Can intermediate input trade liberalization affect worker health in a developing country like China, and if so, how? Do the impacts differ between skilled and unskilled workers? What are the welfare implications of input tariff reductions once health factors are considered? Professors Haichao Fan of Fudan University, Faqin Lin of China Agricultural University, and Shu Lin of the Chinese University of Hong Kong develop...

Recently, a growing strand of literature has started to examine the health effects of trade shocks (e.g., Hummels et al., 2016; McManus and Schaur, 2016; Pierce and Schott 2016; Bombardini and Li, 2020). While most existing work focuses on import competition or export expansion, less is known about the role of input tariffs. Using input tariffs and the China Health and Nutrition Survey data (CNHS), this study (Fan, Lin, and Lin, 2020) finds that input tariff reductions adversely affect worker health through increased working hours. Moreover, input tariff reductions widen both the income and the health gaps between skilled and unskilled workers. Further welfare analysis indicates that ignoring health outcomes would substantially underestimate the welfare disparity between skilled and unskilled workers.

Four Stylized Facts

We first document four stylized facts regarding input tariffs, workers’ working hours, wages, and health before and after China’s WTO accession:

Fact 4: After trade liberalization, unskilled workers experienced larger increases in illness rates and working time but a smaller gain in income.

A Theoretical Model

To rationalize these facts and to motivate our empirical exercises, we build an illustrative model linking input tariff reductions to firm behavior, workers’ supply of labor, and their health status. In our model, firms choose the number of workers and each worker’s optimal working time. The optimal working time and the number of employed workers are substitutes and are both decreasing functions of the wage rate. To hire workers, firms need to pay a search cost (e.g., Helpman and Itskhoki 2010; Helpman et al., 2010). The impact of a reduction in input tariffs on firms’ demand of working time can be illustrated intuitively using a simple diagram (Figure 1). A reduction in input tariffs lowers a firm’s marginal costs and increases its profit. The demand curve for working time is accordingly shifted to the right.

Figure 1. Optimal working hours and wage payment

Empirical Strategy

We then put the predictions of our theoretical model to a test. Following the literature (e.g., Amiti and Konings 2007; Topalova, 2010), we construct a prefecture-level input tariff shock measure using the Chinese provincial-level input-output table and the prefecture-level initial industry composition of labor. Employing the CHNS database, we estimate the following specification.

, where i and t represent the individual and year, respectively. measures the health condition of individual i in year t. Our main measure of workers’ health status is a binary variable that takes the value of 1 if an individual has experienced illness or injury in the past four weeks and 0 otherwise. We also consider two alternative measures for robustness. First, since the CHNS survey provides information about whether a worker has felt uncomfortable or exhausted in the past four weeks, we will use this binary variable as an alternative. Second, we also utilize workers’ subjective evaluations of the severity of illness or injury as another alternative. The self-evaluated score ranges from 1 (the least severe) to 3 (the most severe). In addition, we will also examine the effects on different types of illness using a set of binary indicators.

, the one-year lagged input tariff shock measure, is our variable of interest. We expect the estimated coefficient on this key variable to be negative and significant, which would indicate an adverse effect of input tariff reductions on worker health. We also include output tariffs in the regression and we expect the output tariff shock to have an ambiguous effect on worker health. On the one hand, the declines in output tariffs may lead to better worker health outcomes due to reduced working hours. On the other hand, there can also be some negative effects on worker health associated with unemployment (or underemployment), ranging from stress related to job loss to being unable to afford proper health care.

is a comprehensive set of individual-level controls, including gender, years of education, age, age squared, disease history, smoking, possession of health insurance, occupation type dummies, and employer ownership dummies. We also control for a set of prefecture-level covariates, , in our benchmark specification. Following Facchini et al. (2019), we include in a set of prefecture-level measures of other policy changes, including the elimination of trade uncertainty (Handley and Limao, 2017; Erten and Leight, 2019; Facchini et al., 2019) and export licenses (Bai et al., 2017), changes in quotas due to the expiration of global MFA, and changes in tariffs on China’s exports abroad. In addition, we also include in a prefecture-level measure of air quality to capture the effect of air pollution on health. is the year fixed effects, which captures yearly shocks common to all individuals. We also include prefecture fixed effects, , to control for all time-invariant differences across prefectures in our regressions. is the error term.

What Do We Find?

We find that manufacturing workers in Chinese prefectures that had a larger exposure to input tariff reduction shocks experienced a significantly higher likelihood of suffering from illness or injury. This finding is not affected by controlling for other trade shocks and pollution variables. It is also robust to using only the two waves of the survey data closest to China’s WTO accession. In a comprehensive set of tests, we also rule out the possibility that our results are driven by pre-existing trends. Additional evidence from the placebo tests lends further support to our hypotheses.

Table 1. Benchmark results

Table 2. Welfare disparity between skilled and unskilled workers (for a one standard deviation reduction in input tariff shock)

References

Amiti, Mary, and Jozef Konings. 2007. “Trade liberalization, intermediate inputs, and productivity: Evidence from Indonesia.” American Economic Review, 97: 1611–1638.

Erten, Bilge and Jessica Leight. 2019. “Exporting out of agriculture: The impact of WTO accession on structural transformation in China.” Review of Economics and Statistics, forthcoming.

Facchini, Giovanni, Maggie Liu, Anna Maria Mayda, and Minghai Zhou. 2019. “China’s “Great Migration”: The impact of the reduction in trade policy uncertainty.” Journal of International Economics, 120: 126–144.

Fan, Haichao., Faqin Lin, and Shu Lin. 2020. “The hidden cost of trade liberalization: Input tariff shocks and worker health in China.” Journal of International Economics, forthcoming.

van der Hulst, Monique, 2003. “Long workhours and health.” Scandinavian Journal of Work, Environment & Health, 29(3), 171–188.