized is the role of technology-induced spillovers. This dissertation, after surveying the major approaches to technology adoption found in the literature and stressing their drawbacks, studies the effects on firms' choices of technology-induced production externalities that are largely consistent with the empirical evidence and whose relevance has not been previously assessed. The thesis central claim is the existence of a causal link between the effects of firms' choices of a technology and wages. The choice of technology determines an increase in workers' productivity and consequently an improvement of their occupational alternatives, that can transfer on the wages a firm must pay in order to retain its employees.\nIt is shown, first in an efficiency wage partial equilibrium framework and then in a simple general equilibrium setting, that the presence of production externalities of the sort described above can lead to inefficient technology adoption by firms, so that they may not have an incentive to upgrade to the technological frontier, remaining stuck with old and inefficient technologies. Finally, the possibility of technology misallocation is investigated from a normative point of view, characterizing Pareto-efficient allocations and discussing the role of government interventions, through non-linear (first best) subsidization and second best policy instruments, to overcome or mitigate market failures.

Abstract This paper empirically investigates wage inequality within the group of skilled workers in the recent four decades in the USA using CPS data and finds evidence that the trend of wage growth of the top and bottom 10th percentile of skilled workers significantly diverged starting from 2000. Using a task-based framework of occupation, I find that the changing trend of wage inequality was entirely driven by one category of occupation, namely the nonroutine analytic occupation. Then, I consider in a model task reallocation between two broad task categories, namely the routine and abstract task, induced by an ongoing investment-specific technical change. In my model, the labor in the routine task is replaced by cheaper machines due to investment-specific technical change, then workers that are less productive in the abstract task enter abstract occupations. As a result, the wage inequality in the abstract task widens because of the reallocation of less productive workers from the routine task to the abstract task, that is, the “composition effect.” In addition, since economic agents tend to postpone the investment in machines after the ongoing investment-specific technical change takes place for a while, the expansion path of wage inequality is not linear but features an acceleration of wage dispersion in the middle of the technical change. The quantitative results suggest that the model is able to provide a well-matched timing and magnitude of the nonlinear expansion path in wage inequality that is observed in the data.

University innovation plays an increasingly significant role in regional industrial development. In this paper, we study how local industrial initiatives affect the university innovation activities. We link preferred initiatives from Chinese provincial Five-Year Plans to university–industry patent data. Using difference-in-differences design, we document that local preferred industrial initiatives significantly enhance university innovation. These initiatives increase public Research and Development (R&D) funding (government-push effect) and facilitate regional university–industry collaboration (market-pull effect). The effects exhibit heterogeneity across university administrative affiliation, university research capacity, and industry technology intensity. Furthermore, regional industrial comparative advantages and university technology transfer capabilities strengthen the innovation-enhancing effects. Finally, this paper demonstrates that university research capacity strengthens the effectiveness of industrial initiatives on firm output. These findings underscore the synergy among industrial initiatives, university innovation, and local development.

This paper examines the relationship between subject specialization in high school and university undergraduate degree program choices. Focusing on a reform in England that encouraged students to opt for studying mathematics in the last 2 years of high school, the study analyzes its effect on undergraduate enrollment in Science, Technology, Engineering, and Mathematics (STEM) fields. The findings indicate that the reform increased the likelihood of students pursuing and completing STEM undergraduate degrees. Thus, encouraging mathematics specialization during high school enhances the number of STEM graduates. However, despite the reform’s implementation, gender and socioeconomic disparities in STEM participation remained unchanged, suggesting that interventions during adolescence might not effectively address the underrepresentation of specific groups, such as females, in STEM programs.

We consider a model in which appropriate organization fosters innovation, but because of contractibility problems, this benefit cannot be internalized. The organizational design element we focus on is the division of labor, which as Adam Smith argued, facilitates invention by observers of the production process. However, entrepreneurs choose its level only to facilitate monitoring their workers. Whether there is innovation depends on the interaction of the markets for labor and for inventions. A high level of specialization is chosen when the wage share is low. But low wage shares arise only when there are few entrepreneurs, which limits the market for innovations therefore and discourages inventive activity. When there are many entrepreneurs, the innovation market is large, but the rate of invention is low because there is little specialization. Rapid technological progress therefore requires a balance between these opposing effects, which occurs with a moderate relative scarcity of entrepreneurs and workers. In a dynamic version of the model in which a credit constraint limits entry into entrepreneurship, this relative scarcity depends on the wealth distribution, which evolves endogenously. There is an inverted-U relation between growth rates driven by innovation and the level of inequality. Institutional improvements have ambiguous effects on growth. In light of the model, we offer a reassessment of the mechanism by which organizational innovations such as the factory may have spawned the industrial revolution.

Abstract We propose Computational Thinking (CT) as an innovative pedagogical approach with broad application. Research and current industry trends illustrate that students should have a solid computational thinking ability in order to have the skills required for future jobs in Artificial Intelligence. Due to current social issues regarding COVID-19 and natural disasters, we are rapidly moving towards a cyberspace era where many citizens will conduct their work online. Understanding the foundations and tools of computation – e.g., abstraction, decomposition, pattern recognition – is critical for any student to be prepared for the digital AI age. Believing students should be fully prepared for future jobs that involve computation, we developed a CT module on a Learning Management System (LMS). We have collected data of students who took our CT course module. We looked into the students’ activity records and analyzed the number of students’ views on the pages and the number of participants on each quiz. We counted the total number of engagements of the ten components in the CT course module. Ultimately, we believe that our modules had a greater impact on those students who were newer to computational thinking, over those who had prior experience and were enrolled in upper-level computational courses.

This paper studies how industry-specific shocks affect workers with specialized skills, focusing on the impact of the burst of the dotcom bubble on the careers of IT-specialized college graduates in Sweden. Graduates entering the labor market during the bust faced sharp initial earnings penalties and lower probabilities of IT sector employment compared to boom cohorts. However, they exhibited remarkable resilience, recovering earnings by leveraging their skills in high-paying, non-IT occupations. Incumbent IT workers, while remaining within the IT sector, experienced a decline in earnings as they moved to lower-premium firms. • The dotcom bust sharply cut initial earnings for new IT graduates. • Graduates recovered by using their IT skills in other high-paying occupations. • Incumbent IT workers’ earnings fell as they moved to lower-premium firms within the IT sector.

We consider a model of innovation that distinguishes between ideas and innovations. Although innovation responds to incentives, ideas are a scarce resource that provides an exogenous constraint on the rate of innovation. We investigate how the optimal reward structure is shaped by the scarcity of ideas. Substitute ideas for innovation in the model arrive to random recipients at random times. By forgoing investment in a current idea, society preserves an option to invest in a better idea for the same market niche but with delay. Because successive ideas may occur to different people, there is a conflict between private and social optimality. The social planner does not observe the arrival of ideas and learns over time about the arrival rate of ideas. We represent the social planner’s beliefs using a general continuous density and illustrate that the evolution of beliefs can be tracked by the cumulative hazard function. The reward set by the social planner serves a dual purpose: learning about the arrival rate of ideas and trading off lower delays against lower-cost ideas. We show that the optimal reward increases as time passes because the social planner comes to view ideas as more scarce. This paper was accepted by Joshua Gans, business strategy. Supplemental Material: The online appendix is available at https://doi.org/10.1287/mnsc.2023.03362 .

Abstract An entrant and an incumbent allocate their research funds across a rival and a non‐rival market. The prospect of an acquisition distorts both players' incentives to allocate funding. Allowing for acquisitions may improve both players' innovation direction and consumer surplus. Under conditions, the incumbent, anticipating monopolization rents in the rival market, moves R&D towards that market. This “incumbency for buyout” effect lowers the rents the entrant obtains from the contestable market, which gives it incentives to move R&D resources away from the rival market. Such strategic interaction in the R&D market has implications for the assessment of start‐up acquisitions.

