Access to this full-text is provided by PLOS.
Content available from PLOS One
This content is subject to copyright.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 1 / 23
OPEN ACCESS
Citation: Tang Q, Sun H (2025) Centralized
allocation of decision rights and enterprise
transformation and upgrading: Based on
human capital level and capital allocation
efficiency. PLoS ONE 20(3): e0319063. https://
doi.org/10.1371/journal.pone.0319063
Editor: Ömer Tuğsal Doruk, Adana Alparslan
Turkes Science and Technology University:
Adana Alparslan Turkes Bilim ve Teknoloji
Universitesi, TÜRKIYE
Received: August 23, 2024
Accepted: January 28, 2025
Published: March 17, 2025
Copyright: © 2025 Tang, Sun. This is an open
access article distributed under the terms of
the Creative Commons Attribution License,
which permits unrestricted use, distribution,
and reproduction in any medium, provided the
original author and source are credited.
Data availability statement: All relevant
data are within the paper and its Supporting
Information files.
Funding: China National Social Science Fund
"Research on M&A Behavior of Manufacturing
Enterprises in the Perspective of High-quality
Development" (23BGL126); China Xinjiang
RESEARCH ARTICLE
Centralized allocation of decision rights and
enterprise transformation and upgrading:
Based on human capital level and capital
allocation efficiency
Qingliu Tang, Hongfeng Sun *
School of Economics and Management, Shihezi University, Shihezi, China
* sunhongfeng@shzu.edu.cn
Abstract
Under the pressure of transformation and upgrading caused by complex environments
at home and abroad, it is signicant to explore how the centralized allocation of decision
rights affects enterprise transformation and upgrading. Employing the data of Chinese
A-share listed companies from 2010 to 2021, we demonstrate that the centralized alloca-
tion of decision rights signicantly promotes enterprise transformation and upgrading. Fur-
ther analysis indicates that enhanced human capital level and capital allocation efficiency
are potential mechanisms through which centralized allocation of decision rights affects
enterprise transformation and upgrading. Moreover, we document that the above posi-
tive correlation is more evident for groups with a high matching between cash ow rights
and control rights, groups with strong supervision of major shareholders, and groups
with a strong willingness of subsidiaries to cooperate. Collectively, these ndings conrm
the governance advantages of the centralized allocation of decision rights and provide
important implications for enterprises’ transformation and upgrading in emerging market
countries.
1. Introduction
In recent years, global economic development has faced severe challenges such as slowing
growth, overcapacity, environmental pollution, and the COVID-19 shock, which makes
enterprises face increasing pressure for transformation and upgrading [1,2]. For example,
polluting enterprises need to achieve green transformation through environmental invest-
ment and green innovation to cope with ecological pollution pressure [3,4], and traditional
manufacturing enterprises are in urgent need of intelligent and digital transformation to cope
with the impact of the COVID-19 epidemic [5]. Notably, micro-level studies on the factors
affecting enterprise transformation and upgrading primarily focus on government subsidies
and credit intervention [6, 7], technological innovation [8, 9], environmental regulation [10],
digital innovation [11], and entrepreneurial characteristics [12,13]. However, there is a lack
of research to explore the role of enterprise groups’ centralized allocation of decision rights in
enterprise transformation and upgrading.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 2 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Centralized allocation of decision rights of enterprise group refers to the organizational
management mode of concentrating the decision rights of personnel, finance, R&D and other
aspects on the parent company. There is mixed empirical evidence concerning the effect of
the centralized allocation of decision rights. Some research demonstrates that centralized
allocation of decision rights may cause serious agency problems [14], increase the cost of
knowledge transfer [15], and trigger inefficient investment [16], thus causing stock price crash
risk and damaging company value [17,18]. Other research identifies that centralized alloca-
tion of decision rights would alleviate agency problems [19], improve enterprise efficiency
[20], reduce enterprise decision-making costs [21], restrain resource misallocation [22], and
promote enterprise innovation [23]. However, the above literature primarily focuses on the
institutional background of developed countries, and scarce research thus far has focused on
the effect of the centralized allocation of decision rights on enterprise transformation and
upgrading. In this context, we mainly investigate the influence of centralized allocation of
decision rights on enterprise transformation and upgrading in emerging economies.
As the largest developing country, China provides an ideal scenario for examining the above
issue. First and foremost, enterprise groups composed of several legally independent enter-
prises have become a typical pattern of enterprise organization in emerging market countries
[23], especially in China, where more than 67.1% of public firms control at least one subsidiary.
