How does labour share respond to risk? Theory and evidence from the Chinese industrial sector

• Published date: 01 June 2020
• DOI: http://doi.org/10.1111/ilr.12111
• Date: 01 June 2020

Copyright © The authors 2020

Journal compilation © International Labour Organization 2020

* National Academy of Development and Strategy, Renmin University of China, email: zou_

jingxian@163.com. ** Lingnan College, Sun Yat-sen University, email: hnshgj@126.com (correspond-

ing author). *** Development Research Centre of the State Council, email: js@drc.gov.cn. The au-

thors would like to thank two anonymous referees for their comments. This study was supported

by the Fundamental Research Funds for the Central Universities and the Research Funds of

Renmin University of China (grant No. 19XNB011).

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

publication does not constitute an endorsement by the ILO.

International Labour Review, Vol. 159 (2020), No. 2

How does labour share

respond to risk?

Theory and evidence from

the Chinese industrial sector

Jingxian ZOU,* Guangjun SHEN ** and Shen JIA***

Abstract. This study discusses the role of rm risk in the declining labour share

in China. Based on the model developed by Holmström and Milgrom (1987), the

authors demonstrate that lower rm risk can motivate workers to work harder,

leading to higher output per worker and average wage. However, increased output

will lower the labour share. Using data from the Chinese Industrial Enterprises

Database for the period 1998–2007 and the World Bank’s Investment Climate

Survey 2005 , empirical evidence supports this hypothesis and performs robustly

across various model specications and proxies for rm risk, indicating a positive

correlation between labour share and rm risk.

Keywords: labour share, rm risk, contract theory, industrial sector, China.

1. Introduction

The distribution of national income between capital and labour has always been

of interest to economists and policy-makers. Since the “Kaldor Facts” were rst

put forward (Kaldor, 1957), it has long been claimed that, according to neoclas-

sical economics, the share of national income going to factors of production

should remain constant. However, this claim has been challenged by empirical

evidence documenting drastic changes in labour income share. This evidence

points to a signicant decline in the labour share of developed countries since

the early 1980s, especially in continental European countries (Blanchard, 19 97;

Harrison, 2005; Guscina, 2 006). Although not as pronounced as in Europe, a dis-

cernible downward trend has also been observed in the United States. Indeed,

as argued by Karabarbounis and Neiman (2014), this decline in labour share

has become a global trend.

International Labour Review

260

The literature has furthermore noted the appearance of a similar pattern

in China, where there has been a signicant decline in labour share from the

middle of the 1990s (Bai and Qian, 2 010; Luo and Zhang, 2010). Against the

background of “New Normal Economics”, China is encouraged to boost domes-

tic savings and consumption, rebalance the structure of the external and in-

ternal economy, and change the growth model from the previous factor-driven

model to a new model driven by human capital and innovation. All these tar-

gets depend on the improvement of the population’s income and can hardly be

achieved if the labour share keeps declining.

Important as the study of labour share change is, no agreement has been

reached on the mechanism behind it. The main explanations given for China’s

declining labour share can be roughly divided into four categories: (1) factor-

biased technological progress; (2) structural change; (3) reforms including pri-

vatization, restructuring of state-owned enterprises and ownership reforms; and

(4) globalization, especially in the form of foreign direct investment (FDI).

First, from the perspective of technology, if technological change is labour-

augmenting, then a balanced growth model would imply a constant labour share

(Acemoglu, 2003; Bentolila and Saint-Paul, 20 03). However, this contrasts with

the pattern observed in China. Huang and Xu (2009) argue that it was labour-

saving technological progress that led to the decline of labour share in China

after 1995. Second, in terms of structural change, the relative growth of the ser-

vice sector compared with the agricultural sector is perceived as one of the two

driving forces behind the decline in the labour share over the last three decades

(Bai and Qian, 2010). Third, a series of economic reforms have had an inu-

ential impact on the labour market. Yang (1999) lists various institutional fac-

tors in China that resulted in labour income change, including labour mobility

restrictions, welfare systems and nancial policies oering ination subsidies

and investment credits to the urban sector. Moreover, by investigating the dis-

tortions in product and factor markets, Bai and Qian (2010) ascribe the labour

share decline within the manufacturing sector after 1998 to the restructuring of

state-owned enterprises and to expanded monopoly power. Lastly, international

studies on labour share also identify globalization as having a major role. For

instance, Harrison (20 05) provides evidence of its eect on labour share in de-

veloped countries. In the case of China, Luo and Zhang (2010) point to the nega-

tive inuence of FDI on labour share due to regional competition, which weak-

ens the labour force’s bargaining power and thus reduces labour income share.

In general, the abovementioned explanations for change in labour share

take a fundamentally macroeconomic approach. In contrast, in this article we

focus on microeconomic dimensions, restricting our attention to change at rm

level within the industrial sector.

We restrict our focus to the secondary sector on account of its importance

and representativeness. This sector has long been, and will long continue to be

important for the overall economy. During our sample period (1998–2007 ), it

was consistently the largest sector in terms of value added, accounting for more

than 45 per cent of the total most years (20 02 being the only exception). More re-

cently, although the industrial sector was overtaken in size by the service sector

in 2013, its share remains above 4 2 per cent and still accounts for a large portion