Balancing the demands of work and schooling is a challenging task for an increasing number of students who have to pay their way through college and for workers who intend to upgrade their skills. However, flexible learning and working environments could play an important role in easing many frictions associated with performing both activities simultaneously. Using detailed (work and study effort) data - from a partnership between Arizona State University and Uber that allows eligible drivers to enroll in online college courses for free - we analyze how labor supply and study efforts respond to changes in labor market conditions and college activities/tasks. Our findings indicate that a 10% increase in average weekly online college activities reduces weekly time spent on the Uber platform by about 1%, indicating a low 'short run' opportunity cost of studying when working. We also show that study time is not particularly sensitive to changes in labor market conditions, where a 10% increase in average weekly pay reduces study hours by only 2%. Consistent with these results, we find that workers take advantage of their flexible schedules by changing their usual working hours when their college courses are more demanding. We do not find adverse effects of work hours on academic performance in this context, or of study hours on workplace performance (as measured by driver ratings or tips). Finally, the (elicited) value assigned to flexible working and educational formats is high among the students in our sample, who view online education as an important vehicle for increasing expected future income. Overall, this study underscores that combining flexible working and learning formats could constitute a suitable path for many (low-SES) students who work to afford an increasingly expensive college education and for workers aiming to improve their skill set.

This paper explores the symbiotic relationship between transformative entrepreneurs and inventors, which is crucial for economic growth. We utilize microdata from Denmark to demonstrate that while the relationship between IQ and general entrepreneurship tends to be negative, it is strongly positive among transformative entrepreneurs. Transformative entrepreneurs, often with higher IQ and education levels, significantly drive R&amp;amp;amp;D and business growth, thereby providing substantial opportunities for inventors. In contrast, average entrepreneurs are more influenced by their family's entrepreneurship background. Our economic model links these dynamics to overall economic progress, highlighting how higher education influences career paths in entrepreneurship and invention. We identify talent misallocation caused by unequal education access, particularly affecting lower-income families. Our findings indicate the most effective policies strengthen the interplay between higher education, innovation, and entrepreneurship to foster transformative businesses and achieve long-run economic growth.

This paper explores the symbiotic relationship between transformative entrepreneurs and inventors, which is crucial for economic growth.We utilize microdata from Denmark to demonstrate that while the relationship between IQ and general entrepreneurship tends to be negative, it is strongly positive among transformative entrepreneurs.Transformative entrepreneurs, often with higher IQ and education levels, significantly drive R&D and business growth, thereby providing substantial opportunities for inventors.In contrast, average entrepreneurs are more influenced by their family's entrepreneurship background.Our economic model links these dynamics to overall economic progress, highlighting how higher education influences career paths in entrepreneurship and invention.We identify talent misallocation caused by unequal education access, particularly affecting lower-income families.Our findings indicates the most effective policies strengthen the interplay between higher education, innovation, and entrepreneurship to foster transformative businesses and achieve long-run economic growth.

In the 1970s and 1980s the US position as the global technological leader was increasingly challenged by Japan and Europe. In those years the US skill premium and residual wage inequality increased substantially. This paper presents a two-region quality ladders model of technical change where …rms from the leading region innovate in all sectors of the economy, while the lagging region progressively catches up as its …rms enter global innovation races in a larger number of sectors. As the innovation gap closes, the advanced country experiences …ercer foreign technological competition which forces its …rms to innovate more. Faster technical change then increases the skill premium and residual

This paper presents a model in which technological change increases the share of reproducible factors at the expense of nonreproducible ones. When reproducible factors are abundant, firms have incentives to adopt technologies that are intensive in such resources, and this increases the incentives to invest more in them. This feed-back process may generate growth or also stagnation: when reproducible factors are not abundant, firms do not have incentives to adopt technologies intensive in those resources and technological change does not take place. The paper also analyzes how biased technological change affects interpersonal distribution of income: nonre-producible factors are more equally distributed than reproducible ones, thus biased technological change increases inequality. ∗Special thanks are due to Victor Rios Rull. Comments from Carlos Bethencourt and participants in the European Economic Association Conference and IGIER workshops were very helpful. This

ABSTRACT This study examines how digital transformation (DT) affects firms' internal incentive structures in China from 2010 to 2019. Unlike prior research, we assess firms' DT progress through participation in a nationwide DT certification program. Using a variety of empirical methods, we analyze how DT affects the compensation of both executives and workers, controlling for labor productivity, financial performance, and other firm characteristics. Our results indicate that: (i) DT significantly increases average worker compensation; (ii) this increase stems from compositional shifts toward hiring more skilled workers and creating additional non‐routine jobs; (iii) contrary to skill‐biased or routine‐biased technological change predictions, DT raises worker compensation without uniformly reducing low‐wage jobs in absolute terms; (iv) DT realigns incentive structures by linking corporate growth to executives' future compensation rather than current pay; and (v) DT reduces both the absolute and relative compensation gaps between executives and workers.

Technical change plays a crucial role in improving environmental quality, while the influence of demand-side factors remains insufficiently examined. To clarify the pull effect of consumer demand on the transition to clean technology, this study develops a model of directed technical change incorporating quality innovation in consumer goods. The analysis shows that the relative prices and market sizes of clean consumer goods drive the transition to clean technology, generating a direct demand-induced pull for clean innovations. Income inequality determines the market size of clean relative to dirty goods, thereby shaping innovation incentives and influencing the effectiveness of environmental policies. By integrating learning-by-doing and demand-induced innovation for dirty and clean technologies, respectively, the model captures the path dependence of technological progress and explains the dynamic ‘U-shaped’ evolution of environmental quality under environmental policy intervention. These findings provide theoretical insight into how consumer heterogeneity and income distribution affect the direction of innovation and the long-term transition toward cleaner technologies.

Abstract Many developed countries have faced labor shortages recently, intensifying competition for highly skilled immigrants. This study examines immigrant engineers and the evolving role of innovation played by those from developed and developing countries. Using original panel data for Japanese firms across ten industries between 1970 and 2019, the study employs a knowledge production function model to analyze immigrant engineers’ changing impact on innovation. The analysis reveals that in the 1970s, immigrant engineers had no significant effect on innovation output; however, after the 1980s onward, their influence became significant. Since the 2000s, the impact of engineers from developing countries has surpassed that of those from developed countries. The results show that while individuals from the USA had a positive and significant effect in the 1980s and 1990s, those from China and India have had significantly positive effects since the 2000s and 2010s, respectively. These findings suggest that for developed countries such as Japan, the origins of immigrant engineers, who significantly contribute to innovation, are shifting from developed to developing countries.

Esta investigación analiza el impacto de la inversión en Investigación y Desarrollo (I+D) y del número de investigadores en el crecimiento económico de algunas de las economías de la Organización para la Cooperación y el Desarrollo Económicos (OCDE), para el periodo 1996-2016. Se realiza un análisis de causalidad en el sentido de Granger y se estima un modelo de datos panel. Los datos son obtenidos de Banco Mundial. Se encuentra evidencia empírica de una causalidad bidireccional entre la I+D y el PIB per cápita, pero predominantemente I+D causa PIB. También se encuentra una causalidad bidireccional entre el número de investigadores y el PIB per cápita, pero predominantemente el PIB causa el número de investigadores. Mientras que el modelo de panel dinámico MGM-sistema en una etapa muestra que el crecimiento económico es afectado positivamente por la inversión en I+D y el número de investigadores. Este trabajo se distingue de otros en los siguientes aspectos: 1) considera una muestra de 25 países de OCDE en el periodo 1996-2016; 2) tiene una mayor disponibilidad de datos, y 3) se realiza un análisis de datos panel dinámico que permite utilizar una mayor cantidad de países, variables y períodos.

Mathematics preparation is a key determinant of success in business and other quantitatively oriented majors, yet many students enter college underprepared and are placed into remedial coursework. This paper examines how mathematics placement shapes academic trajectories. Using administrative data from a mid-sized public university and a fuzzy regression discontinuity design, we estimate the causal impact of placement just below versus above key thresholds. Students placed into remediation are significantly less likely to enroll in gateway math courses, pursue quantitatively intensive majors, or graduate within six years. These differences emerge among students with nearly identical placement scores, suggesting that placement itself—through added time, cost, or discouragement—alters behavior. We find little evidence that remediation improves course success, indicating limited academic benefit relative to its costs. Traditional prerequisite-based remediation may therefore act as a barrier rather than a support. We conclude by discussing alternatives appropriate to business schools, including co-requisite models and integrated quantitative instruction, that may better support progression and completion.

We study the impact on the field of study and academic outcomes of a personalized online mentoring program connecting high school students with university students in quantitative fields. Our RCT shows that the likelihood of choosing the field of the mentor increases by 14 to 22 p.p. – a 25% to 45% increase from the baseline. The program shifts preferences towards STEM/Economics, enhancing prospective wages by 3.1-3.7%. Administrative data show that the intervention does not negatively impact performance, even though treated students enroll in more competitive fields. These findings underscore the potential to guide undecided students toward more beneficial educational choices.