This means that a form of “one control many” group system is widespread in China, and most
Chinese enterprise groups adopt the management mode of centralized allocation of decision
rights [24]. That is, enterprise groups concentrate decision rights on the parent company [25].
In China’s economic system, authority and hierarchy are often obvious. Managers of enter-
prises in China usually have strong control rights and mainly implement a centralized manage-
ment mode, while Western enterprises advocate a flat management mode more. In particular,
the state-owned enterprises in which the Chinese government holds the controlling stake are
the key components of the enterprise structure in China, accounting for a relatively high pro-
portion. The commercial and public welfare characteristics of such enterprises are important
attributes that distinguish Chinese enterprises from Western enterprises. The decision-making
power of state-owned enterprise groups is centralized in the parent company, and unified man-
agement is implemented, which makes it easier to achieve the national economic and social
goals. For example, state-owned enterprises need to promote the transformation and upgrading
of enterprises and then drive the upgrading of the entire industrial structure to ensure that their
development direction is consistent with national policies and plans. Second, China’s economic
development begins to transition to a green, sustainable, high-quality model [3,4], which has
made it difficult for the traditional crude development mode to fulfil the demands of economic
development. Consequently, enterprises’ production and operation situation has become
extremely complicated and rigorous, and enterprises urgently need to deal with the increasingly
serious challenges through transformation and upgrading.
Adopting a sample of China’s A-share public companies from 2010 to 2021, we demon-
strate that the centralized allocation of decision rights significantly promotes enterprise trans-
formation and upgrading. Our results remain unchanged after a battery of robustness checks.
Potential mechanism analysis shows that enhanced human capital level and capital allocation
efficiency are two mechanisms through which centralized allocation of decision rights affects
enterprise transformation and upgrading. In addition, we find that the above promotion
relationship is more evident for groups with a high matching between cash flow rights and
control rights, groups with strong supervision of major shareholders, and groups with a strong
willingness of subsidiaries to cooperate.
The contributions of this study are mainly as follows: First, we contribute to the studies
demonstrating the economic significance of the centralized allocation of decision rights.
Uygur Autonomous Region "Tianchi Talents"
introduction plan "Overseas investment strategy
and economic effect of environmental protec-
tion enterprises from the perspective of green
and low-carbon development" (CZ003612);
China Xinjiang Uygur Autonomous Region
Shihezi University High-level Talents Scientific
Research Initiation Project "Research on
Innovation Effect and Spillover Effect of
National Technology Innovation Demonstration
Enterprise Recognition Policy" (RCSK202406).
The funders had no role in study design, data
collection and analysis, decision to publish, or
preparation of the manuscript.
Competing interests: The authors have
declared that no competing interests exist.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 3 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Extant research primarily examines the role of the allocation of decision rights in agency
problems [19], enterprise efficiency [20], enterprise decision-making costs [21], resource
misallocation [22], and enterprise innovation [23]. We supplement this literature by offer-
ing empirical evidence that the centralized allocation of decision rights promotes enterprise
transformation and upgrading, which reveals the governance advantages of the centralized
allocation of decision rights. Second, we enrich the studies on the determinants of enterprise
transformation and upgrading from the perspective of decision rights allocation. While prior
micro-level studies on the determinants of enterprise transformation and upgrading primarily
focus on government subsidies and credit intervention [6,7], technological innovation [8,9],
environmental regulation [10], digital innovation [11] and entrepreneurial characteristics
[12,13], we document that the centralized allocation of decision rights matters in enterprise
transformation and upgrading. Third, different from previous studies, we systematically
analyze how the centralized allocation of decision rights can promote the transformation
and upgrading of enterprises by improving the level of human capital and the efficiency of
capital allocation. In particular, although this study is similar to Lou and Zhu’s research [23]
on centralized allocation of decision rights and enterprise innovation, there are also essential
differences. First of all, although there is a certain correlation between the research variable
enterprise innovation and enterprise transformation and upgrading, there are crucial differ-
ences in definition and scope. Secondly, Lou and Zhu’s research did not explore the theoretical
mechanism in depth, but this paper made a detailed theoretical analysis and tested the mediat-
ing effect. Finally, although there are similarities in the selection of control variables between
this paper and Lou and Zhu’s research, the research model of Lou and Zhu is a U-shaped
relationship model, which is different from the linear relationship model in this paper.
