Selected research papers (September 2026)

JUE Insight: Soft Infrastructure and the Location Choice of Multinational Firms: Evidence from Japanese Investment in the United States in the 1980s (with Eric Olson and Haoyu Wang, Journal of Urban Economics)

(Abstract) We examine whether culture-specific educational infrastructure influenced the location decisions of multinational enterprises by analyzing Japanese foreign direct investment (FDI) in the United States over the 1980s. Using the expansion of Japanese FDI following Reagan-era trade tensions, we test whether preexisting Japanese Studies programs at U.S. universities predict subsequent Japanese investment across 722 commuting zones. We find that zones with Japanese Studies programs in 1975 were approximately 25 percentage points more likely to receive Japanese investment in 1992, controlling for traditional determinants like Japanese population, manufacturing infrastructure, agglomeration, and market access. The results are robust to alternative empirical methods, including the instrumental variable approach. Our results suggest that soft infrastructure--culture-specific human capital pipelines--represents an overlooked location advantage in attracting FDI.

Bankruptcy, Ownership Turnover and the Fate of U.S. Coal Mines (with Sara E. Guffey, Resource and Energy Economics)

(Abstract) The rapid decline of the U.S. coal industry has triggered a wave of bankruptcies in which environmental liabilities such as mine reclamation and water treatment are often left unfunded. Bankruptcy courts frequently relieve firms of these legal obligations to facilitate mine transfers to new owners, aiming to preserve local operations and limit the economic impacts of closures. This study examines whether such transfers sustain mining activity. Using mine-level data from the U.S. Mine Safety and Health Administration (MSHA) on production and ownership turnover, we estimate the effect of ownership turnover on mine survival, distinguishing between court-supervised and private deals. We find that ownership turnover substantially increases the probability of closure within a few years, regardless of court involvement. Our findings indicate that even bankruptcy procedures do not reverse structural decline and may shift environmental costs to local communities.

The Growth of Firms, Markets and Rents: Evidence from China (with Daniel Berkowitz, Journal of Comparative Economics, 52(2), 383-399, 2024)

(Abstract) Using recent methods for estimating firm-level markups and profit shares, we document that Chinese manufacturing firms collected more rents following China's accession to the World Trade Organization (WTO). This is because the net entry of firms lagged the massive growth in the domestic market. These effects were particularly strong in domestic markets where state ownership was pervasive. While selection on large productive firms drove the rise in the aggregate markups in the United State (De Loecker et al, 2020), these competitive forces played a secondary role in Chinese manufacturing.

Political Regimes and Firms' Decisions to Pay Bribes: Theory and Evidence from Firm-level Surveys (with Sumi Sharma and Tuan Le, Journal of Institutional Economics, 19(6), 764-786, 2023)

(Abstract) This paper makes the most of the observed actions of bribe takers and givers from the World Bank Enterprise Surveys and studies how a taker's action influences a giver's decision to pay bribes. To motivate our empirical study, we consider Kaufmann and Wei's (1999) Stackelberg game between a tax authority and a firm that undergoes tax inspection. The model predicts that, when the authority can use its action as a credible threat for the firm's profitability, the authority disturbs the firm by inspecting more, and the firm is more likely to pay bribes. Consistent with the theoretical prediction, we find correlational evidence that the propensity to pay bribes increases with the number of inspection visits, particularly for non-democratic countries.

The Political Effects of Trade with Japan in the 1980s (with Eric Olson, Economic Inquiry, 61(2), 451-471, 2023)

(Abstract) The 1974 trade act substantially increased the executive branch's authority in trade negotiations through the granting of fast-track and Section 301 authority. This paper evaluates the effect on U.S. voting behavior resulting from trade with Japan over 1976-1992 time period after the act was passed. To capture U.S. trade exposures to Japan, we develop the Bartik index from Autor et al (2013) for import competition with Japan and show that local exposure to import competition had statistically significant negative impacts on Republican presidential candidates over the 1976-1984 period. Although the second Reagan administration used Section 301 to open Japan's markets and Japanese firms shifted production to the United States, job-creation effects of exports and foreign direct investment did not have any influence on voting outcomes.

Recasting the Iron Rice Bowl: The Reform of China's State Owned Enterprises

(with Daniel Berkowitz and Hong Ma, Review of Economics and Statistics, 99(4), 735-747, 2017)

(Abstract) Following the enactment of reforms in the mid-1990s China's state owned enterprises (SOEs) became more profitable. Using theoretical insights from Azmat, Manning and Van Reenen (2012) and Karabarbounis and Neiman (2014) and econometric methods in De Loecker and Warzynski (2012) this paper finds that SOE restructuring was nevertheless limited. SOEs became more profitable because their cost of capital fell and their capital-labor elasticity of substitution generally exceeded unity, and also because they were under less political pressure to hire excess labor. Moreover, SOE productivity lagged foreign and private firms.

International Differences in Production Techniques: Implications for the Factor Content of Trade

(Journal of International Economics, 87, 2012: p98-104)

(Abstract) This paper examines how production techniques differ across countries, factors, and industries and considers its implications for previous empirical evidence on the Vanek prediction. I find that production techniques differ substantially across countries and factors, but differ much less across industries within a country. Davis and Weinstein (2001) argue that modeling cross-industry differences (multiple-cone specialization) improves the fit of the Vanek prediction; however, their test statistics are unchanged when one restricts techniques to be identical across industries within a country. Thus, the bulk of world factor content of trade does not arise from specialization.

International Differences in Emissions Intensity and Emissions Content of Global Trade

(with Stratford Douglas, Journal of Development Economics, 99, 2012: p415-427)

(Abstract) Understanding international differences in the emissions intensity of trade and production is essential to understanding the effects of greenhouse gas limitation policies. We develop data on emissions from 41 industrial sectors in 39 countries and estimate the CO2 emissions intensity of production and trade. We find no evidence that developing countries specialize in emissions-intensive sectors; instead, our evidence suggests emissions intensities differ systematically across countries because of differences in production techniques. Our results confirm that international differences in emissions intensity are substantial, but suggest that they do not play a significant factor in determining patterns of trade.

Productivity, Trade and the R&D Content of Intermediate Inputs

(with Marla Ripoll, European Economic Review, 56, 2012, p1573-1592)

(Abstract) This paper explores a novel way to evaluate the extent to which R&D knowledge embodied in intermediate inputs correlates with productivity at the industry level. We propose the concept of the R&D content of intermediates, which represents the knowledge stock embodied in intermediate inputs used in production. Using a sample of 32 countries and 13 manufacturing industries we compute the elasticity of industry-level TFP with respect to the R&D content of intermediates. We find that among high-R&D industries, the R&D embodied in inputs purchased from the own industry is significantly associated with industry-level TFP. In this case, both own-industry domestic inputs as well as those imported from G5 countries are relevant. In contrast, intermediate input trade does not appear to be a significant channel of R&D diffusion among low-R&D industries.