Labor Economics — Wages, Unemployment, Discrimination, Returns to Education

Labor economics studies the markets where human effort is bought and sold. It asks: how are wages determined, why do some people work more hours than others, why do unemployment rates fluctuate, what does education actually pay for, where does discrimination come from and what eliminates it, how do unions and minimum wages affect outcomes, what happens when robots and AI can do tasks previously done by humans? Labor is unusual among markets: it is inalienable from the person who supplies it, its quality depends on motivation and culture as well as skill, it cannot be stored for later, and the same person is on the supply side here and the demand side everywhere else as a consumer. These peculiarities give labor economics its distinctive shape.

The field has long produced more Nobel laureates per capita than almost any other applied subfield: Becker (1992), Heckman (2000), Mortensen-Diamond-Pissarides (2010), Card-Angrist-Imbens (2021), Goldin (2023), and Acemoglu-Johnson-Robinson (2024) all worked centrally on labor. The 2010 and 2021 prizes recognized search-and-matching models and the credibility revolution in empirical methods, respectively — both transformed how economists think about wages and employment.

1. Labor demand

A firm’s demand for labor is a derived demand: it depends on the demand for the firm’s product. A profit-maximizing competitive firm hires workers up to the point where the marginal revenue product of labor (MRPL = MPL × price for a price-taker; MRPL = MPL × MR for a monopolist) equals the wage. Where workers are heterogeneous, the firm chooses the optimal mix of skill levels along the same logic, equating the marginal product of each type to its wage.

The wage elasticity of labor demand — the percent change in hours demanded for a 1% change in the wage — has been extensively estimated. Daniel Hamermesh’s Labor Demand (Princeton, 1993) surveys the literature, with consensus estimates of own-wage elasticities clustering around −0.3 to −0.5 in the short run and somewhat larger in absolute value over longer horizons. The Hicks-Marshall laws of derived demand identify the determinants: labor demand is more elastic when (1) the product demand is more elastic, (2) labor is a larger share of total cost, (3) substitutes for labor (capital, intermediate inputs) are more readily available, and (4) supply of substitute factors is more elastic.

The cross-elasticity of substitution between capital and labor has been estimated for decades. Most estimates of the elasticity of substitution between capital and labor (σ) at the aggregate level fall in the 0.4 to 0.8 range, supporting a Cobb-Douglas-like specification with σ ≈ 1 as a tractable approximation. The capital share of national income rose globally from roughly 30% in the 1970s to over 40% by the 2010s in many advanced economies (Karabarbounis-Neiman 2014), a trend with major implications for inequality and for labor’s bargaining position.

2. Labor supply

Individuals decide how many hours to work by trading off consumption (purchased with wages) against leisure. The substitution effect of a wage increase makes leisure more expensive and pushes hours up. The income effect makes the worker richer at the same hours and, if leisure is normal, pushes hours down. The net effect is the famous backward-bending labor supply curve: at low wages, the substitution effect dominates and hours rise with wages; at high wages, the income effect dominates and additional wages buy more leisure.

Labor force participation (LFP) is the extensive margin — whether to work at all. US LFP peaked at 67.3% in early 2000 and declined to 62.5% in 2024, driven by aging, declining male prime-age participation, and shifts in caregiving. The prime-age (25-54) LFP rate of 83.6% in mid-2024 had recovered from its pandemic trough of 79.7% in April 2020. Female LFP rose from 33.9% in 1950 to a peak of 60.3% in 1999 before easing to roughly 57-58% in 2024 — a half-century rise that Claudia Goldin, the 2023 Nobel laureate, dissected across her career.

The reservation wage is the lowest wage at which an individual will accept work. Above it, the worker participates; below it, the worker stays out of the labor force or continues searching.

Intertemporal labor supply — Lucas and Rapping (Journal of Political Economy, 1969) — modeled work-leisure substitution across periods in response to expected wage changes, generating a positive intertemporal elasticity used in much business-cycle theory.

3. Human capital and the Mincer equation

The human capital revolution, led by Theodore Schultz, Gary Becker (Human Capital, 1964), and Jacob Mincer (Schooling, Experience, and Earnings, 1974), reframed education and training as investment in productive capacity rather than mere consumption.