2. Theoretical analysis and hypothesis development
According to the resource-configuration theory, heterogeneous resource elements produce
different effects when they are configured, and enterprise transformation and upgrading
should be the result of the optimal allocation of resource elements [26]. Numerous studies
have shown that human capital level and capital allocation efficiency are the most essential
resource elements affecting enterprise transformation and upgrading [27]. Therefore, from
the perspective of element allocation, we analyze the effect of the centralized allocation of
decision rights on enterprise transformation and upgrading in terms of human capital level
and capital allocation efficiency.
2.1. Based on the perspective of human capital level
First, the centralized allocation of decision rights can enhance the human capital level by
alleviating financing constraints, thus promoting enterprise transformation and upgrading.
Broadly speaking, enterprise transformation and upgrading have a rigid demand for funds,
while in the early stage of transformation, enterprises usually face greater risks and weak
credit guarantees, resulting in the low willingness of external financing institutions to lend,
thus aggravating enterprises’ financing constraints. Scholars represented by Williamson
[28] argue that companies can alleviate financing constraints by structuring internal capi-
tal markets. Arguably, enterprise groups with a centralized allocation of decision rights can
leverage the internal capital market to achieve complementarity of resources among group
members, thus creating a resource integration effect [29]. Moreover, the centralized allocation
of decision rights allows the group to adopt a centralized debt model to reduce loan risks to
enhance external financing capabilities [30] and play a joint guarantee effect, thereby allevi-
ating financing constraints. This will be expected to guarantee a stable salary and promotion
plan for employees, allow the group to provide preferential policies or special treatment for
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 4 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
outstanding talents, and supplement working capital to increase the share of labour income,
thereby attracting talent aggregation. In addition, the alleviation of financing constraints can
also unify the group’s employment policy and establish a perfect staff training system [20],
further enhancing human capital levels to facilitate enterprise transformation and upgrading.
Second, the centralized allocation of decision rights can enhance the human capital level
by promoting knowledge and information sharing, thus promoting enterprise transformation
and upgrading. First, the centralized allocation of decision rights can use consistent strategic
management to establish mandatory norms of sharing behavior, thereby reducing the cost of
knowledge information transmission within the group and generating knowledge information
sharing effects [31]. This can be reasonably expected to enrich and expand enterprises’ knowl-
edge boundary, create opportunities and platforms for mutual learning, and promote better
exchange and learning among talents [32], thus improving enterprises’ human capital quality
and facilitating enterprise transformation and upgrading. Second, most knowledge is implicit
and difficult to transmit, which is commonly called experience or intuition. The centralized
allocation of decision rights gives the delegated personnel group authority so that they have
higher power and motivation to utilize their own rich experience, knowledge, and ability to
drive the talents’ growth within the group, thereby making implicit knowledge visible and
generating spillover effects [33], and ultimately improving the quality of human capital in the
enterprise groups, thereafter promoting enterprise transformation and upgrading.
2.2. Based on the perspective of capital allocation efficiency
First, the internal capital market constructed by centralized allocation of decision rights can
not only alleviate the financing constraints of enterprises, but also revitalize funds and improve
resource utilisation efficiency, thus promoting enterprises’ transformation and upgrading. Due
to the limited resources of the group, the parent company will choose the “winner” among the
subsidiaries to optimize the resource allocation [34]. This not only curbs the inefficient invest-
ment of subsidiaries, but also avoids the rent-seeking behavior of subsidiaries, thereby enhanc-
ing the group’s investment efficiency to promote enterprise transformation and upgrading.
Second, the centralized allocation of decision rights can establish a unified management
system, such as a fair personnel system, a standard workflow, and a normative financial
system, which will help reduce enterprises’ information transmission cost [21], improve
management efficiency [20], and to a certain extent promote the efficient coordination among
companies within the group, thus optimizing the group’s investment efficiency to promote
enterprise transformation and upgrading.
