Ownership restructuring and wage inequality in urban China

• Published date: 01 March 2016
• Author: Chunbing XING,John WHALLEY
• DOI: http://doi.org/10.1111/ilr.12005
• Date: 01 March 2016

International Labour Review, Vol. 155 (2016), No. 1

Copyright © The authors 2016

Journal compilation © International Labour Organization 2016

Ownership restructuring

and wage inequality in urban China

John WHALLEY* and Chunbing XING**

Abstract. Urban income inequality is an increasingly important driver of China’s

overall income inequality. Using three urban household surveys for 1995, 20 02

and 200 7, the authors examine how change in enterprise ownership structure – one

of the major features of China’s economic transition – has contributed to widen-

ing urban wage inequality. While wage dispersion was higher in the private sector

over the period 1995–2 007, it increased faster in the public sector. However, over

50 per cent of the increase in urban wage inequality is associated with labour re-

allocation from the public to the private sector. Urban wage inequality may thus

widen further if more labour is reallocated to the private sector.

A

s a collectivist socialist economy, China was very egalitarian. In the early

1980s, when economic reform was in its early stages, it had a Gini coef-

cient of around 0.3 (World Bank, 2009). As its economy grew rapidly in the

transition process, however, its income inequality widened. By 2015, its Gini

coefcient had reached 0.462 (National Bureau of Statistics of China, 2016).

Because of the dualistic nature of China’s economy, the rural–urban in-

come gap still contributes in a major way to overall income inequality (Li and

Sicular, 2014).

1

Yet urban inequality is also becoming an increasingly important

component of China’s overall income inequality, as urban income inequality

rises and more and more rural residents migrate to urban areas. According to

Meng, Shen and Sen (2013), the urban Gini coefcient was around 0.23 in the

late 1980s but increased to about 0.37 in 20 09 (see, for example, Ravallion and

Chen, 2007; Li and Sicular, 2014). As urbanization proceeds, urban inequality

will play an even more important role in overall inequality.

* University of Western Ontario, Center for International Governance Innovation (CIGI),

and National Bureau of Economic Research, email: jwhalley@uwo.ca. ** Beijing Normal Univer-

sity, email: xingchb@bnu.edu.cn. The authors wish to express their gratitude to the Ontario Research

Fund for support, and to Audra Bowlus, Chris Robinson, Terry Sicular and a seminar group at the

University of Western Ontario for their comments on an earlier draft of this article.

Responsibility for opinions expressed in signed articles rests solely with their authors, and

publication does not constitute an endorsement by the ILO.

1 The ratio of urban to rural per capita household disposable income was 3.33:1 in 2009, though

it declined to 2.92:1 in 2014. See http://www.ce.cn/xwzx/gnsz/gdxw/201501/20/t201 50120_438 4230.

shtml [accessed 25 January 2016].

International Labour Review58

The literature identies rising skill price as the major reason for the in-

crease in wage inequality (Meng, Shen and Sen, 2013; Park et al., 2008). Yet,

while some research indicates that institutional change was the main driver

of the signicant increase in the skill price (Zhang et al., 2005; Li and Ding,

2003), little attention has been paid to the effect of institutional changes on

wage inequality, including that of changes in ownership structure.

Also emphasized in the literature are the differences in wage-setting be-

haviour under different types of ownership. For example, Zhao (2002), using

data for 1996, nds that unskilled workers in foreign-invested enterprises (FIE)

earned signicantly less than those in state-owned enterprises (SOEs), but that

skilled workers earned more in FIEs than in state enterprises. Chen, Démurger

and Fournier (2005) also nd that FIEs and collective enterprises offered higher

returns to education than state-owned enterprises. Similar ndings also are re-

ported by Meng (2000), Xing (200 8), and Yang, Démurger and Li (2010).2

This article investigates how ownership restructuring has contributed

to China’s rising urban wage inequality. Using urban household surveys for

1995, 2002 and 20 07, we rst document the private-sector’s higher wage dis-

persion and skill price, compared to the public sector, but we also show that

both wage dispersion and skill price are increasing faster in the public sector.

Decomposition exercises show that the privatization process – i.e. the reallo-

cation of labour from the public to the private sector – has played a major

role in increasing urban wage inequality. Yet the increases in wage dispersion

within different types of ownership (particularly in the public sector) also con-

tribute signicantly. Counterfactual analyses indicate that urban wage inequal

-

ity may continue rising if more labour were to be allocated to the private

sector in the future.

The remainder of the article is organized into six sections. The rst pre-

sents some background on ownership restructuring and discusses the chan-

nels through which ownership restructuring can inuence wage inequality. The

second section presents our data, and the third reports a variance decompo-

sition analysis aimed at quantifying the effect of ownership restructuring on

wage inequality. The fourth section provides wage equation estimates by type

of ownership and by year, which enable us to evaluate whether self-selection

is a serious issue in changing wage structures. The fth section reports coun-

terfactual analyses, and the sixth section concludes.

Background: Enterprise ownership restructuring

in China

There are many types of enterprises in China: state-owned, collective, individ-

ual-owned, foreign-invested, joint-ventured or share-holding. Before economic

reform began in the early 1980s, almost all of the labour force was employed

2 Dong and Bowles (2002) is an exception: they nd that there is no signicant difference

in the returns to education between rms under different types of ownership.