Unification of power and responsibilities for state‐owned enterprises: A quasi‐natural experiment

• Published date: 01 November 2023
• Author: Ning Hu,Shilei Yu,Yanan Cao,Savannah (Yuanyuan) Guo,Yu Wang
• Date: 01 November 2023
• DOI: http://doi.org/10.1111/corg.12514

ORIGINAL ARTICLE

Unification of power and responsibilities for state-owned

enterprises: A quasi-natural experiment

Ning Hu

1

|Shilei Yu

1

|Yanan Cao

2

|Savannah (Yuanyuan) Guo

3

|Yu Wang

4

1

Southwestern University of Finance and

Economics, Chengdu, China

2

Shanghai University of International Business

and Economics, Shanghai, China

3

University of Nevada, Reno, Nevada, USA

4

Xiamen University, Xiamen, China

Correspondence

Yanan Cao, Shanghai University of

International Business and Economics,

Shanghai, China.

Email: sufecyn@126.com

Funding information

National Natural Science Foundation of China,

Grant/Award Numbers: 72002173,

72010107001, 72102139; Finance

Department of Sichuan Province, Grant/Award

Number: 2021-sckjkt-026; Shanghai Municipal

Education Commission, Grant/Award Number:

21CGA65

Abstract

Research Question/Issue: Based on Property Rights Theory and Empowerment The-

ory, this paper uses the establishment of local State-Owned Assets Supervision and

Administration Commissions (SASACs) as a quasi-natural experiment to investigate

whether and how SASACs improve the efficiency of state-owned enterprises (SOEs).

Research Findings/Insights: (1) After the establishment of SASACs, and compared to

those not supervised by the SASACs (i.e., the control group), SOEs governed by local

SASACs (i.e., the treatment group) have experienced a significant increase in decen-

tralization and empowerment from the government, proxied by corporate pyramid

levels. We also find increased pay-performance sensitivity for SOE managers and

higher productivity measured by total factor productivity (TFP). (2) SASACs adopt dif-

ferent strategies to manage SOEs in monopolistic and competitive industries. (3) The

above effect of the SASACs is more pronounced in SOEs supervised by high-quality

governments that effectively protect property rights, enforce fair contracts, apply

laws and regulations to everyone, and sufficiently refrain from expropriation.

Theoretical/Academic Implications: Using a quasi-natural experiment, this paper

expands the existing literature on SOE reform from the perspective of incentive

reform at the regulatory level based on Property Rights Theory and Empowerment

Theory.

Practitioner/Policy Implications: (1) Privatization is not necessarily the only optimal

solution for SOE reform. We show that the unification of power and responsibilities

can be very effective and is perhaps less costly and more practical than privatization.

Thus, our study provides an encouraging solution for SOE reform for other countries.

(2) Countries experiencing SOE reform should also work on strengthening their gov-

ernment quality in order to fully maximize the benefit of the reform.

KEYWORDS

corporate governance, Empowerment Theory, establishment of SASACs, flexible supervising

methods, Property Rights Theory, total factor productivity

1|INTRODUCTION

State-owned enterprises (SOEs) are a common form of business enti-

ties worldwide and across industries, accounting for 25 out of the top

100 Global Fortune companies (Du et al., 2018; Musacchio

et al., 2014). Despite their undeniable importance to the world's econ-

omy, SOEs are often perceived as less efficient than other entities,

and SOE managers tend to be undermotivated to improve perfor-

mance. Research finds that problems, such as ownership vacancy

(Bradshaw et al., 2019; Gu et al., 2020; Wei, 2021), a lack of

Received: 27 February 2022Revised: 3 November 2022Accepted: 27 December 2022

DOI: 10.1111/corg.12514

Corp Govern Int Rev. 2023;31:971–993. wileyonlinelibrary.com/journal/corg © 2023 John Wiley & Sons Ltd.971

delegation of power and responsibilities (Chen et al., 2011;Du

et al., 2018; Fan et al., 2013; Xu, 2011), and “soft budget constraint”

(see Lin & Li, 2008, among others), result in various moral hazards and

agency problems in SOEs. How to improve SOE efficiency has thus

become a crucial question for researchers and practitioners in many

economies.

As the country with the largest scale of SOEs—one third in the

number of firms and two thirds in market capitalization (Jiang &

Kim, 2020)—China has introduced a series of measures since the

1970s to improve SOE productivity. Most noticeable was the estab-

lishment of the State-Owned Assets Supervision and Administration

Commission (SASAC) under the State Council (i.e., the central SASAC).

The SASAC consolidated various industry-specific ministries and

became the direct supervisor and manager of state-owned assets.

