"Identifying Chinese Supply Shocks - Effects of Trade on Labor Markets" (joint with Andreas M Fischer and Philipp Herkenhoff)

Abstract

In a seminal paper, Autor et al. (2013a) analyze the effect of Chinese exports on U.S. employmentvariables. To identify causality, Chinese exports to the United States are instrumented withChinese exports to other advanced economies, an identification strategy that relies on the absenceof common demand shocks to all advanced economies. Our paper first questions this identificationassumption and then adjusts the reduced-form identification strategy in Autor et al. (2013a). In thefirst part, we document that China’s sector-level export growth correlates positively with that ofother emerging market economies around and after China’s WTO accession. This positivecorrelation is inconsistent with the view that China-specific supply shocks dominated China’ssectorial export growth. In a second part, we develop an identification strategy derived from aparsimonious structural model. Our findings validate the estimated negative effects of Chineseimport penetration on U.S. labor markets in Autor et al. (2013a). However, important differencesarise when decomposing the China-specific supply shock effects by gender, education, and (non)manufacturing sector.