The Determinants of China's Minimum Wage Rates

• Published date: 01 May 2023
• Author: Achim Schmillen,Michael Stops,Dewen Wang
• Date: 01 May 2023
• DOI: http://doi.org/10.1111/cwe.12489

©2023 Institute of World Economics and Politics, Chinese Academy of Social Sciences

China & World Economy / 59–91, Vol. 31, No. 3, 2023 59

The Determinants of China’s Minimum Wage Rates

Achim Schmillen, Michael Stops, Dewen Wang*

Abstract

We use a highly disaggregated panel of macro data and minimum wages at the county

level to investigate the processes behind minimum wage adjustments in China. Relying

on random eff ects models, spatial econometrics techniques, and multilevel analyses, we

document that a comparatively small number of economic variables – including the local

price level and GDP per capita – are important determinants of minimum wage rates.

Interactions between adjacent counties and counties of the same administrative type, and

centralized mechanisms, particularly at the provincial level, also play an important role

in explaining the variance in minimum wage rates across counties. Finally, we show that

China’s provinces are the key players for setting minimum wage rates and that, when

they do so, they are not uniform in the way they weigh diff erent economic variables.

Keywords: applied spatial econometrics, China, highly disaggregated macro data,

minimum wage setting, multilevel analysis, strategic interactions between subnational

governments

JEL codes: C23, J30, R23

I. Introduction

Since the seminal contribution by Card and Krueger (1994), a large body of literature

has studied the effects of minimum wage rates on wages, employment, and other

outcome variables.1 In contrast, the setting and determinants of minimum wages have

*Achim Schmillen, Practice Leader for Human Development, World Bank, Indonesia. Email: aschmillen@

worldbank.org; Michael Stops, Senior Researcher, Institute for Employment Research, Germany. Email:

michael.stops@iab.de; Dewen Wang (corresponding author), Senior Social Protection Economist, World Bank,

China. Email: dwang2@worldbank.org. The authors thank Mario Bossler, Tobias Haepp, Michael Murach,

Josefi na Posadas, Juliane Scheff el, Hans-Jörg Schmerer, and Helmut Wagner as well as participants at several

seminars and workshops for helpful comments and suggestions, and Xichen Li, Carl Lin, Chunyang Pan, and

Jin Song for graciously sharing relevant literature and data. Linghui (Jude) Zhu provided outstanding research

assistance and the Nordic Trust Fund provided funding support. Findings, interpretations, and conclusions

expressed in this paper are those of the authors and do not necessarily represent the views of the Nordic Trust

Fund, the World Bank, its affi liated organizations, its executive directors, or the governments these represent.

1For China, studies that explore the eff ects of minimum wages on wages, employment, and other outcome

variables include those by Huang et al. (2014), Lin and Yun (2016), and Démurger et al. (2021).

Achim Schmillen et al. / 59–91, Vol. 31, No. 3, 2023

©2023 Institute of World Economics and Politics, Chinese Academy of Social Sciences

60

received relatively little attention in the literature. Whitaker et al. (2012, p. 631) note

that “the fact that [minimum] wage increases are understudied is particularly puzzling

given the fi erce ideological confl ict the policy engenders.” To contribute to closing this

knowledge gap, we use panel data on county-level minimum wage rates and highly

disaggregated macro variables from 2005 to 2014 to investigate the processes behind

minimum wage adjustments in China.

China has no uniform nationwide minimum wage. Instead, the country’s Five-Year

Plans set general targets regarding the development of minimum wage levels, and the

Minimum Wage Regulations of 2004 (2004 Regulations) list a range of factors capturing

local economic or social conditions that provincial governments should consider when

setting their own minimum standards. The 2004 Regulations also give provincial

governments some discretion to set different minimum wages in different locations

within a province. This, in principle, could result in diff erent minimum wages for each

of China’s 2,843 county-level administrative divisions.

