Long‐term Effect of Export Expansion on Human Capital

• Published date: 01 November 2024
• Author: Jintao Fu
• Date: 01 November 2024
• DOI: http://doi.org/10.1111/cwe.12556

©2024 Institute of World Economics and Politics, Chinese Academy of Social Sciences

China & World Economy / 1–34, Vol. 32, No. 6, 2024 1

Long-term Ef‌f ect of Export Expansion on

Human Capital

Jintao Fu*

Abstract

This paper illustrates how trade liberalization during an individual’s early life can

affect long-term health and cognitive outcomes 15 years later (i.e., in adolescence).

China’s accession to the World Trade Organization has brought significant economic

benefits but has also given rise to some environmental concerns. Exploiting variations

in each prefecture’s exposure to trade shock and trade-induced pollution shock, this

study constructed a shift-share instrument variable model. The trade shocks in early life

improved health and cognitive outcomes significantly during adolescence while trade-

induced pollution shock had the opposite effect, decreasing these two outcomes. These

impacts were more pronounced among households with low economic status. Further

investigation provided evidence for several underlying mechanisms, such as improved

health and education resources and reduced SO2 emissions by firms. This study provides

useful insights into how to evaluate the long-term effects of trade liberalization on

human capital in terms of economic benefits and environmental costs in China.

Keywords: health and cognitive, human capital, trade-induced pollution, trade liberalization

JEL codes: F16, F64, Q50

I. Introduction

How does trade liberalization affect household welfare? There is evidence that the

economic windfall induced by trade expansion has increased regional productivity,

wages, employment, and GDP per capita significantly (Lu and Yu, 2015; Han et al.,

2016; Erten and Leight, 2021). However, these economic benefits have come with

environmental costs. Exten sive literature documents that the pollution caused by trade

expansion has increased infant mortality, has damaged workers’ health status, and has

reduced human capital accumulation (Li et al., 2019; Bombardini and Li, 2020; Fan et al.,

2020; Lin and Long, 2020). Although these studies provide causal links between trade

*Jintao Fu, PhD Candidate, Department of Economics, University of Ottawa, Canada. Email: jfu041@uottawa.ca.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit

sectors. I thank my advisor Nicholas Rivers for valuable comments and guidance. I am grateful to participants at

the Canadian Resource and Environment Economics Association conference 2023 for helpful discussions.

Jintao Fu / 1–34, Vol. 32, No. 6, 2024

©2024 Institute of World Economics and Politics, Chinese Academy of Social Sciences

2

liberalization and contemporaneous measures of economic and environmental well-

being, there is less evidence regarding the effects of trade expansion effects over a longer

period. This paper examines the impact of early life exposure to trade liberalization

on health and cognitive outcomes approximately 15 years later, specifically during

adolescence, within the context of China’s accession to the World Trade Organization

(WTO). The focal point of this study is China’s expansive trade activities, given its

status as the largest manufacturing economy globally and the presence of relatively high

pollution levels. China’s trade policies and export expansions have yielded substantial

economic and productivity gains but they have also been accompanied by environmental

challenges. Comprehensive evidence on firms, households, and regional development

is essential to evaluate the benefits and costs of trade-induced pollution. On the one

hand, China’s WTO accession has led to heightened emissions from firms and elevated

regional pollution levels, as documented in previous research (Bombardini and

Li, 2020; Chen et al., 2020), potentially detrimentally impacting individual well-being.

On the other hand, recent investigations have indicated that trade liberalization can

contribute to a reduction in firms’ emissions (Cui et al., 2020; Rodrigue et al., 2022;

Lin et al., 2024). Moreover, the trade shock exerts both positive and negative impacts on

human capital accumulation as well as household income (Fan et al., 2020). The mixed

evidence underscores the need to analyze how trade shocks and trade-induced pollution

may affect long-term outcomes. Indeed, this is crucial for policy evaluations.

This study extended the work of Bombardini and Li (2020) by investigating the

enduring ramifications of trade expansion on health and cognitive test performance,

examining both income channel of trade shocks and trade-induced pollution shocks. To

ascertain the causal influence of these shocks, a shift-share (Bartik-style) instrumental

variable (IV) for trade shock was constructed, leveraging foreign countries’ tariffs on

Chinese export goods to estimate export flows. Each prefecture’s exposure to trade

shock was then determined by weighting the estimated export flows according to its

initial employment share, employing an approach akin to methodologies employed in

prior studies (Bombardini and Li, 2020). In assessing trade-induced pollution shock,

industry-specific SO2 emission intensity was interacted with the trade shock. Health

outcomes and cognitive test scores were evaluated using data from the China Family

Panel Studies (CFPS), spanning from 2010 to 2020, tracing the effects into later

adolescence across seven birth cohorts from 1999 to 2005.

The study incorporated measures of income channel of trade shock and trade-induced

pollution shock. Prefecture-level income channel of trade shock measures the extent

to which a prefecture initially specialized in industries that subsequently experienced

substantial export growth. This measure was constructed using a shift-share framework,

©2024 Institute of World Economics and Politics, Chinese Academy of Social Sciences

Long-term Ef‌f ect of Export Expansion on Human Capital 3

which combined the initial industry shares with changes in trade flows over time. The

trade-induced pollution shock incorporated each industry’s emission intensity into

the trade shock measure. This captured the interaction between export expansion and

pollution intensity: prefectures with a higher initial employment share in industries facing

significant trade shocks and displaying high emission intensity were expected to encounter

elevated pollution levels. In the empirical analysis, this paper employed regression models

to examine the relationship between early life exposure to the two measures and a range of

long-term human capital indicators. This enabled exploration of the coefficients associated

with each measure and their implications for human capital development over time.

One possible danger when attempting to investigate the coefficients in this way is

that trade shocks within each prefecture could influence households’ behavior changes

in ways that are not captured in my specification. Simultaneity arises from the fact

that export flows are not only influenced by regional economic conditions like GDP

growth but also have feedback effects on these variables. This mutual determination can

complicate the analysis of export flows and their impact on regional welfare.

Omitted variable bias could occur if important factors affecting both export flows

and regional welfare, such as productivity improvements, are not included in the

analysis. This omission could lead to biased estimates of the relationship between

export flows and welfare outcomes. To address potential endogeneity, I constructed

an IV for Chinese export flows by leveraging the tariffs reported by the rest of the

world. By regressing actual export flows on these reported tariffs, I estimated the fitted

value of Chinese export flows. The fundamental assumption here posits that the tariffs

imposed by foreign countries do not influence domestic household behaviors within

China directly. Consequently, the estimated fitted value of export flows varies across

prefectures in China, reflecting regional disparities in trade exposure.

Another empirical challenge is the establishment of a link between childhood

circumstances and later life outcomes. This requires detailed household information and

insights into migration status. Leveraging the comprehensive household data available in the

CFPS dataset, I narrowed the sample to households that remained in their original prefectures.1

There were substantial effects of exposure to trade expansion during gestation and

early childhood on adolescent health and cognitive development. Specifically, exposure

to trade shock correlated with a 0.025 standard deviation reduction in the unhealthiness

index, which comprised self-reported health, hospitalization, and underweight indicators,

alongside a 0.022 standard deviation increase in the cognitive index, encompassing

1The causal effects on human capital accumulation can be identified by restricting the sample to exclude

selective migration. For example, if children’s parents with higher education levels are more likely to migrate

to regions with trade liberalization, I would capture education rather than trade liberalization.