The Mincer wage equation is the workhorse empirical specification:

log(wage) = α + β·schooling + γ₁·experience + γ₂·experience² + ε

where β is the average return per year of schooling, γ₁ and γ₂ capture the typical concave experience-earnings profile, and ε absorbs unobserved heterogeneity. Estimates of β typically fall in the 7-10% range per year of schooling for the developed world, with substantial variation across countries (Psacharopoulos and Patrinos meta-analyses).

The fundamental identification problem is that schooling is correlated with unobserved ability, so OLS estimates of β are biased upward. Strategies to address it include twin studies (Ashenfelter-Krueger 1994 — within-twin-pair returns of about 9%), instrumental variables (Angrist-Krueger 1991 quarter-of-birth, Card 1995 college proximity), and regression discontinuity. The credibility revolution in labor economics, recognized by the 2021 Nobel to David Card, Joshua Angrist, and Guido Imbens, transformed this and a hundred other questions by emphasizing quasi-experimental identification.

Signaling models (Michael Spence, Quarterly Journal of Economics, 1973; Nobel 2001) provide the alternative interpretation: education raises wages not because it builds productive capacity but because it signals pre-existing ability to employers in a separating equilibrium. The empirical literature finds both human-capital and signaling effects matter, with the balance varying by field and level.

General versus specific human capital (Becker 1962) — general human capital is portable across employers (literacy, programming); specific human capital is valuable only at a particular employer (knowledge of the firm’s systems, customers, internal politics). General human capital is paid for by workers (lower wages during training); specific human capital is paid for jointly because neither side can hold up the other ex post.

4. Discrimination

Two foundational theories explain wage gaps by group identity. Taste-based discrimination (Gary Becker, The Economics of Discrimination, 1957) models employers, customers, or coworkers as having “tastes” against members of a group, willing to pay (in foregone profits or higher prices) to avoid them. Competitive markets in the long run should drive prejudiced employers out, since non-discriminating firms with access to the underpaid group can undercut them — though many frictions slow this dynamic. Statistical discrimination (Edmund Phelps, American Economic Review, 1972; Kenneth Arrow 1973) models discrimination as a rational response to imperfect information: if employers cannot observe a particular worker’s productivity, they substitute group averages, leading to disparate treatment even without prejudice.

Audit studies provide some of the cleanest evidence of discrimination. Marianne Bertrand and Sendhil Mullainathan’s “Are Emily and Greg More Employable Than Lakisha and Jamal?” (American Economic Review, 2004) sent fictitious résumés varying only the name’s racial association to Boston and Chicago employers, finding callback rates 50% higher for white-sounding names than Black-sounding names, with the gap larger at higher résumé quality. Subsequent large-scale audit work — Kline, Rose, and Walters (Quarterly Journal of Economics, 2024) sent 84,000 applications to 11,000 entry-level jobs at 108 of the largest US firms — found persistent gaps across firms with substantial firm-level heterogeneity, identifying specific Fortune 500 firms with statistically significant discrimination.

Blinder-Oaxaca decomposition (1973) partitions group wage gaps into a component “explained” by observable characteristics (education, experience, occupation, hours) and an “unexplained” residual often interpreted as a discrimination upper bound. The decomposition has well-known caveats: characteristics may themselves be discrimination outcomes (e.g., differential access to schooling); residuals capture unobservables of all kinds, not only discrimination.

5. Wage determination

The competitive model — wage equals marginal product — is a starting point but rarely the whole story. Real labor markets feature search frictions, bargaining, asymmetric information, and behavioral elements that drive wedges between productivity and pay.

Search and matching

The Diamond-Mortensen-Pissarides (DMP) model (Peter Diamond, Dale Mortensen, Christopher Pissarides — Nobel 2010) treats unemployment and vacancies as the outcome of a matching technology. Unemployed workers and vacant jobs meet at a rate determined by an aggregate matching function M(u,v), with constant returns and Cobb-Douglas form typically assumed. The model generates the Beveridge curve — a downward-sloping relationship between unemployment u and vacancies v — and explains why both u and v can be positive simultaneously despite continuous flows. Wages emerge from Nash bargaining between worker and firm over the joint surplus of the match.

