LAWRENCE D. W. SCHMIDT

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.

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 exposure to labor-saving technologies negatively affects the earnings of exposed workers. This negative effect is pervasive across both blue- and white-collar workers and across workers of different ages or earnings relative to their peers. In contrast, labor-augmenting technologies have a heterogeneous impact on exposed workers. While the wage bill paid to affected groups rises, this increase is driven primarily by an increase in employment, while earnings rise for new entrants but decline for incumbent workers. This decline is primarily present among white-collar, older, and higher-paid workers, highlighting the importance of vintage-specific human capital. Last, we find positive spillovers of both types of innovation at the industry level, benefiting other workers in the same industry who are not directly exposed to these innovations.

We study how financial conditions shape labor income risk through endogenous firm insurance, job creation, and job destruction. We develop a directed search model with dynamic wage contracts, two-sided limited commitment, and time-varying risk premia where two forces govern the optimal contract: an insurance motive that smooths wages and a retention motive that deters poaching. When risk premia rise, limited commitment constraints tighten and insurance erodes—especially for low-wage workers near the separation margin—generating state-dependent labor income risk. Using U.S. administrative data, we document consistent evidence: the pass-through of firm productivity shocks to worker earnings rises when risk premia are elevated, concentrated among lower-paid workers and operating through job destruction. The calibrated model matches labor market dynamics, asset prices, and the heterogeneous pass-through of firm shocks. Beyond these targeted moments, it generates realistic variation in non-Gaussian earnings dynamics across workers and over time, sizable welfare losses from idiosyncratic risk, depressed human capital valuations, and large welfare gains from state-contingent policies.

Using U.S. administrative data on worker earnings, we show that increases in risk premia lead to lower labor earnings, particularly for lower-paid workers. These declines are primarily driven by job separations. We build an equilibrium model of labor market search that quantitatively replicates the observed heterogeneity in labor market dynamics across worker earnings levels. Our findings underscore the role of time-varying risk premia as a key driver of labor market fluctuations and highlight the importance of both the job creation and the job destruction margins in understanding the heterogeneity in worker outcomes over the business cycle.

U.S. employers and the federal government devote the equivalent of 1.5% of GDP annually toward promoting defined contribution (DC) retirement savings. Using a new employer-employee linked dataset covering millions of Americans, we show that tax and employer matching incentives disproportionately benefit White and Asian workers compared to their similar-income Hispanic, Black, and American Indian or Alaska Native coworkers. Similarly, these incentives disproportionately benefit those with richer parents compared to those from lower-income families. Breaking the link between contribution choices and saving subsidies through revenue-neutral reforms could close up to one-third of the DC wealth gaps by race and parental income.

For whom has earnings risk changed, and why? We answer these questions by combining the Kalman filter and EM-algorithm to estimate persistent and temporary earnings for every individual at every point in time. We apply our method to administrative earnings linked with survey data. We show that since the 1980s, persistent earnings risk rose by 12.5% for both employed and unemployed workers and the scarring effects of unemployment doubled. At the same time, temporary earnings risk declined. Using education and occupation codes, we show that rising persistent earnings risk is concentrated among high-skill workers and related to technology adoption.