Session 20: Models of the Labor Market and the Aggregate Economy: New Developments and Policy Implications

Micro estimates of the Marshallian elasticity of labor supply are small and typically positive, whereas cross-country and time-series patterns of hours imply a strong negative relationship between wages and hours. I reconcile these two apparently contradictory observations using a single utility specification and taking into account heterogeneity in non-labor income. Micro estimates condition on non-labor income, while macro variation allows capital income to adjust alongside labor in- come, which strengthens the income effect. A model with heterogeneous households and exogenous capital income yields closed-form expressions in which the distribution of the labor share shapes the gap between the micro and the macro elasticities. A cross-sectional regression of hours on wages that conditions on the labor share recovers the macro elasticity. A dynamic model with heterogeneous households and incomplete asset markets reproduces both elasticities as outcomes when disciplined by joint moments of wages, hours, consumption, and wealth. The income effects that bridge the gap between the two elasticities imply marginal propensities to earn that lie in the range of estimates of micro studies on lottery winners.

In frictionless economies, the risk-free rate equals the household discount rate. With uninsurable income risk, precautionary savings depress the risk-free rate below the discount rate. We develop a model in which market power affects equilibrium asset returns through household portfolio allocation between safe assets and risky equity under financial frictions. Higher markups raise profits and increase the share of risky equity in household wealth. Because equity returns are risky, this shift increases exposure to equity risk and strengthens precautionary demand for safety. At the same time, higher markups reduce capital accumulation and contract the supply of safe assets through firms’ collateral constraints. The equilibrium risk-free rate declines to reconcile stronger demand for safety with reduced safe asset supply, widening the equity premium. Quantitatively, the rise in market power and labor income risk account for most of the observed decline in the risk-free rate, while changes in markups explain the bulk of the movement in the return on equity. We also characterize two new channels through which markups decrease welfare: distribution and risk exposure, which add to the standard deadweight loss effect.

We document and explain the gap between measures of AI exposure and measures of AI adoption in the workplace. This leads us to propose a new AI adoption index based on comparative advantage. Using the representative German DiWaBe employee survey linked to worker and establishment information, we compare worker-reported AI use to prominent exposure measures and find that the relationship is weak. Motivated by this gap, we develop a framework in which adoption depends not only on technical feasibility—AI’s absolute advantage measured by exposure—but on profitability—AI’s comparative (dis)advantage relative to a specific worker—balancing AI productivity against AI user costs and worker productivity against wages. We operationalize this framework at the task level by (i) estimating worker productivity relative to pay, (ii) mapping exposure indices into AI productivity, and (iii) inferring task-specific AI user costs from revealed-preference adoption. The resulting occupation-level index accounts for 60% of cross-occupation variation in observed AI adoption, compared to 14% for an exposure-only model. The two approaches diverge substantially for approximately 30% of workers, highlighting that comparative advantage—not exposure alone—is crucial for assessing AI’s labor-market impact.

We propose a novel empirical strategy to infer the extent to which rm-worker matches are inspection or experience goods. We argue that the informative content of the signals that rms and workers receive about the productivity of their match before entering an employment relationship can be inferred from the gaps between the separation rates of workers hired from unemployment, employment at low-tenure jobs, and employment at high-tenure jobs. We implement the strategy using German administrative data. We nd that, before entering an employment relationship, a rm and a worker receive a signal that reduces the variance of their beliefs about the productivity of the match by 67%. The informative content of the signal varies according to the gender and the education of the worker, and it has increased over time. If matches were pure inspection goods, labor productivity and output would be 2% higher. If matches were pure experience goods, labor productivity would be 3% lower, and output 5% lower.

