« Back to Results

Firm Dynamics

Paper Session

Sunday, Jan. 5, 2020 8:00 AM - 10:00 AM (PDT)

Marriott Marquis, Vista
Hosted By: American Economic Association
  • Chair: James Tybout, Pennsylvania State University

Estimated Dynamic Industry Equilibrium Model with Firing Costs and Subcontracting

Maria Francisca Perez
,
University of Chile

Abstract

In Chile every permanent worker is entitled to one monthly wage per year of service with a maximum of eleven wages in case of dismissal. To regain flexibility this has triggered a widespread use of subcontracting. To assess the "true costs" of the regulation, I estimate a dynamic industry equilibrium model in which firms optimally choose the division labor between subcontracted workers that are totally flexible, and permanent workers that entail tenure-dependent firing costs. To overcome the potential costs of dismissing workers, firms hire permanent workers only up to the point where their expected firing costs equal the wage premium on subcontracted workers. Hence, firms’ use of subcontracted workers is decreasing in their relative cost and increasing in the volatility of shocks. I use a simulated method of moments by fitting plant-level employment dynamics and the size distribution in the manufacturing sector in Chile. I use data from the Chilean Annual National Manufacturing Survey, which reports permanent and subcontracted workers in the establishment where they physically work.

I find that severance payments are equivalent to seven monthly wages, and that workers get tenure after 4 years. Firms are willing to pay a “subcontracting wage premium” of 10% to substitute for hiring permanent workers. A naive researcher willing to estimate firing costs without considering the subcontracting margin of adjustment, would conclude that the Chilean labor market is rather flexible. Allowing firms to subcontract in a heavily regulated environment increases output, employment and productivity; firms respond more aggressively to productivity shocks, which enhances the allocation of labor across firms and hence total factor productivity. Removing the regulation leads in steady state to an increase in average labor productivity around 1%. Restricting the use of subcontracting to improve working conditions will lead to a decrease in total output, employment and productivity.

Import Exposure and Skill Content: Plant-Level Evidence from the United States

Mina Kim
,
U.S. Bureau of Labor Statistics
Jooyoun Park
,
Kent State University
Georg Schaur
,
University of Tennessee
T. Clay McManus
,
University of Tennessee

Abstract

We use the China trade shock to identify how import competition from low-wage countries affects the distributions of occupations and skills in U.S. plants. We find that import competition contributes to job polarization in the manufacturing sector; more import exposure decreases the employment of routine skills and increases the employment of non-routine skills. Adjustment occurs both within and
across plants, but through different skill channels. Within-plant adjustment occurs in large
plants toward non-routine skills. The decrease in routine skills occurs through the exit of highly routine plants. When looking at the economy as a whole, the link between import competition and job polarization is less clear even after taking into account input-output linkages.

Learning and Information Transmission within Multinational Corporations

Cheng Chen
,
University of Hong Kong and Clemson University
Chang Sun
,
University of Hong Kong
Hongyong Zhang
,
Research Institute of Economy, Trade and Industry (RIETI)

Abstract

We propose that multinational firms learn about their profitability in a particular market by observing their performance in similar markets. We first develop a model of firm expectation formation with noisy signals from multiple markets. A positive signal about demand inferred from similar markets raises the probability of entry into a new market, or raises the firm's sales expectation in an existing market. The latter effect is stronger when (1) the firm is less experienced in the focal market (2) the signals from the focal market are noisier and (3) the firm is more experienced in similar markets where signals are extracted. We confirm these predictions using a unique dataset of Japanese multinational corporations in which we observe sales expectations of each affiliate.

The Distribution and Evolution of Firm Productivity

Ming Xu
,
Queen's University
Mons Chan
,
University of Toronto
Sergio Salgado
,
University of Pennsylvania

Abstract

We study the distribution of firm-level shocks experienced by firms using panel data for the entire Danish private sector. We make several contributions. First, unlike the previous literature which focuses mostly on manufacturing, we measure firm-level Total Factor Productivity (TFP) for the entire private sector. Moreover, because ours is an Employer-Employee matched panel data set, we can control for workforce characteristics at the firm-level. Second, we show how to directly control for unobserved variation in labor quality across firms by combining a two-sided fixed-effect model with a novel nonparametric TFP estimation strategy. We use this novel methodology on the Danish MEE data to identify and separately characterize the distribution of permanent and transitory productivity shocks. Our empirical results show that controlling for labor quality with our method reduces the cross-sectional dispersion of firm productivity by up to 80% relative to estimates using standard measures of labor inputs such as hours or the wagebill. We also find that firm shocks display substantial deviation from the standard assumption of log-normality. In fact, the distribution of productivity shocks displays skewness and kurtosis that is more than seven times that of a Normal distribution. Higher kurtosis means that in a given year, most firms experience small fluctuations in productivity, but a non-negligible fraction of firms experience very large shocks. We also show that these deviations from log-normality persist when we control for observable differences across firms (such as age, industry, and size) and entry and exit of firms.

The ‘Missing Top’ of the Firm-Size Distribution in Developing Countries: What Is the Evidence?

Andrea Ciani
,
Düsseldorf Institute for Competition Economics
Alexandros Ragoussis
,
World Bank
Trang Thu Tran
,
World Bank

Abstract

Poorer markets tend on average to host smaller firms. That average conceals important differences within markets. Using a unique set of firm level data from five developing countries we show that that development is not only associated with higher frequency of the larger among large firms, but also the more productive and outward-oriented. If an ‘optimal’ number of large firms is defined proportionately against firms of smaller sizes, then we count less in developing countries than what is commonly observed in the rest of the world. Our findings provide empirical support for the ‘missing top’ rather than a ‘missing middle’ – a gap to the right of the firm size distribution in developing countries – and extend it to other features that characterize top performers. Failures preventing smaller as well as middle-sized firms from growing in many aspects could be the story behind these observations.
JEL Classifications
  • E2 - Consumption, Saving, Production, Investment, Labor Markets, and Informal Economy