The DMP framework has become standard for analyzing labor market policy: how unemployment insurance affects job finding rates, how minimum wages interact with monopsony power, how labor demand shocks transmit through hiring and quits.

Compensating differentials

Adam Smith’s Wealth of Nations (1776) Chapter X observed that wages adjust for the non-pecuniary disamenities of work: dangerous, unpleasant, irregular, or low-status jobs must pay more to attract workers. Sherwin Rosen’s “Hedonic Prices and Implicit Markets” (Journal of Political Economy, 1974) gave the theory its modern form. Estimates of the value of a statistical life (VSL) from wage-fatality risk regressions cluster around $7-13 million in 2020 US dollars (Viscusi and Aldy meta-analyses), values used in cost-benefit analysis of safety regulation (see environmental-and-resource-economics).

Efficiency wages

Efficiency wage theories explain why firms might pay above the market-clearing wage and tolerate unemployment as an outcome. Shapiro and Stiglitz’s “Equilibrium Unemployment as a Worker Discipline Device” (American Economic Review, 1984) models firms paying above-market wages so that workers fear the cost of being fired and exert effort; in equilibrium, unemployment must exist to make the threat credible. Akerlof and Yellen’s Efficiency Wage Models of the Labor Market (1986) collects related theories — fair-wage / effort hypothesis (Akerlof-Yellen 1990), gift exchange (Akerlof 1982), turnover models (Salop 1979), and adverse selection (Weiss 1980).

Compensation structures and tournament theory

Lazear and Rosen’s “Rank-Order Tournaments as Optimum Labor Contracts” (Journal of Political Economy, 1981) modeled organizational pay hierarchies as tournaments where workers compete for promotion. The spread between pay levels — particularly the CEO premium — incentivizes effort across the entire hierarchy, not just at the top. The S&P 500 CEO-to-median-worker pay ratio reached approximately 290:1 in 2024 (AFL-CIO Executive Paywatch), up from roughly 20:1 in 1965 (Mishel and Kandra, EPI). The rise has been attributed variously to globalization and scale (Gabaix-Landier 2008), skill-biased technical change, governance failures, and tournament-theoretic effects.

6. Unemployment

Unemployment has multiple types. Frictional unemployment reflects the time it takes for workers and firms to find each other. Structural unemployment arises from mismatch between worker skills/location and available jobs. Cyclical unemployment is the component that rises during recessions. Seasonal unemployment reflects predictable annual fluctuations.

The natural rate of unemployment U*— the rate consistent with stable inflation — was introduced by Milton Friedman (1968) and Edmund Phelps (Nobel 2006) in the expectations-augmented Phillips curve framework. The Phillips curve relating inflation to unemployment is conditional on expectations: any inflation rate is consistent with U = U* in the long run, but inflation rises if U < Uand falls if U > U. CBO and Fed estimates of U* have evolved from approximately 6% in the early 1990s to around 4-4.5% in 2024, with substantial uncertainty.

Hysteresis — the proposition that actual unemployment affects the natural rate, so deep recessions raise long-run unemployment — was developed by Blanchard and Summers in “Hysteresis and the European Unemployment Problem” (NBER Macroeconomics Annual, 1986) to explain persistent high European unemployment after the early 1980s recession. Hysteresis mechanisms include skill atrophy among the long-term unemployed, insider-outsider bargaining (Lindbeck-Snower 1986), and discouragement.

The Beveridge curve plots vacancy rates against unemployment rates. Movements along the curve reflect labor demand shifts; outward shifts reflect deteriorated matching efficiency. The post-2020 Beveridge curve shifted outward dramatically during the pandemic recovery: 2022 saw both high vacancies (peaking at 12.2 million job openings in March 2022) and elevated unemployment, with the curve returning closer to its pre-pandemic position by 2024.

7. The minimum wage

The minimum wage is among the most contested topics in labor economics. The classical competitive model predicts that a binding minimum wage above the market-clearing wage reduces employment, with the magnitude depending on the elasticity of labor demand.

