Assitant Professor, School of Finance, Shanghai University of Finance and Economics
2017. 09 - now
PhD, Economics, University of Western Ontario
This paper quantifies the novel role of intermediate goods frictions, i.e. time-to-order and borrowing constraints, in accounting for the documented substantial misallocation in China Industrial Enterprise Survey (CIES) (Hsieh and Klenow, 2009; Brandt, Van Biesebroeck, and Zhang, 2012). The empirical part of this paper documents substantial intermediate goods misallocation and suggestive evidence of time-to-order and borrowing constraint frictions on intermediate goods in China's data. Quantitaive works find that a standard dynamic investment model with intermediate goods frictions can account for 70% of gross output misallocation in China, twice of the magnitude generated in an investment model with adjustment costs and borrowing constraints on capital only.
Substantial misallocation of inputs across firms is documented in the China Industrial Enterprise Survey Data (CIES) (e.g. Hsieh and Klenow, 2009). Is this measured misallocation due to distortions in reallocation across existing firms or distortions in entry and exit? To answer this question, I follow Bailey, Hulten and Campbell (1992) and decompose 5-year output-weighted aggregate productivity growth into contributions from reallocation, net entry, and firm-level productivity growth in the CIES. I find that net entry and firm-level productivity growth account for 91 percent and 18 percent of aggregate productivity growth, respectively. Reallocation across existing firms lowers growth by 9 percent. This is surprising, since the literature finds that reallocations account for over 30 percent of the U.S. manufacturing productivity growth (Bailey, Hulten and Campbell, 1992; Foster, Haltiwanger and Krizan, 2003). One potential explanation for this difference is that the CIES includes only privately owned firms with sales above 5 million yuan, while the U.S. census data covers all manufacturing firms. Over 1998-2003, only 57 percent of entrants in the CIES are new firms, while the rest are existing firms whose sales rise above 5 million yuan. The measured 17 percent exit rate from the CIES is also biased upwards, since many exiters are continuing firms whose sales fall below 5 million yuan. To quantify the impact of this measurement bias, I redo the decomposition with several alternative exit rates in the CIES. Varying the exit rate from 8 percent to 5 percent implies a decreasing contribution of reallocation from an upper bound of 20 percent to 2 percent. Given that large firms have a lower exit rate than the 8 percent for all manufacturing firms, I conclude that reallocation in China is more distorted than in the U.S., but less than what the direct decomposition in the CIES suggests.
Intermediate Macroeconomics, Econometrics, PhD Econometrics (Teaching Assitant)
International Finance, International Finance Theory (PhD)