Impacts of enterprise zones on local households in Vietnam

• Published date: 01 December 2023
• Author: Tien Manh Vu,Hiroyuki Yamada
• Date: 01 December 2023
• DOI: http://doi.org/10.1111/1468-0106.12432

ORIGINAL ARTICLE

Impacts of enterprise zones on local households

in Vietnam

Tien Manh Vu

1

| Hiroyuki Yamada

2

1

Faculty of Global Management, Chuo

University, Tokyo, Japan

2

Faculty of Economics, Keio University,

Tokyo, Japan

Correspondence

Tien Manh Vu, Faculty of Global

Management, Chuo University. 742-1

Higashinakano, Hachioji city, Tokyo

192-0393, Japan.

Email: mvu264@g.chuo-u.ac.jp

Funding information

Japan Society for the Promotion of

Science, Grant/Award Numbers:

22K18534, 21K01455, 20H01506,

19H00619, 18K01580, 18K12784

Abstract

We examine the possible impacts of enterprise zones

(EZs) on local Vietnamese households between 2002

and 2008, using differences-in-differences and a panel-

event study. We layer four waves of household surveys

using a census of EZs in 2007, based on the same com-

mune identity for our household and individual ana-

lyses. Within 5 years of EZ establishment, we find they

are associated with higher household incomes, an

increase in private property prices, and an increase in

working hours. However, we do not find a significant

impact on household living expenditure or school atten-

dance/working probabilities among members aged

between 7 and 17 years. Neither do we find a significant

impact on health outcomes.

KEYWORDS

enterprise zone, health, household, income, school attendance,

Vietnam

JEL CLASSIFICATION

O12, O18, D1, P36

1|INTRODUCTION

Although the term has appeared throughout the history of development in various forms, such

as industrial zone/park/cluster, (special) economic zone, export-processing zone and so forth

(see Akinci & Crittle, 2008 for the variations in history), a zone is often considered an area with

defined boundaries with specific policies, typically referred to as place-based policies. We focus

on zones mainly to attract businesses and firms and refer to these as enterprise zones (EZs).

Received: 17 May 2023 Accepted: 17 May 2023

DOI: 10.1111/1468-0106.12432

Pac Econ Rev. 2023;28:593–612. wileyonlinelibrary.com/journal/paer ©2023 John Wiley & Sons Australia, Ltd 593

The EZ is designed for specific policy objectives such as free trade, export manufacturing,

regional development, and policy experiments.

More specifically, this study examines the effect of Vietnamese EZs on several outcomes

regarding households and individuals, namely household income per capita, household living

expenditure per capita, individual labour force participation and school attendance probability

by age cohort. In addition, we test the pollution haven hypothesis of whether the establishment

of EZs is associated with more health issues and loss proxied by absent days from routine work.

Our research is motivated by several important facts and existing literature as follows.

First, whether EZs impact household income is debatable. Papke (1994), Ham et al. (2011),

Ali and Peerlings (2012), Busso et al. (2013), Chaurey (2017) and Zheng et al. (2017) report

higher incomes due to clustered firms in both developed and developing countries. However,

Neumark and Kolko (2010) document that the EZ programme in California does not reduce

unemployment. Reynolds and Rohlin (2015) find limited evidence to show that the Federal

Empowerment Zone programme in the United States aided underprivileged individuals. Mean-

while, if agglomeration economies exist along with improved infrastructure as a result of EZs,

quality of life could increase. This could lead to a higher cost of living due to the influx of

demand and labour mobility. Therefore, household income increases, and whether EZs improve

the real net household disposable income, is not a direct result.

Second, increased income and the rising demand for labour, if any, would complicate

household or individual decisions regarding investment in education. Parents might invest

more in their children's education, either in terms of quantity or quality, due to the wealth

effect and increased returns to education (Edmonds & Pavcnik, 2005; Glewwe & Jacoby, 2004;

Glewwe & Patrinos, 1999). However, since parents can suddenly become full-time employees,

intra-household work of children may increase. Therefore, children may have to compensate

(Basu & Van, 1998), as a substitution effect. This would affect children's education (Beegle

et al., 2009) in terms of school absenteeism or dropping out of school altogether. Similarly, if

job opportunities surge with the possibility of higher incomes, parents may pressure their chil-

dren to discontinue their education and start working, especially if the parents have to pay for

their children's education and it is very expensive.

Third, Vietnamese EZs boom during 2000 and 2007, as illustrated in Figure 1. Newly

established EZs during this period are four times more compared to all previous existing EZs

combined. The literature for environmental standards for EZs were homogeneous nationwide.

However, with the intense outbreak of EZs and to compete with other regions, local govern-

ment possibly relaxed law enforcement and monitoring efforts to attract new firms and inves-

tors. For example, during the Formosa incident in Vietnam in 2016, Hoang et al. (2020) reveals

that a single firm in the Vung Ang Economic Zone discharged toxic wastewater into the ocean.

This resulted in an immediate loss of an estimated 42–46% of revenue suffered by the fishing

industry in the affected area (the sector accounted for 3.8% of employment and 7.3% of income

in coastal areas).

We analyse Vietnamese households and individuals using the Household Living Standard

Survey (2002, 2004, 2006 and 2008) and the 2007 Establishment Census (containing a census of

EZs). We combine the two main sources through communal identity. We select districts

1

with-

out any EZs before 2003. We select communes with EZs as the treatment group, and communes

without EZs that are located in the same district, as the control group until 2008. However, dis-

tricts containing both groups should not have any zones established prior to 2003. We compare

households in the two groups using both differences-in-differences (DID) and a panel-event

study approach. Our findings indicate that household income per capita is higher in the

594 VU and YAMADA