David Card and Alan Krueger’s “Minimum Wages and Employment: A Case Study of the Fast-Food Industry in New Jersey and Pennsylvania” (American Economic Review, 1994) used a difference-in-differences design comparing fast-food employment in New Jersey (which raised the minimum from $4.25to$5.05 on April 1, 1992) and eastern Pennsylvania (control). They found no significant negative employment effect, and possibly a small positive effect. The paper, expanded in their book Myth and Measurement (1995), launched a decades-long debate and helped earn Card the 2021 Nobel Prize.

Subsequent meta-analyses and bunching estimators have refined the picture. Cengiz, Dube, Lindner, and Zipperer’s “The Effect of Minimum Wages on Low-Wage Jobs” (Quarterly Journal of Economics, 2019) used 138 state-level minimum wage increases from 1979-2016 and bunching-estimator methods to find that minimum wage increases substantially raised earnings at the bottom of the distribution with negligible employment effects, consistent with low-wage labor markets featuring monopsony power.

The contrasting view, associated with David Neumark and others, finds more negative employment effects, particularly for the least-skilled workers, in specifications using synthetic control or different comparison groups. The Seattle minimum wage studies — UW (Jardim et al. 2018) finding large negative hours effects vs. Berkeley (Reich et al.) finding negligible effects — illustrate the empirical contestation.

The federal US minimum wage has been $7.25/hoursinceJuly24,2009—thelongeststretchwithoutanincreasesincetheFLS{A}^{\prime }s1938enactment.Stateandlocalminimumsdivergewidely,from$7.25 in 20+ states to $16inCalifornia,$16.28 in Washington, and even higher in some cities (DC at $17/hour as of mid-2024).

8. Unions and collective bargaining

US private-sector union membership peaked at approximately 35% of the workforce in 1954 and fell to 6.0% in 2024 (BLS). Public-sector unionization is much higher at 32.5%, and total union density is 10.0%. The long decline reflects the shift to service industries, right-to-work laws (now in 27 states), legal hostility (Taft-Hartley 1947, NLRB v. Mackay Radio 1938 permitting permanent striker replacements), globalization, and management opposition.

The Supreme Court’s Janus v. AFSCME (June 27, 2018) overturned Abood v. Detroit Board of Education (1977), holding that public-sector unions cannot collect agency fees from non-member workers, weakening public-sector union finances.

The 2023-24 period saw a notable revival of labor militancy. The United Auto Workers struck the Big Three automakers (GM, Ford, Stellantis) in September-October 2023 in the first simultaneous strike against all three, winning approximately 25% pay raises over 4.5 years plus restored cost-of-living adjustments. The Teamsters secured a strong contract with UPS in 2023 without a strike, raising the average top-rate pay package to over $170,000 (wages plus benefits). SAG-AFTRA struck the studios from July to November 2023 (118 days, the longest in the union’s history); the Writers Guild of America struck from May to September 2023 (148 days, the second-longest WGA strike). Starbucks Workers United organized over 400 stores from 2021-24 amid intense corporate resistance and slow contract progress. The Amazon Labor Union won an election at the Staten Island JFK8 warehouse in April 2022; Amazon has not voluntarily recognized it, and the case remains in NLRB and federal court litigation.

European labor markets retain much higher union density and sector-level collective bargaining: Iceland 91%, Denmark 67%, Sweden 65%, Finland 58%, Belgium 49%, Italy 32%, Germany 16%, France 7% (OECD 2024). Sector-level bargaining covers 80%+ of workers in much of continental Europe.

9. Gig economy and worker classification

The classification of platform workers (Uber, Lyft, DoorDash, Instacart) as independent contractors vs. employees has become a defining labor-policy battle. Employee classification triggers wage-and-hour protections (minimum wage, overtime), unemployment insurance contributions, workers’ compensation, employer payroll tax obligations, and collective-bargaining rights under the NLRA.

California’s AB5 (Assembly Bill 5, effective January 1, 2020) codified the “ABC test” from Dynamex Operations West v. Superior Court (2018), classifying workers as employees unless they meet all three of: (A) free from control, (B) work outside the usual course of the hiring entity’s business, (C) engaged in an independently established trade. Ride-hailing and delivery companies funded Proposition 22, passed by 58% of California voters in November 2020, carving out a separate classification for app-based drivers with limited benefits but contractor status. The California Supreme Court upheld Prop 22 in Castellanos v. State of California on July 25, 2024.