This paper examines how public four-year institutions adjust their program offerings in different fields in response to changes in state appropriations for higher education. Using a shift-share identification approach that interacts state-level fluctuations in funding for higher education with baseline institutional reliance on appropriations, I show that public institutions offer fewer programs in the natural sciences, social sciences, and education, but more programs in other professional and vocational fields, following a decrease in state funding. Program cuts in the sciences are more pronounced at research-intensive universities, while social science majors are more affected at non-research institutions; program cuts in education are observed at both types of schools. Program offerings are not responsive to changes in state higher education budgets at private universities, which provides further evidence that the results cannot be fully explained by shocks to aggregate economic conditions. • I construct a measure of the number of programs by field offered by an institution. • Public universities offer fewer programs when state appropriations are lower. • The social sciences and education are most strongly impacted by state budget cuts. • State appropriations do not impact program offerings at private universities.

This chapter examines key aspects of tertiary education, focusing on access, financing, and cost-effectiveness. The expansion of higher education has increased human capital and economic productivity, yet disparities persist in access due to socioeconomic barriers. Financing mechanisms vary across countries, with some relying on public subsidies and others on private contributions. The analysis highlights that income-contingent loans can enhance equity while maintaining fiscal sustainability. Additionally, the returns to higher education remain positive, although they differ by field of study and institution. The chapter underscores the need for policies that balance affordability, efficiency, and accessibility to maximize the benefits of tertiary education.

In 2007, China launched a nationwide Free Teacher Education (FTE) program, offering conditional tuition waivers and stipends to teaching-track students at six elite teachers' colleges. This study examines the impacts of three phases of the FTE program on college admission outcomes in Ningxia, a rural and racially diverse province. Using administrative data on College Entrance Exams (CEEs) and admission records between 2003 and 2018, we find that the program's initial phase motivated applications from disadvantaged students but did not significantly improve the academic qualification of admitted students. In contrast, later phases attracted high-performing applicants, especially in STEM disciplines, raising the admission standards by up to 7 percentiles. However, the increased competition unintentionally deterred financially constrained students, reducing their representation among the program's admitted cohorts. We supplement our analysis with a randomized survey experiment at a large high school in Ningxia to uncover the behavioral mechanisms underlying our findings.

This study examines the long-term effect of a pandemic on a crucial human capital decision, namely college major choice. Using China's 2008-2016 major-level National College Entrance Examination (Gaokao) entry grades, we find that the 2003 severe acute respiratory syndrome (SARS) had a substantial deterrent effect on the choice of majoring in medicine among high school graduates who experienced the pandemic in their childhood. In provinces with larger intensities of SARS impact, medical majors become less popular as the average Gaokao grades of enrolled students decline. Further evidence from a nationally representative survey shows that the intensity of the SARS impact significantly decreases children's aspirations to pursue medical occupations, but does not affect their parents' expectations for their children to enter the medical profession. Our discussion on the effect mechanism suggests that the adverse influence of SARS on the popularity of medical majors likely originates from students' childhood experiences.

This article examines effects of the US Immigration Act of 1990 on STEM (science, technology, engineering, and mathematics) education and labor market outcomes for native-born Americans. The Act increased the inflow and stock of foreign STEM workers in the United States, potentially altering the relative desirability of STEM fields for natives. The authors examine effects of the policy on STEM degree completion, STEM occupational choice, and employment rates separately for black and white men and women. The novel identification strategy measures exposure to foreign STEM workers of age 18 native cohorts immediately before and after the policy change via geographic dispersion of foreign-born STEM workers in 1980, which predicts subsequent foreign STEM flows. The Act affected natives in three ways: 1) black male students moved away from STEM majors; 2) white male STEM graduates moved away from STEM occupations; and 3) white female STEM graduates moved out of the workforce.

Abstract Economists find supply and demand to be a very useful way to describe markets, including labor markets. Yes, supply and demand models sometimes fail, but such cases are notable precisely because such models usually work so well. In fact, arguably no model in social science works as well; it is the crown jewel of economic theory. In a supply and demand based labor market, buyers and sellers mostly take prices as given, and assume that they can’t change prices much. Given this assumption, they try to achieve their goals by varying how much labor they buy or sell. Note that supply and demand doesn’t require that everyone know everything, or that they always do exactly what is best for them. It is actually a pretty robust and useful model of human behavior. True, workers often acquire very specific job skills, after which there may be too few sellers or buyers of each specific skill to make for a competitive market. At that point people may reasonably believe that their behaviors can change relevant prices. But for each specific skill there is usually a large pool of workers who are similarly able to learn that skill, and another large pool, this time of employers, with skills they’d like this same pool of workers to learn in order to do their jobs. There is thus a pre-skill labor market with pools of similarly-able-to-learn workers, and with employers who have similar-tasks-to-learn. If these pools are large, and if they do not coordinate to limit the wages they accept, then supply and demand analysis will apply well to this pre-skill market. Thus while it may be hard to predict the specific wages that workers will earn after they learn a specific skill, we can more confidently predict that, in the pre-skill labor market, similar workers will reasonably expect to earn a similar net compensation after they train. Also, employers trying to attract similar workers should expect to pay a similar net compensation. (Of course “wages” include not just cash, but other forms of compensation such as status markers, connections, and resources including information access and computing power.) Consider how such pre-skill labor markets change when we introduce ems built from cheap signal-processing hardware, who are able to substitute in most jobs for ordinary human workers after they’ve acquired relevant skills.

The study of economic growth is concerned with long-run changes, and therefore, historical data should be especially influential in informing the development of new theories. In this review, we draw on the recent literature to highlight areas in which study of history has played a particularly prominent role in improving our understanding of growth dynamics. Research at the intersection of historical data, theory, and empirics has the potential to reframe how we think about economic growth in much the same way that historical perspectives helped to shape the first generation of endogenous growth theories.

Throughout much of the United States, career and technical education (CTE) is offered as one or more elective sequences within comprehensive high schools. We know very little about why students select into CTE in such systems, or more generally, whether the labor market affects students choice of coursework in high school. We test whether changes in the local labor market align with and affect course enrollments in three in-demand CTE career clusters in Tennessee: advanced manufacturing, information technology, and health science. We find evidence of roughly proportionate alignment between changes in advanced manufacturing CTE course-taking and changes in local manufacturing employment in recent years. Instrumental variable estimates, however, suggest that this is not necessarily due to students responding to local employment dynamics. We do not detect evidence of alignment, causal or otherwise, for health science or information technology CTE.

STEM fields are viewed as being important for global economic development, as well as for the well-being of society. Many factors, including knowledge of future pay and other occupational insights, influence university major selection. This paper reports the findings from an empirical study of diploma, undergraduate, and postgraduate on the relationship between gender equality and university support with students’ views on STEM careers, as well as their persistence and attrition in STEM majors. The findings from PLS-SEM analysis shows that gender equality did positively affect students’ views on STEM careers and students’ persistence in STEM majors. It was also found that gender equality did not affect students’ attrition. In contrast, the university support did not positively affect students’ views on STEM careers and students’ attrition in STEM majors. However, university support was found to positively affect students’ persistence in STEM majors. The implications of the findings are that the university can channel its support systems in nurturing the students’ skills and knowledge by providing physical and psychosocial support for the students to persist in STEM majors. Hence, encouraging more students to opt for STEM majors is necessary to enhance the global economy so that it can contribute to the well-being not just of the STEM graduates, but the society and nation as well.

Številni izzivi, na katere naletimo pri razvoju programskih rešitev, nas silijo, da neprestano iščemo nove pristope in prakse, s katerimi bi IT projekte realizirali boljše, hitreje in predvsem z nižjimi stroški. Želja po hitri in cenovno ugodni realizaciji IT projektov, višji stopnji njihove kakovosti ter nenazadnje v zadnjem času že kroničnem pomanjkanju usposobljenih IT strokovnjakov, so samo nekateri izmed izzivov, s katerimi se srečujemo v programskem inženirstvu. Pri naslavljanju omenjenih izzivov si v zadnjem času veliko obetamo od vpeljave umetne inteligence v proces razvoja programske opreme. Možnosti se kažejo predvsem v vpeljavi z naprednimi metodami umetne inteligence podprtih orodij, ki razvojno skupino razvijalcev aktivno podpirajo pri razvoju. Z umetno inteligenco podprta orodja odpirajo vrata odmiku od avtomatizacije ponavljajočih se trivialnih opravil in obljubljajo možnost avtomatizacije intelektualno zahtevnejših in kompleksnih opravil, kar bi občutno razbremenilo razvijalce informacijskih rešitev.

