Effects of the Resident Basic Medical Insurance Reform on Household Consumption in China

• Published date: 01 January 2024
• Author: Shuheng Yu,Xinxin Ma,Peng Zhan
• Date: 01 January 2024
• DOI: http://doi.org/10.1111/cwe.12517

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

China & World Economy / 96–129, Vol. 32, No. 1, 2024

96

Ef‌f ects of the Resident Basic Medical Insurance

Reform on Household Consumption in China

Shuheng Yu, Xinxin Ma, Peng Zhan*

Abstract

This paper investigates the causal relationship between the Urban and Rural Resident

Basic Medical Insurance (URRBMI) reform and household consumption in urban China

using data from the Chinese Household Income Project survey and employing combination

of the propensity score matching and difference-in-differences methods. The results yield

three conclusions. First, the reform affected both the amount and share of household

consumption. Specifically, the reform led to an increase in total household consumption,

encompassing both medical and nonmedical expenditures. Among these, the proportion

of medical consumption increased, while the proportion of nonmedical consumption

reduced. Second, the impact of the URRBMI reform varied across consumption categories

within nonmedical expenditure. Specifically, consumption for education and entertainment

industries are positively impacted by the URRBMI reform in terms of both quantity

and proportion. Third, low-income households benefit more from the URRBMI reform

compared to middle- and high-income households. The main channels through which the

URRBMI reform affects household consumption were the price effect and the crowding out

effect on precautionary savings.

Keywords: household consumption, nonmedical consumption, out-of-pocket expenses,

reimbursement ratio

JEL codes: D12, H53, I13

I. Introduction

With its aging population, the establishment and reform of social security systems has

become an important issue in China. Since the 1990s, the Chinese government has

*Shuheng Yu, PhD Candidate, School of Public Af‌f airs, Institute for Common Prosperity and Development,

Zhejiang University, China. Email: yushuheng@zju.edu.cn; Xinxin Ma, Professor, Faculty of Economics,

Hosei University, Japan. Email: xxma@hosei.ac.jp; Peng Zhan (corresponding author), School of Public

Af‌f airs, Institute for Common Prosperity and Development, Zhejiang University, China. Email: zhanpeng@

zju.edu.cn. The authors are grateful for support from the National Social Science Foundation of China (No.

19ZDA116), the Japan Society for the Promotion of Science (No. 20H01512) and the Fundamental Research

Funds for the Central Universities.

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

Medical Insurance Reform and Consumption 97

implemented several public medical insurance reforms, including the introduction of

three types of medical insurance: Urban Employee Basic Medical Insurance (UEBMI),

Urban Resident Basic Medical Insurance (URBMI), and the New Rural Cooperative

Medical Insurance Scheme (NRCMS). The reforms were intended to enforce the

integration of the URBMI and the NRCMS into the Urban and Rural Resident Basic

Medical Insurance (URRBMI), and greatly increased the reimbursement ratios of

URRBMI. Specifically, the reform stipulated that the ratio of reimbursement for

inpatient expenses would increase to 75 percent, and the ratio of reimbursement for

outpatient expenses would gradually increase. Huang and Wu (2020) and Feng et al.

(2022) indicated that the URRBMI reform has signif‌i cantly improved the reimbursement

ratio and efficiency. As URRBMI implementation differs by region, the URRBMI

reform can be utilized as a quasi-natural experiment. Using two iterations (2013 and

2018) of the Chinese Household Income Project (CHIP) survey, this study investigates

the causal relationship between URRBMI reform and household consumption

in China.