The federal Department of Labor’s January 2024 rule on independent contractor classification under the FLSA returned to a multifactor “totality of circumstances” test similar to pre-2021 law.

10. Returns to education and skill-biased technical change

The college wage premium — the average earnings advantage of a four-year college graduate over a high school graduate — rose dramatically in the US from roughly 35% in 1979 to over 80% by the late 1990s and 95-100% by the 2010s, before plateauing and modestly declining for newer cohorts.

Skill-biased technical change (SBTC) — Lawrence Katz and Kevin Murphy’s “Changes in Relative Wages, 1963-1987” (Quarterly Journal of Economics, 1992) — argued that technology has shifted demand toward more-skilled workers faster than supply has expanded, raising the skill premium. The task framework of Autor, Levy, and Murnane’s “The Skill Content of Recent Technological Change” (Quarterly Journal of Economics, 2003) refined this by partitioning tasks into routine cognitive, routine manual, non-routine cognitive (analytical, interactive), and non-routine manual. Computers substitute for routine tasks (both cognitive and manual) and complement non-routine cognitive tasks, predicting wage growth at the top, hollowing-out of the middle, and resilience at the bottom — the polarization pattern.

Autor and Dorn’s “The Growth of Low-Skill Service Jobs and the Polarization of the U.S. Labor Market” (American Economic Review, 2013) documented the labor-market polarization empirically. Autor, Dorn, and Hanson’s “The China Syndrome” (American Economic Review, 2013) showed that geographic exposure to Chinese import competition substantially reduced manufacturing employment in affected commuting zones, with persistent wage and employment scars not fully offset by trade gains. Enrico Moretti’s The New Geography of Jobs (2012) traced the divergence between “innovation hubs” and declining manufacturing regions.

11. Immigration and labor

Immigration shifts labor supply. The classical model predicts wage declines for native workers who substitute closely with immigrants and gains for those who complement them.

David Card’s “The Impact of the Mariel Boatlift on the Miami Labor Market” (Industrial and Labor Relations Review, 1990) used the 1980 Mariel boatlift — when Fidel Castro temporarily allowed Cubans to leave, sending 125,000 immigrants to Miami in a few months — as a natural experiment, finding negligible effects on native wages or unemployment. George Borjas’s reanalysis (“The Wage Impact of the Marielitos,” ILR Review, 2017) re-examined the data with attention to low-skill workers and high school dropouts, finding substantially negative effects. Card and others (Peri, Yasenov 2019) re-reexamined and disputed Borjas’s results, citing sample-composition changes in the comparison years. The debate continues to define immigration economics.

Borjas’s broader work emphasizes negative effects on native workers most substitutable with immigrants. The Card-led consensus emphasizes complementarities, capital adjustment, and downward effects on the order of zero to small for natives overall. The National Academies’ The Economic and Fiscal Consequences of Immigration (2017) survey concluded “the impact of immigration on overall native wages may be small.”

The H-1B program admits up to 65,000 skilled-worker visas annually plus 20,000 with US graduate degrees, with heavy oversubscription each year. STEM workforce growth has been disproportionately driven by immigration, with foreign-born workers comprising over 25% of US STEM employment in 2024.

12. Gender and the labor market

The raw gender wage gap — the ratio of female to male median earnings — has narrowed substantially over the past half century. The US ratio for full-time, full-year workers stood at approximately 84% in 2023 (Census), up from 60% in 1979. After controlling for occupation, experience, hours, and other observables, the residual gap shrinks to roughly 8-10%.

Claudia Goldin’s 2023 Nobel Prize recognized her career studying women’s labor force participation and the gender gap. Goldin’s “A Grand Gender Convergence: Its Last Chapter” (American Economic Review, 2014) and her book Career and Family (Princeton, 2021) identify the central remaining barrier as the “greedy work” premium — occupations that pay disproportionately more for long, inflexible hours penalize the parent (typically the mother) who must accommodate childcare. Convergence requires either changes in workplace flexibility, more equal parental responsibility, or technological substitutes for time-intensive child investment.

