UC San Diego
Firm Learning, Unemployment, and Self-Employment in Growth and Development
- Author(s): Feng, Ying
- Advisor(s): Rauch, James E
- Lagakos, David
- et al.
Differences in average income levels across countries are vast. This dissertation investigates the interaction between heterogeneous firms or workers' decisions and economic growth.
Chapter 1 of this dissertation studies firm life-cycle learning and misallocation. Misallocation is one of the most prominent theories of Total Factor Productivity in recent years. Specifically, this study focuses on misallocation of resources across producers. Dispersion in marginal revenue products of capital (MRPK) across firms may lower aggregate productivity through misallocation. Using firm-level panel data from China, I document that MRPK dispersion decreases substantially with firm age, particularly before age five. Building on this fact, I provide a new interpretation of MRPK dispersion as firm life-cycle learning. I formalize this idea in a dynamic model, in which firms learn about their fundamental productivity as they age and choose capital inputs in a frictional market based on their priors. Within each cohort of firms, imprecise priors lead firms to differ in their ex-post MRPK even in the absence of firm-level distortions. As firms learn over time and adjust their capital stocks, possibly through exiting the market, dispersion in MRPK decreases. Quantitative analysis of the model shows that omitting firm life-cycle learning leads to sizable overestimation of the aggregate productivity losses from misallocation.
Chapter 2 of this dissertation asks: How does the average unemployment rate change with GDP per capita? This chapter draws on household survey data from countries of all income levels to measure how unemployment varies with income. We document that unemployment is increasing with GDP per capita. Furthermore, we show that this fact is accounted for almost entirely by low-educated workers, whose unemployment rates are strongly increasing in GDP per capita, rather than by high-educated workers, whose unemployment rates are not correlated with income. To interpret these facts, we build a model with workers of heterogeneous ability and two sectors: a traditional sector, in which self-employed workers produce output without reward for ability; and a modern sector, in which firms hire in frictional labor markets, and output increases with ability. Countries differ exogenously in the productivity level of the modern sector. The model predicts that as productivity rises, the traditional sector shrinks, as progressively less-able workers enter the modern sector, leading to a rise in overall unemployment and in the ratio of low-educated to high-educated unemployment rates. A calibrated version of the model accounts for some, but not all, of the cross-country patterns we document.
Chapter 3 of this dissertation proposes a universal division of different types of self-employment. It is well-known that self-employment rate declines with GDP per capita (Gollin, 2008). However, when dividing self-employment into employers and own-account workers (self-employed without employees), this paper documents that the labor share of employers increases with income levels, and the share of own-account workers decreases. Using household surveys from countries of all income levels, we show these facts are robust across main industries and educational categories. We also show nearly universal negative selection on ability into own-account status, and positive selection into employer and wage earning statuses in our data. We develop a simple two-sector model to explain these facts. In general equilibrium, agents with ability below a threshold become own-account workers in the traditional sector, and agents with ability higher than the threshold enter the modern sector, becoming wage workers or employers. Higher aggregate productivity is driven by higher returns to ability in the modern sector due to skill biased technological change, which reduces the threshold ability level. By distinguishing between own-account workers and employers consistently across 56 countries, our database and model help reconcile diverse findings about development and entrepreneurship.