1

After the central SASAC was established in March 2003, provincial-

and municipal-level SASACs (i.e., local SASACs) started to develop as

well. The formation of the SASACs arguably alleviated the free-rider

problems that plagued SOEs when multiple administrative institutions

shared the power and responsibilities of supervising SOEs, blaming

each other for poor performance while SOEs went unregulated. We

use the creation of the SASACs as a quasi-natural experiment and

empirically investigate whether and how SASACs improve the effi-

ciency of SOEs.

We hypothesize that the establishment of the SASACs introduces

a series of positive effects on SOEs. Our theoretical development is

fostered on the foundation of Property Rights Theory and Empower-

ment Theory. The Property Rights Theory contends that clearly

defined ownership is the premise of realizing efficiency (Coase, 1960),

which suggests that the creation of the SASACs as the sole responsi-

ble governmental division for managing and supervising state-owned

assets lays the groundwork for improving the operating efficiency of

SOEs. The Empowerment Theory emphasizes the importance of dele-

gating decision-making power to lower level managers with local

information and specialized knowledge to improve firm performance

(Hayek, 1945; Jensen & Meckling, 1992). This theory suggests that

the SASACs are more likely to decentralize and delegate responsibili-

ties to SOE managers, compared to the period when multiple govern-

mental divisions jointly managed SOEs. In addition, following prior

studies that find performance-based payment is often used in combi-

nation with the delegation of decision rights (Baiman & Rajan, 1995;

Bushman et al., 2000; Jensen & Meckling, 1992; Melumad &

Reichelstein, 1987), we predict that pay-performance sensitivity will

increase following the establishment of the SASACs to prevent SOE

managers from abusing their private information. Lastly, SOE man-

agers are no longer required to fulfill their social roles to serve the

public after SASACs were created. We expect SOE productivity to sig-

nificantly improve due to a combination of enhanced delegation to

the local level, increased pay-performance sensitivity, and a more

focused role for SOE managers.

We use the establishment of the local SASACs as a quasi-natural

experiment and construct staggered difference-in-differences (DID)

models to empirically investigate the economic consequences of the

creation of the SASACs. As predicted, we find that, after the

establishment of the SASACs, and compared to those not supervised

by the SASACs (i.e., the control group), SOEs governed by local

SASACs (i.e., the treatment group) experience a significant increase in

decentralization and empowerment from the government, proxied by

corporate pyramid levels. We also find increased pay-performance

sensitivity for SOE managers and higher productivity measured by

total factor productivity (TFP).

In addition, because the optimal configuration of a business's con-

trol system requires balancing the benefits of delegating decision

rights to lower managers against control losses from increased infor-

mation asymmetries, we hypothesize that the SASACs will adopt dif-

ferent strategies in managing SOEs in different industries. Our results

show that the beneficial effects are primarily observed in SOEs oper-

ating in competitive industries, where competition can act as a strong

disciplinary governance mechanism (Shleifer & Vishny, 1997) to lower

the agency costs of delegation. For monopolistic industries, however,

we do not observe any significant increase in decentralization,

performance-based payment, or productivity, presumably because

monopolistic SOEs derive much of the profit from their monopolistic

position and are not disciplined by competition. Thus, the benefits of

delegation are likely to be much smaller than the costs, and the

SASACs maintain tight control. Collectively, these results suggest that

SASACs are able to apply flexible approaches to oversee firms in dif-

ferent industries.

In our cross-sectional analyses, we study the heterogeneous

effect based on government quality. We define a high-quality gov-

ernment as a government that effectively protects property rights,

enforces fair contracts, applies laws and regulations to everyone,

and sufficiently refrains from expropriation (Levine, 2005). Accord-

ing to the administrative hierarchy, the central SASAC manages

central SOEs and is a subordinate division of the central govern-

ment. The local SASACs, which manage local SOEs, are subordinate

divisions of the local governments and also follow the leadership

of the central SASAC. We argue that high-quality local govern-

ments can refrain from expropriating and intervening in local

SASACs' management of the SOEs. Thus, high-quality governments

provide a better institutional foundation for local SASACs to super-

vise SOEs. Applying a widely used government quality index based

on survey responses of corporate executives from 120 major Chi-

nese cities (Chen, Li, Xiao, & Zou, 2014), we find that the effect of

SASACs is more pronounced in SOEs supervised by high-quality

governments. We interpret this result as evidence that, compared

with low-quality governments, high-quality governments can more

efficiently manage SOEs, presumably because they can more suc-

cessfully enforce policies and rules and apply more appropriate

administrative approaches that respect the difference between

competitive and monopolistic industries. Finally, we confirm that

our primary results are robust to a battery of analyses, including a

parallel trend test, a falsification test, applying two alternative sam-

ples and several key variable measurements, changing the industry

classifications of competitive versus monopolistic industries, and

including additional control variables and fixed effects at the

municipal level.

972 HU ET AL.