From an economic theory perspective, three channels can explain the effects of

medical insurance on household consumption. First, medical insurance may crowd

out precautionary savings by reducing the risk of higher medical expenses associated

with illness (Gruber and Yelowitz, 1999; Chou et al., 2003; Chou et al., 2004; Kuan

and Chen, 2013) (the crowding out ef‌f ect on precautionary savings). Households may

consequently adjust their levels of medical and nonmedical consumption (Sheu and

Lu, 2014). Second, medical insurance can reduce medical care service costs through

payment regulations (changing the price channel). For example, an increase in the

reimbursement ratio or a decline in the share of out-of-pocket (OOP) expenses may

induce a greater utilization of medical services and increase household medical expenses

(Manning et al., 1987; Baker et al., 2001; Ameyaw et al., 2017). Moral hazard may

also encourage overconsumption of medical services, which increases total household

consumption and accordingly impacts consumption patterns (e.g., by changing the ratio

of medical and nonmedical consumption) (Feldstein, 1973). Third, medical insurance

premium payments decrease disposable income, which may reduce consumption due to

income constraints (i.e., the insurance contribution payment ef‌f ect). As these channels

could either increase or decrease household consumption, the full effects of medical

insurance on consumption remain unclear from an economic theory perspective and

deserve to be investigated empirically.

Regarding existing empirical studies on the issue, He and Shi (2014) indicated

that medical insurance might crowd out precautionary savings and promote household

consumption in China. Meanwhile, Song and Song (2018) revealed that increases in

Shuheng Yu et al. / 96–129, Vol. 32, No. 1, 2024

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

98

medical security significantly affected the household savings rate among low-income

groups. Regarding previous studies on household consumption, empirical research has

also confirmed that social insurance positively affects household consumption. For

instance, Gruber and Yelowitz (1999) found that basic medical insurance (Medicaid)

for low-income groups in the US, which expanded signif‌i cantly in the 1980s and 1990s,

increased consumption by 5.2 percent. Wagstaf‌f and Pradhan (2005) identif‌i ed that, in

Vietnam, basic medical insurance increased household consumption significantly and

that this ef‌f ect was the greatest for nonfood consumption.

The present study contributes to the existing literature in three ways. First,

although numerous studies have investigated the effects of medical insurance on

medical service utilization (Wagstaff et al., 2007; Ma, 2022), OOP medical expenses

(Lindelow and Wagstaff, 2005; Wagstaff et al., 2007; Huang and Gan, 2010), total

household consumption (Gruber and Yelowitz, 1999; Wagstaff and Pradhan, 2005;

Zang et al., 2012), and the crowding out effect associated with precautionary saving

(Chou et al., 2004; Bai et al., 2012; Kuan and Chen, 2013) in China and other countries,

scant empirical research has been conducted to demonstrate either how medical

insurance reforms impact medical and nonmedical consumption or how such reforms

impact particular types of nonmedical consumption (e.g., food, education, or housing

consumption). To the best of our knowledge, only Sheu and Lu (2014) have focused

on this issue; however, their study was set in Chinese Taiwan province. The present

study is the f‌i rst to focus on this issue on the Chinese mainland. Specif‌i cally, this study

investigates the effects of reforms to the Resident Basic Medical Insurance (RBMI)

program on medical and nonmedical consumption on the Chinese mainland, which has

one of the world’s largest populations and consumption markets.

Second, the existing literature has not suf‌f‌i ciently addressed endogeneity concerns,

the present study makes advancements in addressing these issues. For instance, Sheu

and Lu (2014) did not consider the endogeneity issue caused by the nonrandom selection

resulting from participation in public medical insurance programs. Specif‌i cally, some

observable and unobservable variables may af‌f ect the likelihood of participation. This

study uses a combination of propensity score matching (PSM) and the difference-in-

dif‌f erences (DID) method, enabling more robust evidence on this issue.

Third, Sheu and Lu’s (2014) study was set shortly after a new public medical

insurance program had been implemented. Meanwhile, the present study investigates

whether increases in the reimbursement ratio impact consumption against the

background of universal insurance (since 2011, China’s public medical insurance

has covered almost its entire population). Compared to studying the impact of

establishing new institutions, studying institutional reforms can help us gain a deeper