The motherhood penalty has been documented across rich countries. Kleven, Landais, and Søgaard’s “Children and Gender Inequality: Evidence from Denmark” (American Economic Journal: Applied Economics, 2019) used event-study methods around first childbirth in Danish administrative data, finding that mothers’ earnings drop approximately 20% relative to fathers’ and remain persistently lower a decade later — the “child penalty.” Similar magnitudes emerge across rich countries. The Goldin-Katz-Bertrand work on female MBAs (Bertrand, Goldin, and Katz, American Economic Journal: Applied Economics, 2010) showed that the gender pay gap widens substantially in the decade after MBA receipt, driven by hours and career interruption around childbirth.

US childcare costs as a share of income for low- and middle-income families exceed 20% in many states, far above the OECD average, creating sharp work-incentive distortions for second earners (typically mothers).

13. Aging and demographic shifts

The post-WWII Baby Boom generation (born 1946-64) reached retirement age starting in 2011 and the entire cohort will have reached 65 by 2030. The US old-age dependency ratio (population 65+ / 15-64) rose from approximately 20% in 2010 to 28% in 2024 and is projected to reach 35% by 2034.

The Social Security trust funds are projected to be depleted in 2033 (OASI fund, per 2024 Trustees Report), at which point payroll taxes alone would fund approximately 79% of scheduled benefits absent legislative action.

Japan’s old-age dependency ratio reached 51% in 2024, the highest in the OECD. Italy, Germany, South Korea, and increasingly China face similar demographic pressure. China’s working-age population peaked around 2015 and has been declining; its total population peaked in 2022 and is now shrinking. Demographic pressure on labor supply, growth potential, pension systems, and political economy is among the major medium-run forces in macroeconomics (see macroeconomics-foundations).

14. Remote work and the pandemic shock

The COVID-19 pandemic produced the largest single labor market dislocation since the Great Depression. US unemployment spiked from 3.5% in February 2020 to 14.8% in April 2020 (the highest since the Great Depression’s monthly estimates), before recovering relatively quickly thanks to massive fiscal support and the absence of underlying balance-sheet problems.

The pandemic’s most durable effect may be the normalization of remote and hybrid work. The Survey of Working Arrangements and Attitudes (SWAA) — Barrero, Bloom, and Davis’s “Why Working from Home Will Stick” (NBER, 2021, updated annually) — tracks the share of full paid workdays performed from home, which rose from approximately 5% pre-pandemic to over 60% in spring 2020 and stabilized around 28-30% by 2024, indicating roughly a five-fold permanent increase. The shift has been concentrated in high-skill information work; manufacturing, hospitality, retail, healthcare, and most service jobs remain overwhelmingly on-site.

The implications are still being worked out: commercial real estate has been substantially disrupted (US office vacancy at 20.1% in mid-2024 per CBRE, the highest on record); urban downtowns have struggled with reduced foot traffic; labor markets have widened geographically as workers can live further from headquarters; commute times have fallen by hours per week for remote workers.

15. Automation, robots, and AI

Daron Acemoglu and Pascual Restrepo’s research program on automation — including “Robots and Jobs” (Journal of Political Economy, 2020) and Power and Progress (Acemoglu and Johnson, 2023) — argues that automation has displaced workers without fully offsetting them through productivity gains, contributing to wage stagnation and inequality. Acemoglu’s 2024 Nobel Prize (with Simon Johnson and James A. Robinson) recognized work on institutions and economic outcomes that included this strand.

Brynjolfsson and McAfee’s The Second Machine Age (2014) and Frey and Osborne’s “The Future of Employment” (2013, estimating 47% of US jobs at high risk of automation) initiated a major debate. Subsequent task-level analyses (Arntz, Gregory, Zierahn 2016) produced more moderate estimates of approximately 9% of jobs fully automatable, though much larger shares of tasks within jobs.