Because of the fast development in technologies, our lives have been constantly changing and the impact of changes on labor markets is essential for economic growth and inequality. This dissertation explores the relationship between economic growth, inequality, and the challenges in adapting to changes in the labor market. The study focuses on understanding the difficulties workers face when their skills need to change or when they need to switch to different industries. It examines how these adjustments affect productivity, wages, and the perception of technological changes that favor certain skills. Additionally, it explores the obstacles workers encounter when trying to move between different industries due to factors like high costs or specific skills needed in particular industries. The research also discusses the role of labor market policies in managing unemployment and inflation. It considers the trade-offs involved and the challenges of implementing effective policies. Overall, the dissertation highlights the importance of understanding and addressing these challenges to achieve both economic growth and a fairer society.

Abstract In an era of rapid technological advancements and macroeconomic shifts, worker reallocation is necessary; yet responses to labor market shocks remain sluggish, making it crucial to identify bottlenecks in occupational transitions to understand labor market dynamics and improve mobility. In this study, we analyze French occupational data to uncover patterns of worker mobility and pinpoint specific occupations that act as bottlenecks, impeding rapid reallocation. We introduce two metrics, transferability and accessibility, to quantify the diversity of occupational transitions and find that bottlenecks can be explained by a condensation effect of occupations with high accessibility but low transferability. Transferability measures the variety of transitions from one occupation to another, while accessibility assesses the variety of transitions into an occupation. We provide a comprehensive framework for analyzing occupational complexity and mobility patterns, offering insights into potential barriers and pathways for efficient retraining programs. We argue that our approach can inform policymakers and stakeholders aiming to enhance labor market efficiency and support workforce adaptability.

This paper studies how gradualism affects the welfare gains from trade, technology, and reforms. When workers face adjustment frictions, gradual shocks create less adverse distributional effects in the short run. We show that there are welfare gains from inducing a more gradual transition via temporary taxes on trade and technology and provide formulas for the optimal path for taxes. Our formulas account for the possibility that reallocation effort responds to policy and for the existence of income taxes and assistance programs. Using these formulas, we compute the optimal temporary taxes needed to mitigate the distributional consequences of rising import competition from China and the deployment of automation technologies substituting for routine jobs. Our formulas can also be used to compute the optimal timing of economic reforms or trade liberalizations. We study Colombia’s trade liberalization in 1990 and conclude that optimal policy called for a more gradual reform. *Restrepo thanks the National Science Foundation for its support under award No. 2049427. We thank David Autor and Marcela Eslava for sharing their data and providing feedback on this project. We also thank Joao Guerreiro, Chad Jones, Michael Peters, and Nathan Zorzi for providing valuable comments and Nicolas Werquin for discussing our paper. Technological progress, trade, and economic reforms can generate periods of adjustment during which some workers fall behind, lose their jobs, experience wage declines, and see their livelihoods disrupted.1 Even if technology and trade are positive developments in the long run, dealing with these short-run disruptions remains an important policy concern, especially in the wake of rapid changes in the economy.2 Existing evidence points to large disruptions. Autor et al. (2014) document that an average US worker in an industry exposed to Chinese import competition experienced a cumulative income loss equal to half their annual earnings in 1990 over the 1992–2007 period relative to unexposed workers. Cortes (2016) shows that US workers who in 1985 held routine jobs—those that can be more easily automated—experienced a subsequent decline in wages of 16% by 2007 relative to similar workers in other occupations. How should policy respond during these adjustment periods? Do short-run disruptions imply that more gradual advances in technology and trade are preferable? This paper shows that short-run disruptions create potential gains from gradualism and justify temporary taxes on new technologies and trade or embracing gradual reforms. Our main contribution is to provide formulas for the optimal path for taxes on new technologies and trade that capture the gains from gradualism. We evaluate these formulas in a calibrated version of our model that matches the empirical estimates of Autor et al. (2014) for trade and Cortes (2016) for the automation of routine jobs. Our formulas call for temporary taxes on trade and automation technologies of 10%, phased out over time. We also use our formulas to study Colombia’s trade liberalization in 1990 and show that optimal policy called for a more gradual reform. We derive these formulas in a model where workers are displaced by technology or trade. Ex-ante identical workers are allocated across islands à la Lucas and Prescott (1974). Some islands represent jobs automated by new technologies (e.g., welding or data-entry clerks) or segments of industries disrupted by international trade (e.g., low-cost apparel or household electronics). At time t0, a new technology arrives, capable of replacing workers in these For evidence in the context of trade, see Autor and Dorn (2013); Autor et al. (2014). For evidence in the context of automation technologies, see Cortes (2016); Adão et al. (2021); Acemoglu and Restrepo (2020, 2022). Finally, see Goldberg and Pavcnik (2005) for evidence of how dismantling trade protection reduces the relative wages of workers in exposed industries. In the US, industrial robots installations and imports from China tripled in a few years (see Autor et al., 2013; Acemoglu and Restrepo, 2020, respectively), and the share of e-commerce in retail went from 0.6% to 10% from 1999 to 2019 (see US Census, 2022). As Erik Brynjolfsson and Andrew McAfee put it in The Second Machine Age, “People are falling behind because technology is advancing so fast and our skills and organizations aren’t keeping up” (Brynjolfsson and McAfee, 2014). Managing short-run disruptions is also a key policy concern when it comes to policy reforms (see, for example, Rodrik, 1995).

Abstract The German economy is currently not only in a tense cyclical situation but is also undergoing a profound structural transformation – driven by technological innovations, decarbonisation, (de-)globalisation, and demographic change. These developments have far-reaching effects on the labour market. The labour market is characterised by increasing polarisation, where skill shortages and rising unemployment can occur simultaneously. The deliberate retention of workers by companies despite declining capacity utilisation (labour hoarding) further delays the necessary reallocation. This article analyses these structural trends in the context of the accelerating structural transformation. Finally, economic and labour market policy measures to address these challenges are discussed.

ABSTRACT This paper studies the impact of trade openness on welfare through alterations in workers' skill acquisition. Guided by empirical evidence, we integrate endogenous choices of learning investments into a multisector Eaton–Kortum model. Our model reveals that trade openness influences skill acquisition by two channels: (1) reallocating labor between sectors with varying skill intensities and on‐the‐job learning opportunities and (2) allowing producers in each country to source varieties from more cost‐effective suppliers in other countries, thus reducing costs of material inputs for learning. Our quantification indicates that the gains in skill acquisition account for 5% of the total gains from trade.

In this study, we construct heterogeneous agent general equilibrium models to investigate the relative importance of labor endowment in driving structural transformation. We aim to explore the following question: beyond the demand-side and supply-side structural transformation driving forces extensively studied in the existing literature, does labor, as a crucial endowment, play a pivotal role in facilitating structural transformation and the energy economy? In contrast to the prevalent partial equilibrium analyses, our study employs a general equilibrium framework to conduct a policy evaluation of college expansion, a significant policy that has altered the labor endowment structure in China. Our approach begins with developing a multi-sector model that integrates a nested CES production function and incorporates workers with different skill levels to assess the macroeconomic impact of college expansion on structural transformation. We calibrate the base model to reflect labor allocations across sectors and skill levels using the simulated method of moments (SMM), ensuring that the model-generated data align closely with actual labor allocation data. Utilizing this calibrated model, we perform counterfactual experiments to assess the impact and relative importance of the college expansion policy. Our counterfactual analysis demonstrates that the policy has resulted in an average decrease of 7.7% in labor allocation in the agricultural sector, alongside an average increase of 8.9% in the industry sector and 28.7% in the services sector. These results highlight the significant, yet often overlooked, contribution of labor in endowment-driven structural transformation. Furthermore, we extend the base model by constructing an industry-level heterogeneous agent general equilibrium model, enabling us to pinpoint which industries have developed as a result of the college expansion policy and recalibrate it at the industry level. This approach allows us to analyze the impact of changes in labor endowment on the energy economy. Counterfactual experiments conducted show that the college expansion policy has prompted a labor shift from industries with low energy efficiency and high pollution to high-end services. This macroeconomic pattern of structural transformation suggests that the college expansion policy has facilitated a transition toward a low-carbon economy by reducing dependency on high energy-consuming industries and promoting high-end services.