This paper studies gender gaps in labor-market outcomes, with a focus on job ladder dynamics. We show that women experience substantially lower wage growth conditional on prior wages despite nearly identical job-to-job transition rates for men and women. To reconcile these observations, we document gender differences in the valuation of nonwage job amenities and in job search behavior, and develop a multi-dimensional job-ladder model with endogenous search effort where workers value both wages and amenities. The model allows for gender heterogeneity in separation rates, search effort, the value of nonemployment, amenity valuations, and bargaining power, enabling a joint analysis of gender wage and employment gaps. A quantitative decomposition shows that differences in preferences for nonwage amenities account for nearly 40 percent of the gender pay gap. Differences in the value of nonemployment and bargaining power explain most of the remainder, with only a limited role for differences in separation rates and search behavior. Finally, we show that increases in job amenities—such as the expansion of remote work—raise the gender wage gap while reducing gender differences in employment.

Technical change often simplifies jobs. This increases productivity, but it also makes workers more substitutable—or more “commoditized”. Commoditization of labor drives down worker bargaining power: anyone can do the job, implying workers are disposable, which improves the outside option of firms and can lower worker wages. We develop a model that captures both the productivity enhancing and wage depressing effects of commoditizing technical change. Commoditizing technical change involves firms standardizing tasks. This reduces the sensitivity of output to worker quality. Firms benefit because they can more easily fill vacancies for their durable jobs. We show that our model can help explain the divergence between productivity and wages in the service sector, increasing wage markdowns despite falling local concentration, and the decline of the large-firm wage premium.

We estimate a macroeconomic model of the dynamics of the wealth distribution, driven by stochastic earnings, differential savings, and wealth-dependent capital income risk. We do not impose stationarity: we initialize the model with the 1962 SCF distribution and estimate its parameters by matching the transitional dynamics of wealth in the U.S. through 2022, together with intergenerational social mobility. The model accurately reproduces the dramatic rise in wealth concentration observed over this period - for both White and Black households. We then study the racial wealth gap, allowing for race-specific return processes, and find that differences in base rates of return-compounded by wealth-dependent returns-account for a large share of the persistent gap between White and Black wealth. Counterfactual simulations indicate that reverting tax progressivity to 1962 levels would substantially attenuate the rise in top wealth shares.

Many of the wealthiest Americans are professional capitalists: business owners and founders tied to businesses in which they remain personally involved. This paper develops a Q-theoretic model in which financial and agency frictions create a distinction between the market value of these businesses, which governs a professional capitalist’s measured wealth, and their replacement-cost value, which governs the professional capitalist’s consumption and private valuation of the firm. The model implies that cross-firm dispersion in market-value-to-replacement-cost ratios can make measured wealth more concentrated than consumption. It also implies that increases in the level and dispersion of valuation ratios can raise top wealth without a proportional rise in replacement-cost value. Increases in market-value wealth driven by higher valuation ratios therefore have different welfare implications than increases driven by greater replacement-cost value. Consistent with the mechanism, very-top wealth concentration co-moves with the level and book-equity-weighted dispersion of public-firm market-to-book ratios.

We offer new empirical facts and insights into the determinants, dynamics, and consequences of extreme poverty in an advanced market economy, based on a novel administrative data set from France’s largest food assistance charity that is unique in its comprehensive coverage of low-income households– in particular of homeless individuals. We document several novel facts. First, extreme poverty is largely a transient phenomenon (1.5 years on average), but a very large fraction of households is at risk of falling into food and housing insecurity (8% of the population transited through our sample in the past 7 years). Second, children and teens are vastly over-represented in food and housing insecurity relative to their share in the overall French population (20% of our sample are below age 9, vs. 10% in the population). Third, flows into and out of extreme poverty are asymmetric: inflows are highly sensitive to local macroeconomic conditions and policies (in particular, a 10% increase in the unemployment rate or in private housing rents lead to a 10% increase in homelessness) while outflows only depend on individual characteristics, with an especially strong duration dependence. Fourth, housing is a luxury good, with marginal propensities to consume and price elasticities equal to zero below a subsistence threshold. Fifth, the high inflation period starting in 2022 played a central role in accounting for the fast rising rates of extreme poverty in the last four years.