The generative AI wave starting with ChatGPT’s November 2022 release has reopened the displacement debate. Eloundou et al. (“GPTs are GPTs,” 2023) estimated that 19% of US workers have at least 50% of their tasks affected by LLMs, with effects skewing toward higher-wage knowledge work. David Autor’s 2024 essay “Why Generative AI Could Help More Than Hurt Workers” argued that AI’s elite-task-substitution profile differs from prior automation waves and could compress wages by enabling less-trained workers to perform expert-level tasks. The empirical labor-market effects of generative AI as of 2024-25 are too early to measure definitively but appear modest at the macro level so far.

16. The care economy and labor

Employment in care services — childcare, eldercare, healthcare, home health aide work — has grown faster than overall employment for decades and is among the largest projected sources of new jobs. The BLS Occupational Outlook projects home health and personal care aides as the single largest source of US job growth from 2022-32 (over 800,000 additional jobs), with healthcare practitioners and technical occupations close behind. Care work is disproportionately female, frequently low-wage, and often performed by immigrants. Public investment in care infrastructure (childcare, paid leave, universal pre-K) was a major component of the Biden Build Back Better agenda that was substantially reduced in the Inflation Reduction Act.

17. Monopsony and the non-compete debate

The textbook competitive labor market has many employers competing for workers, driving wages to marginal product. Modern empirical work has found substantial monopsony power — concentrated employer-side bargaining power — in many local labor markets.

Naidu, Posner, and Weyl’s “Antitrust Remedies for Labor Market Power” (Harvard Law Review, 2018) catalogued the institutional and empirical case. Azar, Marinescu, and Steinbaum (Journal of Labor Economics, 2022) measured high concentration in many US local labor markets using Burning Glass vacancy data. Manning’s Monopsony in Motion (2003) and his subsequent work emphasized search frictions as a source of employer power even in apparently unconcentrated markets.

The FTC’s non-compete rule, issued April 23, 2024 under Chair Lina Khan, would have banned most new non-compete agreements and voided most existing ones (except for senior executives earning above $151,164). Federal courts in Texas (Ryan LLC v. FTC, August 20, 2024) and Florida set the rule aside on statutory and constitutional grounds; the FTC’s appeal is pending. State action continues independently: California, North Dakota, Oklahoma, and Minnesota broadly ban non-competes; many states limit them by wage threshold or industry.

DOJ-FTC joint guidance on no-poach agreements (2016) signaled criminal antitrust enforcement against employer wage-fixing and no-poach cartels. Several criminal prosecutions followed (railroad industry no-poach indictments, healthcare staffing no-poach cases), with mixed trial results but a clear policy reorientation toward labor-market antitrust.

18. Major economists and Nobels

• Jacob Mincer — human capital and the wage equation; never received the Nobel.
• Gary Becker (Nobel 1992) — human capital, discrimination, household production, crime.
• James Heckman (Nobel 2000) — selection bias correction (Heckman 1979), evaluation of social programs, early-childhood investment.
• Peter Diamond, Dale Mortensen, Christopher Pissarides (Nobel 2010) — search-matching unemployment theory.
• David Card, Joshua Angrist, Guido Imbens (Nobel 2021) — credible empirical labor economics, instrumental variables identification.
• Claudia Goldin (Nobel 2023) — gender labor history and the structural sources of the gender gap.
• Daron Acemoglu, Simon Johnson, James A. Robinson (Nobel 2024) — institutions and economic outcomes, including labor-related work on automation and inequality.

Other major figures without the Nobel: Henry Farber, Larry Katz, David Autor, Alan Krueger (died 2019), Marianne Bertrand, Sendhil Mullainathan, Raj Chetty, Emmanuel Saez, Henrik Kleven, Patrick Kline, and many more.

Adjacent

• microeconomics-foundations — labor market supply-demand framework, elasticity, deadweight loss.
• macroeconomics-foundations — Phillips curve, unemployment dynamics, business cycles.
• behavioral-economics — biases in labor supply, retirement saving, occupational choice.
• industrial-organization — monopsony, non-competes, and labor-market antitrust.
• development-economics — labor markets in developing economies, structural transformation.
• employment-and-labor-law — statutory framework: FLSA, NLRA, Title VII, ADA, ADEA, FMLA, OSHA.