This paper examines how patent strength affects R & D competition in a model where two symmetric firms simultaneously choose their research projects. A stronger patent system induces not only greater R & D investment but also more research duplication. It is shown that there exists a range of patent strength in which increasing patent strength discourages innovation. Moreover, as firms behave more collusively in the post-innovation market, the strongest patent system becomes less likely to maximize the probability of innovation.

Wesley M. Cohen, Daniel A. Levinthal, Absorptive Capacity: A New Perspective on Learning and Innovation, Administrative Science Quarterly, Vol. 35, No. 1, Special Issue: Technology, Organizations, and Innovation (Mar., 1990), pp. 128-152

This paper examines whether the Solow growth model is consistent with the international variation in the standard of living. It shows that an augmented Solow model that includes accumulation of human as well as physical capital provides an excellent description of the cross-country data. The paper also examines the implications of the Solow model for convergence in standards of living, that is, for whether poor countries tend to grow faster than rich countries. The evidence indicates that, holding population growth and capital accumulation constant, countries converge at about the rate the augmented Solow model predicts.

This paper develops a model based on Schumpeter's process of creative destruction. It departs from existing models of endogenous growth in emphasizing obsolescence of old technologies induced by the accumulation of knowledge and the resulting process or industrial innovations. This has both positive and normative implications for growth. In positive terms, the prospect of a high level of research in the future can deter research today by threatening the fruits of that research with rapid obsolescence. In normative terms, obsolescence creates a negative externality from innovations, and hence a tendency for laissez-faire economies to generate too many innovations, i.e too much growth. This business-stealing effect is partly compensated by the fact that innovations tend to be too small under laissez-faire. The model possesses a unique balanced growth equilibrium in which the log of GNP follows a random walk with drift. The size of the drift is the average growth rate of the economy and it is endogenous to the model ; in particular it depends on the size and likelihood of innovations resulting from research and also on the degree of market power available to an innovator.

It is by now incontrovertible that increases in per capita income cannot be explained simply by increases in the capital-labor ratio. Though doubtless no economist would ever have denied the role of technological change in economic growth, its overwhelming importance relative to capital formation has perhaps only been fully realized with the important empirical studies of Abramovitz [1] and Solow [l 1]. These results do not directly contradict the neo-classical view of the production function as an expression of technological knowledge. All that has to be added is the obvious fact that knowledge is growing in time. Nevertheless a view of economic growth that depends so heavily on an exogenous variable, let alone one so difficult to measure as the quantity of knowledge, is hardly intellectually satisfactory. From a quantitative, empirical point of view, we are left with time as an explanatory variable. Now trend projections, however necessary they may be in practice, are basically a confession of ignorance, and, what is worse from a practical viewpoint, are not policy variables.KeywordsLabor ForceProduction FunctionWage RateTechnical ChangeSerial NumberThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The procedures and the nature of "technologies" are suggested to be broadly similar to those which characterize "science". In particular, there appear to be "technological paradigms" (or research programmes) performing a similar role to "scientific paradigms" (or research programmes). The model tries to account for both continuous changes and discontinuities in technological innovation. Continuous changes are often related to progress along a technological trajectory defined by a technological paradigm, while discontinuities are associated with the emergence of a new paradigm. One-directional explanations of the innovative process, and in particular those assuming "the market" as the prime mover, are inadequate to explain the emergence of new technological paradigms. The origin of the latter stems from the interplay between scientific advances, economic factors, institutional variables, and unsolved difficulties on established technological paths. The model tries to establish a sufficiently general framework which accounts for all these factors and to define the process of selection of new technological paradigms among a greater set of notionally possible ones. The history of a technology is contextual to the history of the industrial structures associated with that technology. The emergence of a new paradigm is often related to new "schumpeterian" companies, while its establishment often shows also a process of oligopolistic stabilization. © 1982.

Giovanni Dosi is recognized as one of the world’s leading scholars in industrial economics and corporate change. This volume contains a selection of his most important work and provides an excellent overview of the contribution he has made to the economics of innovation and technical change.

We offer a unified analysis of the growth of low-skill service occupations between 1980 and 2005 and the concurrent polarization of US employment and wages. We hypothesize that polarization stems from the interaction between consumer preferences, which favor variety over specialization, and the falling cost of automating routine, codifiable job tasks. Applying a spatial equilibrium model, we corroborate four implications of this hypothesis. Local labor markets that specialized in routine tasks differentially adopted information technology, reallocated low-skill labor into service occupations (employment polarization), experienced earnings growth at the tails of the distribution (wage polarization), and received inflows of skilled labor. (JEL J24, J31, R23)

Household-level panel data from a nationally representative sample of rural Indian households describing the adoption and profitability of high-yielding seed varieties (HYVs) associated with the Green Revolution are used to test the implications of a model incorporating learning by doing and learning spillovers. The estimates indicate that imperfect knowledge about the management of the new seeds was a significant barrier to adoption; this barrier diminished as farmer experience with the new technologies increased; own experience and neighbors' experience with HYVs significantly increased HYV profitability; and farmers do not fully incorporate the village returns to learning in making adoption decisions. Copyright 1995 by University of Chicago Press.

This paper documents the pervasiveness of job polarization in 16 Western European countries over the period 1993–2010. It then develops and estimates a framework to explain job polarization using routine-biased technological change and offshoring. This model can explain much of both total job polarization and the split into within-industry and between-industry components. (JEL J21, J23, J24, M55, O33)

This paper investigates the factors determining how rapidly the use of a new technique spreads from one firm to another. A simple model is presented to help explain differences among innovations in the rate of imitation. Deterministic and stochastic versions of this model are tested against data showing how rapidly firms in four industries came to use twelve important innovations. The empirical results seem quite consistent with both versions of the model.

Between 1973 and 1978, the Indonesian government engaged in one of the largest school construction programs on record. Combining differences across regions in the number of schools constructed with differences across cohorts induced by the timing of the program suggests that each primary school constructed per 1,000 children led to an average increase of 0.12 to 0.19 years of education, as well as a 1.5 to 2.7 percent increase in wages. This implies estimates of economic returns to education ranging from 6.8 to 10.6 percent. (JEL I2, J31, O15, O22)

This paper reviews research that uses longitudinal microdata to document productivity movements and to examine factors behind productivity growth. The research explores the dispersion of productivity across firms and establishments, the persistence of productivity differentials, the consequences of entry and exit, and the contribution of resource reallocation across firms to aggregate productivity growth. The research also reveals important factors correlated with productivity growth, such as managerial ability, technology use, human capital, and regulation. The more advanced literature in the field has begun to address the more difficult questions of the causality between these factors and productivity growth.

The nonrivalry of technology, as modeled in the endogenous growth Uterature, implies that high population spurs technological change. This paper constructs and empirically tests a model of long-run world population growth combining this implication with the Malthusian assumption that technology limits population. The model predicts that over most of history, the growth rate of population will be proportional to its level. Empirical tests support this prediction and show that historically, among societies with no possibility for technological contact, those with larger initial populations have had faster technological change and population growth.

Abstract The labor market increasingly rewards social skills. Between 1980 and 2012, jobs requiring high levels of social interaction grew by nearly 12 percentage points as a share of the U.S. labor force. Math-intensive but less social jobs—including many STEM occupations—shrank by 3.3 percentage points over the same period. Employment and wage growth were particularly strong for jobs requiring high levels of both math skill and social skills. To understand these patterns, I develop a model of team production where workers “trade tasks” to exploit their comparative advantage. In the model, social skills reduce coordination costs, allowing workers to specialize and work together more efficiently. The model generates predictions about sorting and the relative returns to skill across occupations, which I investigate using data from the NLSY79 and the NLSY97. Using a comparable set of skill measures and covariates across survey waves, I find that the labor market return to social skills was much greater in the 2000s than in the mid-1980s and 1990s.

The author, using data on 694 U.S. manufacturing establishments from a 1992 survey, examines the incidence of innovative work practices (teams, job rotation, quality circles, and Total Quality Management) and investigates what variables, including human resource practices, are associated with the adoption of these practices. He finds that about 35% of private sector establishments with 50 or more employees made substantial use of flexible work organization in 1992. Some factors associated with an establishment's adoption of these practices are being in an internationally competitive product market, having a technology that requires high levels of skill, following a “high road” strategy that emphasizes variety, service, and quality rather than low cost, and using such human resource practices as high levels of training and innovative pay systems.