We exploit this cross-sectional variation to explore economic and noneconomic

infl uences underlying minimum wage setting in China with the help of random eff ects

models (REM). The REMs allow us to control for county-specifi c explanatory variables,

province-specific fixed effects, and time fixed effects. They correlate minimum wage

rates with proxies for the factors that are to be considered by provincial governments in

setting these rates according to the 2004 Regulations. We extract these variables from a

range of Chinese administrative data sources. In addition, we rely on spatial econometric

techniques to study the role of both central mechanisms and spatial dependence between

subnational governments. Spatial dependence might be an indication of strategic

interaction between subnational governments and is thus expected to play a prominent

role in determining minimum wages in a context like China’s, where provinces compete

with each other for investors or mobile workers. There are a plethora of newspaper

reports and other anecdotal evidence about China’s minimum wages being, first,

infl uenced by such considerations and, second, seen as a strong signal for the level of

local labor costs (Financial Times, 2010; Reuters, 2011; CBS News, 2014). Finally,

we employ multilevel analyses to assess the relative importance of two hierarchical

levels of subnational governments (the provincial and county levels) in explaining the

variance of minimum wage rates and to explore in more detail how China’s provincial

governments apply the legal rules on setting minimum wages rates spelled out in the

2004 Regulations.

We find, first, that the local price level and the GDP per capita are the most

important economic determinants of minimum wage rates. Localities with higher

prices and higher GDP per capita tend to exhibit higher minimum wage rates. Other

©2023 Institute of World Economics and Politics, Chinese Academy of Social Sciences

The Determinants of China’s Minimum Wage Rates 61

explanatory variables, like the average annual salary or the employment-to-population

ratio, have no signifi cant eff ects on minimum wage rates.

Second, in addition to the factors that are to be considered by provincial governments

in setting minimum wage rates, strategic interaction eff ects also have explanatory power.

We test two approaches: First, following Dreger et al. (2019) and Li et al. (2019), we

compare strategic interactions through the setting of minimum wages for neighboring

localities; in contrast with earlier studies, we do not compare minimum wages between

neighboring prefectures but between neighboring counties. Second, we examine

strategic interactions through the setting of minimum wages for counties of the same

administrative county types as a proxy for similar socio-economic conditions. Both

approaches indicate that strategic interaction explains an important share of the variance

of minimum wage rates across counties.2 This finding is consistent with competition

between provinces influencing their minimum wage-setting behavior, possibly in

addition to central adjustment mechanisms.3

Finally, as envisaged by the 2004 Regulations, the provincial level is of prime

importance in explaining the variance of minimum wage rates in China. Our analysis

also indicates that provinces diff er in how they weigh diff erent economic variables when

setting minimum wage rates.

This study contributes to the literature on the setting and determination of minimum

wages. Relevant theoretical contributions include Hungerbühler and Lehmann (2009)

and Lee and Saez (2012). Hungerbühler and Lehmann (2009) argue that a minimum

wage can be optimal if workers’ bargaining power is relatively low in a framework

where search frictions generate unemployment. Lee and Saez (2012) show that even

in a perfectly competitive labor market, a binding minimum wage – while leading to

unemployment – can nevertheless be desirable if one values redistribution toward low-

wage workers.

In reality, most governments change minimum wages not based on rigorous

economic theory but by following rules of thumb, engaging in (formal or informal)

bargaining with employers and labor unions, and considering the country’s context and

certain economic or noneconomic criteria. The number of empirical studies evaluating

the exact processes underlying minimum wage changes is very limited. This is because

2This fi nding is validated by a sensitivity analysis based on arbitrarily chosen spatial dependence structures,

which results in estimates of the strategic interaction coeffi cients that are not statistically signifi cant.

3All our specifications include year dummies. Coefficient estimates for these are all positive and

signifi cant and reveal a positive time trend, even when we control for local economic conditions. This

points toward central adjustment mechanisms being at least partly responsible for positive and gradual

rate adjustments.