Research

Abstract: This paper develops a novel theory of domestic outsourcing grounded in search frictions and worker-firm heterogeneity. The theory is motivated by new empirical evidence on the outsourcing of skilled services in the United States, notably its rising employment share, outsourcing wage premium, and the discernible sorting of higher-ability workers into outsourcing. In the model, workers, output-producing firms, and labor intermediaries participate in a single frictional labor market. These intermediaries then play the role of a competitive matching market for worker abilities and firm characteristics. A quantitative formulation of the model unveils that, by mitigating the adverse effects of search frictions, the increase in outsourcing share of employment since the 1980s  has resulted in output gains of 4 percent, with workers being better off by 2.5 percent

Abstract: Per capita hours worked in the United States have shown virtually no secular trend in the postwar period. This absence of trend masks two offsetting forces: hours per worker steadily decrease throughout the period, whereas the employment-to-population ratio increases. Using data and quantitative theory, we attribute these patterns to a slowdown in the rate of technical progress. 

Abstract: Increasing separations and higher labor force attachment among the nonemployed during recessions result in a rise in unemployment. This study presents new evidence that this labor attachment, i.e., the participation margin, contributes to around 38% of observed fluctuations in unemployment. There is also a difference in how this margin affects high- and low-wage workers, leading to a shift in the average skill composition of the unemployment pool during recessions. The paper proposes a representative agent model that can replicate the patterns seen in the data, revealing that rising opportunity cost of employment in recessions drives the increase in labor force attachment. Importantly, this finding relies on how the utility function is parameterized, particularly in terms of its intertemporal elasticity. The participation margin increases only when the income effects outweigh the substitution effects of a recession. Furthermore, the varying sensitivity of different wage groups can be explained by differences in unemployment insurance benefits and differences in steady-state separation and job-finding rates.

Abstract: Standard business cycle models cannot account for the magnitude and persistence of the cyclical fluctuations in the labor share of output in the United States. A model with search frictions in the labor market and a technology choice can resolve this shortcoming. The technology choice allows for general patterns of capital- and labor-augmenting technical change in the short run while being consistent with Hicks-neutral technical change in the long run when adjustment costs vanish. Production is CES in the short run and Cobb-Douglas in the long run, consistent with a balanced growth path under standard restrictions on preferences. We calibrate the model to U.S. data and quantify that the technology choice with adjustment costs enhances propagation relative to the model with fixed Cobb-Douglas technology. The model also reproduces the labor share overshooting property in the data.

Abstract: What kind of occupations are outsourced? This paper provides evidence that occupations requiring a high degree of task complexity are primarily allocated to the skilled outsourcing sector. Using data on job requirements from O*NET, I construct an occupational-level index of relative task complexity, which is positively correlated with the probability of a worker being outsourced. This index is used to inform a task-based model of a representative firm, where the fraction of outsourced tasks is determined endogenously. The model reveals that the shift in the task composition of outsourcing has contributed an additional 6 percent to aggregate output growth in the past four decades.

Abstract: We examine the interaction between unemployment benefits and subsistence self-employment in developing countries, where limited opportunities in formal labor markets push low-productivity individuals into self-employment. We develop a model that incorporates occupational choices, search frictions, and borrowing constraints, allowing individuals to choose between wage work, own-account work, and entrepreneurship. Our model highlights how low-wealth, low-ability individuals select into own-account work due to their inability to sustain the search for wage employment in a frictional labor market. Calibrating the model to Mexico, we explore the effects of introducing U.S.-style unemployment benefits. Our results indicate that if own-account workers can conceal their employment status, unemployment benefits lead to an excessive increase in own-account work and a reduction in aggregate output. However, when employment status is effectively monitored, unemployment benefits can reduce subsistence self-employment by 15 percentage points and generate significant output gains of nearly 2%.

Abstract: A worker gets better at her job over time because of two channels: 1) climbing up the skill ladder or 2) learning about match quality. In this paper, we evaluate the quantitative roles of these two channels in a life cycle perspective. We build up a life cycle search and matching model with a skill ladder and learning about matching quality. We find that learning provides additional attachment to the labor force, and it accounts for 5% of the growth of life cycle earnings.

Abstract: How much capital should the central bank of a country hold? There is no consensus on this matter. We review the balance sheets of 45 central banks from around the world to describe actual practices. The principal findings are: (a) the average capital-asset ratio of central banks globally (net of revaluation capital which is purely an accounting entry) is 6.56 percent while the number in emerging economies is 6.96 percent; and (b) our Value-at-Risk estimates for the RBI excluding exchange rate risk indicate that the current level of the core capital of the RBI as mandated by the Jalan committee may be too low. We also discuss the policy moral hazards associated with mandating RBI equity payouts to the government.