We examine the impact of Chinese import competition on broad measures of technical change—patenting, IT, and TFP—using new panel data across twelve European countries from 1996 to 2007. In particular, we establish that the absolute volume of innovation increases within the firms most affected by Chinese imports in their output markets. We correct for endogeneity using the removal of product-specific quotas following China's entry into the World Trade Organization in 2001. Chinese import competition led to increased technical change within firms and reallocated employment between firms towards more technologically advanced firms. These within and between effects were about equal in magnitude, and account for 14% of European technology upgrading over 2000–7 (and even more when we allow for offshoring to China). Rising Chinese import competition also led to falls in employment and the share of unskilled workers. In contrast to low-wage nations like China, developed countries had no significant effect on innovation.

According to endogenous growth theory, permanent changes in certain policy variables have permanent effects on the rate of economic growth. Empirically, however, U. S. growth rates exhibit no large persistent changes. Therefore, the determinants of long-run growth highlighted by a specific growth model must similarly exhibit no large persistent changes, or the persistent movement in these variables must be offsetting. Otherwise, the growth model is inconsistent with time series evidence. This paper argues that many AK-style models and R&D-based models of endogenous growth are rejected by this criterion. The rejection of the R&D-based models is particularly strong.

The dominant explanation for the spread of technological innovations emphasizes processes of influence and information flow. Firms which are closely connected to pre-existing users of an innovation learn about it and adopt it early on. Firms at the periphery of communication networks are slower to adopt. This paper develops an alternative model which emphasizes the role of know-how and organizational learning as potential barriers to adoption of innovations. Firms delay in-house adoption of complex technology until they obtain sufficient technical know-how to implement and operate it successfully. In response to knowledge barriers, new institutions come into existence which progressively lower those barriers, and make it easier for firms to adopt and use the technology without extensive in-house expertise. Service bureaus, consultants, and simplification of the technology are examples. As knowledge barriers are lowered, diffusion speeds up, and one observes a transition from an early pattern in which the new technology is typically obtained as a service to a later pattern of in-house provision of the technology. Thus the diffusion of technology is reconceptualized in terms of organizational learning, skill development, and knowledge barriers. The utility of this approach is shown through an empirical study of the diffusion of business computing in the United States, reporting survey and ethnographic data on the spread of business computing, on the learning processes and skills required, and on the changing institutional practices that facilitated diffusion.

This paper provides structural estimates of a dynamic model of schooling, work, and occupational choice decisions based on eleven years of observations on a sample of young men from the 1979 youth cohort of the National Longitudinal Surveys of Labor Market Experience (NLSY). The authors find that a suitably extended human capital investment model can in fact do an excellent job of fitting observed data on school attendance, work, occupational choices, and wages in the NLSY data on young men and also produces reasonable forecasts of future work decisions and wage patterns. Copyright 1997 by the University of Chicago.

The Dynamo and the Computer: An Historical Perspective on the Modern Productivity Paradox Author(s): Paul A. David Source: The American Economic Review, Vol. 80, No. 2, Papers and Proceedings of the Hundred and Second Annual Meeting of the American Economic Association (May, 1990), pp. 355-361 Published by: American Economic Association Stable URL: http://www.jstor.org/stable/2006600 Accessed: 08/12/2010 01:40

This paper develops a simple general-equilibrium model of product cycle trade. There are two countries, innovating North and noninnovating South. Innovation consists of the development of new products. These can be produced at first only in North, but eventually the technology of production becomes available to South. This technological lag gives rise to trade, with North exporting new products and importing old products. Higher Northern per capita income depends on the quasi rents from the Northern monopoly of new products, so that North must continually innovate not only to maintain its relative position but even to maintain its real income in absolute terms.

The authors propose a theory of economic development in which technology adoption and barriers to such adoptions are the focus. The size of these barriers differs across countries and time. The larger these barriers, the greater the investment a firm must make to adopt a more advanced technology. The model is calibrated to the U.S. balanced growth observations and the postwar Japanese development miracle. For this calibrated structure, the authors find that the disparity in technology adoption barriers needed to account for the huge observed income disparity across countries is not implausibly large. Copyright 1994 by University of Chicago Press.

Economists emphasize the link between market returns to education and investments in schooling. Though many studies estimate these returns with earnings data, it is the perceived returns that affect schooling decisions, and these perceptions may be inaccurate. Using survey data for eighth-grade boys in the Dominican Republic, we find that the perceived returns to secondary school are extremely low, despite high measured returns. Students at randomly selected schools given information on the higher measured returns completed on average 0.20–0.35 more years of school over the next four years than those who were not.

Abstract We investigate the determinants of regional development using a newly constructed database of 1,569 subnational regions from 110 countries covering 74% of the world’s surface and 97% of its GDP. We combine the cross-regional analysis of geographic, institutional, cultural, and human capital determinants of regional development with an examination of productivity in several thousand establishments located in these regions. To organize the discussion, we present a new model of regional development that introduces into a standard migration framework elements of both the Lucas (1978) model of the allocation of talent between entrepreneurship and work, and the Lucas (1988) model of human capital externalities. The evidence points to the paramount importance of human capital in accounting for regional differences in development, but also suggests from model estimation and calibration that entrepreneurial inputs and possibly human capital externalities help understand the data.

Neoclassical theory suggests that when an industry is shaken by radical technological change, incumbent firms will be replaced by entrants because entrants have greater strategic incentives to invest in radical innovation. Organizational theory suggests that incumbent firms fail in the face of radical innovation because they fall prey to inertia and complacency. I show that if organizational effects are significant, tests of neoclassical theory in isolation will yield spurious or noisy results. Using data derived from a detailed field study of the photolithographic alignment equipment industry, I show that as neoclassical theory predicts, established firms invested more than entrants in incremental innovation, but that in agreement with organizational theory, the research efforts of incumbents seeking to exploit radical innovation were significantly less productive than those of entrants.

Results from the Displaced Worker Surveys show that the wage cost of switching industries following displacement is strongly correlated with predisplacement measures of both work experience and tenure. Workers apparently receive compensation for some skills that are neither completely general nor firm-specific but rather specific to their industry or line of work. Further, among displaced workers who find new jobs in their predisplacement industry, postdisplacement returns to predisplacement job tenure resemble cross-section estimates of the returns to current seniority. This suggests that firm-specific factors may contribute little to the observed slope of wage-tenure profiles. Copyright 1995 by University of Chicago Press.

We develop an endogenous growth model in which clean and dirty technologies compete in production. Research can be directed to either technology. If dirty technologies are more advanced, the transition to clean technology can be difficult. Carbon taxes and research subsidies may encourage production and innovation in clean technologies, though the transition will typically be slow. We estimate the model using microdata from the US energy sector. We then characterize the optimal policy path that heavily relies on both subsidies and taxes. Finally, we evaluate various alternative policies. Relying only on carbon taxes or delaying intervention has significant welfare costs.

Large earnings and ability differences exist across majors. This paper seeks to estimate the monetary returns to particular majors as well as find the causes of the ability sorting across majors. In order to accomplish this, I estimate a dynamic model of college and major choice. Even after controlling for selection, large earnings premiums exist for certain majors. Differences in monetary returns explain little of the ability sorting across majors; virtually all ability sorting is because of preferences for particular majors in college and the workplace, with the former being larger than the latter. © 2003 Elsevier B.V. All rights reserved.

A large, empirical literature reports estimates of the rate of return to R&D ranging from 30 percent to over 100 percent, supporting the notion that there is too little private investment in research. This conclusion is challenged by the new growth theory. We derive analytically the relationship between the social rate of return to R&D and the coefficient estimates of the empirical literature. We show that these estimates represent a lower bound on the true social rate of return. Using a conservative estimate of the rate of return to R&D of about 30 percent, optimal R&D investment is at least four times larger than actual investment.

In recent years, there has been a revival of concerns that automation and digitalisation might after all result in a jobless future. The debate has been fuelled by studies for the US and Europe arguing that a substantial share of jobs is at “risk of computerisation”. These studies follow an occupation-based approach proposed by Frey and Osborne (2013), i.e. they assume that whole occupations rather than single job-tasks are automated by technology. As we argue, this might lead to an overestimation of job automatibility, as occupations labelled as high-risk occupations often still contain a substantial share of tasks that are hard to automate. Our paper serves two purposes. Firstly, we estimate the job automatibility of jobs for 21 OECD countries based on a task-based approach. In contrast to other studies, we take into account the heterogeneity of workers’ tasks within occupations. Overall, we find that, on average across the 21 OECD countries, 9 % of jobs are automatable. The threat from technological advances thus seems much less pronounced compared to the occupation-based approach. We further find heterogeneities across OECD countries. For instance, while the share of automatable jobs is 6 % in Korea, the corresponding share is 12 % in Austria. Differences between countries may reflect general differences in workplace organisation, differences in previous investments into automation technologies as well as differences in the education of workers across countries.

We measure the extent to which skilled immigrants increase innovation in the United States. The 2003 National Survey of College Graduates shows that immigrants patent at double the native rate, due to their disproportionately holding science and engineering degrees. Using a 1940–2000 state panel, we show that a 1 percentage point increase in immigrant college graduates' population share increases patents per capita by 9–18 percent. Our instrument for the change in the skilled immigrant share is based on the 1940 distribution across states of immigrants from various source regions and the subsequent national increase in skilled immigration from these regions. (JEL J24, J61, O31, O33)

This paper focuses on the role of capabilities in economic development. In recent years, the quality and availability of data on different aspects of development have improved, and this provides new opportunities for investigating the reasons behind the large differences in economic development. Using factor analysis on data for 25 indicators and 115 countries between 1992 and 2004, we identify four different types of "capabilities": the development of the "innovation system", the quality of "governance", the character of the "political system" and the degree of "openness" of the economy. Innovation systems and governance are shown to be of particular importance for economic development.

Information problems and lack of collateral value should make R&D more susceptible to financing frictions than other investments, yet existing evidence on whether financing constraints limit R&D is decidedly mixed, particularly in the studies of non-U.S. firms. We study a large sample of European firms and also find little evidence of binding finance constraints when we estimate standard investment-cash flow regressions. However, we find strong evidence that the availability of finance matters for R&D once we directly control for: (i) firm efforts to smooth R&D with cash reserves and (ii) firm use of external equity finance. Our study provides a framework for evaluating financing constraints when firms rely extensively on external finance and endogenously manage buffer stocks of liquidity to keep investment smooth, and our findings show that controlling for this smoothing behavior is critical for uncovering the full effect of financing constraints. Our findings also indicate a major role for external equity in financing R&D, highlighting a causal channel through which stock market development and liberalization can promote economic growth by increasing firm-level innovative activity. © 2012 Elsevier B.V.

This paper studies the nature and consequences of competition in R&D and the relationship between this form of competition and competition in the product market, by focusing on comparisons of speed of research, number of independent research laboratories, and level of risk undertaken. Among the results: competition in the current product market reduces the level of innovation (relative to monopoly); competition in R&D increases the level of innovation, possibly beyond the socially optimal level. Under certain conditions, it pays a monopolist to preempt potential competitors, thereby enabling the monopoly to persist. Market equilibrium may entail excessively fast research with insufficient risk-taking.

This paper examines the impact of fiscal incentives on the level of R & D investment. An econometric model of R & D investment is estimated using a new panel of data on tax changes and R & D spending in nine OECD countries over a 19-year period (1979-1997). We find evidence that tax incentives are effective in increasing R & D intensity. This is true even after allowing for permanent country-specific characteristics, world macro shocks and other policy influences. We estimate that a 10% fall in the cost of R & D stimulates just over a 1% rise in the level of R & D in the short-run, and just under a 10% rise in R & D in the long-run. © 2002 Elsevier Science B.V. All rights reserved.

Many countries are pursuing innovation-led “smart” growth, which requires long-run strategic investments and public policies that aim to create and shape markets, rather than just “fixing” markets or systems. Market creation has characterized the kind of mission-oriented investments that led to putting a man on the moon and are currently galvanizing green innovation. Mission-oriented innovation has required public agencies to not only “de-risk” the private sector, but also to lead the direct creation of new technological opportunities and market landscapes. This paper considers four key issues that arise from a market-creating framework for policy: (1) decision-making on the direction of change; (2) the nature of (public and private) organizations that can welcome the underlying uncertainty and discovery process; (3) the evaluation of mission-oriented and market-creation policies; and (4) the ways in which both risks and rewards can be shared so that smart growth can also result in inclusive growth.

Studies of innovation have focused on the effects of patent laws on the number of innovations, but have ignored effects on the direction of technological change. This paper introduces a new dataset of close to fifteen thousand innovations at the Crystal Palace World's Fair in 1851 and at the Centennial Exhibition in 1876 to examine the effects of patent laws on the direction of innovation. The paper tests the following argument: if innovative activity is motivated by expected profits, and if the effectiveness of patent protection varies across industries, then innovation in countries without patent laws should focus on industries where alternative mechanisms to protect intellectual property are effective. Analyses of exhibition data for 12 countries in 1851 and 10 countries in 1876 indicate that inventors in countries without patent laws focused on a small set of industries where patents were less important, while innovation in countries with patent laws appears to be much more diversified. These findings suggest that patents help to determine the direction of technical change and that the adoption of patent laws in countries without such laws may alter existing patterns of comparative advantage across countries.

The authors study the problem of optimal taxation in three infinite-horizon, representative-agent endogenous growth models. The first model is a convex model in which physical and human capital are perfectly symmetric. The authors' second model incorporates elastic labor supply through a Lucas-style technology. Analysis of these two models points out the danger of assuming that government expenditures are exogenous. In their third model, the authors include government expenditures as a productive input in capital formation, showing that the limiting tax rate on capital is no longer zero. In numerical simulations, they find similar effects on growth and welfare in all three models. Copyright 1993 by University of Chicago Press.

A multicountry Schumpeterian growth model is constructed. Because of technology transfer, R&D-performing countries converge to parallel growth paths; other countries stagnate. A parameter change that would have raised a country's growth rate in standard Schumpeterian theory will permanently raise its productivity and per capita income relative to other countries and raise the world growth rate. Transitional dynamics are analyzed for each country and for the world economy. Steady-state income differences obey the same equation as in neoclassical theory, but since R&D is positively correlated with investment rates, capital accumulation accounts for less than estimated by neoclassical theory. (JEL E10, O40)

A general model of community formation and human capital accumulation with social spillovers and decentralized school funding is used to analyse the causes of economic segregation and its consequences for equity and efficiency. Significant polarization arises from minor differences in endowments, preferences or access to capital markets. This makes income inequality more persistent across generations, but the same need not be true for wealth. Equilibrium stratification tends to be excessive, resulting in low aggregate surplus. Whether state equalization of school resources can remedy these problems hinges on how purchased, social and family inputs interact in education and in mobility decisions.

A growing body of literature suggests that variations across countries, in entrepreneurial activity and the spatial structure of economies could potentially be the source of different efficiencies in knowledge spillovers, and ultimately in economic growth. We develop an empirical model that endogenizes both entrepreneurial activity and agglomeration effects on knowledge spillovers within a Romerian framework. The model is tested using the GEM cross-national data to measure the level of entrepreneurship in each particular economy. We find that after controlling for the stock of knowledge and research and development expenditures, both entrepreneurial activity and agglomeration have a positive and statistically significant effect on technological change in the European Union. © Springer 2005.

During college, many students switch from their planned major to another, particularly so when that planned major was in a Science, Technology, Engineering, or Mathematics (STEM) field. A worrying statistic shows that persistence in one of these majors is much lower for women and minorities, suggesting that this may be a leaky joint in the STEM pipeline for these two groups of students. This paper uses restricted-use data from the National Longitudinal Survey of Freshmen (NLSF) and the National Education Longitudinal Study of 1988 (NELS:88) to examine which factors contribute to persistence of all students in STEM field majors, and in particular the persistence of women and minorities. Although descriptive statistics show that a smaller percentage of women and minorities persist in a STEM field major as compared to male and non-minority students, regression analysis shows that differences in preparation and the educational experiences of these students explains much of the differences in persistence rates. Students at selective institutions with a large graduate to undergraduate student ratio and that devote a significant amount of spending to research have lower rates of persistence in STEM fields. A higher percentage of female and minority STEM field graduate students positively impacts on the persistence of female and minority students. However, there is little evidence that having a larger percentage of STEM field faculty members that are female increases the likelihood of persistence for women in STEM majors. These results suggest that the sorting of women and minorities into different types of undergraduate programs, as well as differences in their backgrounds have a significant impact on persistence rates. © 2010 Elsevier Ltd.

I present a model where firms decide what types of jobs to create and then search for suitable workers. When there are few skilled workers and the skilled-unskilled productivity gap is small, firms create a single type of job and recruit all workers. An increase in the proportion of skilled workers or skill-biased technical change can create a qualitative change in the composition of jobs, increasing the demand for skills, wage inequality, and unemployment. I provide some evidence that there has been a change in the composition of jobs in the United States during the past two decades. (JEL E24, J31, J64)

We develop a model that, at the aggregate level, is similar to the one-sector neoclassical growth model; at the disaggregate level, it has implications for the path of observable measures of technology adoption. We estimate it using data on the diffusion of 15 technologies in 166 countries over the last two centuries. Our results reveal that, on average, countries have adopted technologies 45 years after their invention. There is substantial variation across technologies and countries. Newer technologies have been adopted faster than old ones. The cross-country variation in the adoption of technologies accounts for at least 25 percent of per capita income differences. (JEL O33, O41, O47)

This graduate level text on economic growth surveys neoclassical and more recent growth theories, stressing their empirical implications and the relation of theory to data and evidence. The authors have undertaken a major revision for the long-awaited second edition of this widely used text, the first modern textbook devoted to growth theory. The book has been expanded in many areas and incorporates the latest research. After an introductory discussion of economic growth, the book examines neoclassical growth theories, from Solow-Swan in the 1950s and Cass-Koopmans in the 1960s to more recent refinements; this is followed by a discussion of extensions to the model, with expanded treatment in this edition of heterogenity of households. The book then turns to endogenous growth theory, discussing, among other topics, models of endogenous technological progress (with an expanded discussion in this edition of the role of outside competition in the growth process), technological diffusion, and an endogenous determination of labor supply and population. The authors then explain the essentials of growth accounting and apply this framework to endogenous growth models. The final chapters cover empirical analysis of regions and empirical evidence on economic growth for a broad panel of countries from 1960 to 2000. The updated treatment of cross-country growth regressions for this edition uses the new Summers-Heston data set on world income distribution compiled through 2000.

Abstract What shapes the optimal degree of progressivity of the tax and transfer system? On the one hand, a progressive tax system can counteract inequality in initial conditions and substitute for imperfect private insurance against idiosyncratic earnings risk. On the other hand, progressivity reduces incentives to work and to invest in skills, distortions that are especially costly when the government must finance public goods. We develop a tractable equilibrium model that features all of these trade-offs. The analytical expressions we derive for social welfare deliver a transparent understanding of how preference, technology, and market structure parameters influence the optimal degree of progressivity. A calibration for the U.S. economy indicates that endogenous skill investment, flexible labor supply, and the desire to finance government purchases play quantitatively similar roles in limiting optimal progressivity. In a version of the model where poverty constrains skill investment, optimal progressivity is close to the U.S. value. An empirical analysis on cross-country data offers support to the theory.

The intellectual breakthrough contributed by the new growth theory was the recognition that investments in knowledge and human capital endogenously generate economic growth through the spillover of knowledge. However, endogenous growth theory does not explain how or why spillovers occur. This paper presents a model that shows how growth depends on knowledge accumulation and its diffusion through both incumbents and entrepreneurial activities. We claim that entrepreneurs are one missing link in converting knowledge into economically relevant knowledge. Implementing different regression techniques for the Organisation for Economic Co-operation and Development (OECD) countries during 1981 to 2002 provides surprisingly robust evidence that primarily entrepreneurs contributed to growth and that the importance of entrepreneurs increased in the 1990s. A Granger test confirms that causality goes in the direction from entrepreneurs to growth. The results indicate that policies facilitating entrepreneurship are an important tool to enhance knowledge diffusion and promote economic growth.

We review the literature on national innovation systems. We first focus on the emergence of the concept of innovation systems, reviewing its historical origins and three main flavors (associated to three "founding fathers" of the concept). After this, we discuss how the notion of innovation systems filled a need for providing a broader basis for innovation policy. We conclude with some perspectives on the future of the innovation systems literature. © 2010 Elsevier B.V.

The empirical analysis in "International R&D Spillovers" [Coe, D., Helpman, E., 1995. International R&D Spillovers. European Economic Review, 39, 859-887] is first revisited on an expanded data set that we have constructed for the purpose of this study. The new estimates confirm the key results reported in Coe and Helpman about the impact of domestic and foreign R&D capital stocks on TFP. In addition, we show that domestic and foreign R&D capital stocks have measurable impacts on TFP even after controlling for the impact of human capital. Furthermore, we extend the analysis to include institutional variables. Our results suggest that institutional differences are important determinants of TFP and that they impact the degree of R&D spillovers. Countries where the ease of doing business and the quality of tertiary education systems are relatively high tend to benefit more from their own R&D efforts, from international R&D spillovers, and from human capital formation. Strong patent protection is associated with higher levels of total factor productivity, higher returns to domestic R&D, and larger international R&D spillovers. Finally, countries whose legal systems are based on French and, to a lesser extent, Scandinavian law benefit less from their own and foreign R&D capital than countries whose legal origins are based on English or German law. © 2009 Elsevier B.V. All rights reserved.

Economists have long treated technological phenomena as events transpiring inside a black box and, on the whole, have adhered rather strictly to a self-imposed ordinance not to inquire too seriously into what transpires inside that box. The purpose of Professor Rosenberg's work is to break open and examine the contents of the black box. In so doing, a number of important economic problems be powerfully illuminated. The author clearly shows how specific features of individual technologies have shaped a number of variables of great concern to economists: the rate of productivity improvement, the nature of learning processes underlying technological change itself, the speed of technology transfer, and the effectiveness of government policies that are intended to influence technologies in particular ways. The separate chapters of this book reflect a primary concern with some of the distinctive aspects of industrial technologies in the twentieth century, such as the increasing reliance upon science, but also the considerable subtlety and complexity of the dialectic between science and technology. Other concerns include the rapid growth in the development of costs associated with new technologies as well as the difficulty of predicting the eventual performance characteristics of newly emerging technologies

Abstract This article examines the labor market payoffs to different types of postsecondary education, including field and institution of study. Instrumental variables (IV) estimation of the payoff to choosing one type of education compared to another is made particularly challenging by individuals choosing between several types of education. Not only does identification require one instrument per alternative, but it is also necessary to deal with the issue that individuals who choose the same education may have different next-best alternatives. We address these difficulties using rich administrative data for Norway’s postsecondary education system. A centralized admission process creates credible instruments from discontinuities that effectively randomize applicants near unpredictable admission cutoffs into different institutions and fields of study. The admission process also provides information on preferred and next-best alternatives from strategy-proof measures of individuals’ ranking of institutions and fields. The results from our IV approach may be summarized with three broad conclusions. First, different fields of study have substantially different labor market payoffs, even after accounting for institution and peer quality. Second, the effect on earnings from attending a more selective institution tends to be relatively small compared to payoffs to field of study. Third, the estimated payoffs to field of study are consistent with individuals choosing fields in which they have a comparative advantage. Comparing our estimates to those obtained from other approaches highlights the importance of using instruments to correct for selection bias and information on individuals’ ranking of institutions and fields to measure their preferred and next-best alternatives.

This paper analyzes a model of economic growth, with technological innovations that reduce labor requirements but raise capital requirements. The paper has two main results. The first is that such technological innovations are not everywhere adopted, but only in countries with high productivity. The second result is that technology adoption significantly amplifies differences in productivity between countries. This paper can, therefore, add to our understanding of large and persistent international differences in output per capita. The model also helps to explain other growth phenomena, like divergence or periods of rapid growth.

Questioning the underlying assumptions of the process of creative destruction, we conceptualize an alternative process of creative construction that may characterize the dynamics between entrants and incumbents. We discuss the underlying mechanism of knowledge spillover strategic entrepreneurship whereby knowledge investments by existing organizations, when coupled with entrepreneurial action by individuals embedded in their context, results in new venture creation, heterogeneity in performance, and subsequent growth in industries, regions, and economies. The framework has implications for future research in entrepreneurship, strategy, and economic growth. Copyright (C) 2008 Strategic Management Society.