Third, the centralized allocation of decision rights can improve the group authority of the
parent company to appoint the head and optimize the allocation of resources to promote the
transformation and upgrading of enterprises. Generally, the parent company usually appoints
directors, supervisors, and senior administrators to exercise coordination, supervision, and
appraisal functions over the subsidiary. At the same time, the parent company generally
retains the right to appraise and make decisions on the salaries of delegated individuals [35],
so as to avoid collusion between the appointed personnel and the subsidiaries and harm the
interests of the company. When the decision rights of the enterprise groups are decentralized,
the head appointed by the parent company tends to lack authority and show weaker gover-
nance capacity. In contrast, when the decision rights of the enterprise groups are centralized,
the appointed person can take advantage of the group authority and give full play to the
governance role. This can not only weaken the discretion of the subsidiary’s management
and curb their opportunistic behavior but also enable them to know the subsidiary’s business
activities, reducing information asymmetry between the parent and the subsidiary to improve
investment efficiency [19], thereby promoting enterprise transformation and upgrading.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 5 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Taken together, given the foregoing arguments, we posit the following hypothesis:
H. The centralized allocation of decision rights significantly promotes enterprise transfor-
mation and upgrading.
3. Research methodology
3.1. Data
We select Chinese A-share public companies from 2010 to 2021 as the initial sample. Empiri-
cal data derives from the CSMAR database, WIND database, EPS database, and Juchao Infor-
mation Network. Specifically, R&D expense data are derived from the WIND database. GDP,
foreign direct investment, and financial development level data are downloaded from the EPS
database. The data on the geographical distance between parent and subsidiary companies are
hand-collected from the Juchao Information Network. Other financial and corporate gover-
nance data are retrieved from the CSMAR database. Referring to previous studies, we use the
following criteria to clean the sample. First, we remove the samples marked as abnormal by
the stock exchange. Second, we drop the samples of financial insurance, negative net assets,
and no subsidiaries. Third, missing data samples are excluded. To attenuate the influence of
outliers, we winsorize all continuous variables at the 1% level. After these steps, we obtained
19141 firm-year sample observations.
3.2. Models and variables
We use the following equation to examine the hypothesis:
TFPCen Controls Year Ind
it it it
it
,,
,,
=+ ++∑+
∑+
αα
αε
01 2 (1)
where the dependent variable TFPi,t represents the enterprise transformation and upgrading.
TFPi,t is a very broad and vague concept. There are different methods of measuring TFPi,t, but
a unified standard has not yet been established. Wen et al. [6] emphasize that total factor pro-
ductivity combines technical level, management ability, and other factors, which can compre-
hensively measure the effect of enterprise transformation and upgrading. Therefore, following
Wen et al. [6], we employ total factor productivity to measure enterprise transformation and
upgrading. In particular, we adopt the OP method to calculate TFPi,t [36].
The independent variable Ceni,t denotes the centralized allocation of decision rights.
According to Lou and Zhu [23], the employee remuneration payments by the enterprise
group capture the parent company’s control over the group’s personnel rights; it further cap-
tures the parent company’s control over the group’s various decision rights. Hence, following
previous studies [23,25], we construct the following equation (2) and regress it by year and
industry, with the estimated residual as a measure of Ceni,t.
PSalaryPAsset
it it it
Year Ind
,, ,
=+ ++
ββ ε
01 ∑+∑ (2)
where Psalaryi,t is the proportion of employee remuneration payments by the parent company.
It is measured as the “Cash paid to and for employees” item in the parent company’s cash
flow statement, scaled by the corresponding item in the consolidated statement. Passeti,t is the
percentage of the parent company’s assets, measured by dividing total parent company assets
by total consolidated statement assets. Before the regression, PSalaryi,t and PAsseti,t are tailed
by the interval [0,1].
The control variables are taken from prior studies [6,23] involving company size (Size),
asset-liability ratio (Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D
intensity (Rd), independent director ratio (Indep), institutional shareholding ratio (Organ),
regional economic level (GDP), foreign direct investment (FDI), financial development level
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 6 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
(FIN). Simultaneously, industry and year-fixed effects are controlled, which is the same as the
model (1). Detailed variable definitions are presented in Table 1.
4. Empirical results and discussions
4.1. Descriptive statistics
Table 2 gives the descriptive statistical results of key variables. The mean, standard deviation
and maximum and minimum values of TFP indicate great differences in the transformation
and upgrading levels of sample enterprises. The mean and standard deviation of Cen indicate
Table 1. Variable definitions.
Variable Definition Data Source
TFP Total factor productivity calculated by OP method CSMAR Database
Cen See model 2 CSMAR Database
Size The natural logarithm of total assets CSMAR Database
Lev Total debt divided by total assets CSMAR Database
Age The natural logarithm of listed years of enterprises CSMAR Database
Roa Net profit divided by total assets CSMAR Database
Cash Net cash flow from operating activities divided by total assets CSMAR Database
Rd R&D expenditure scaled by assets WIND Database
Indep The ratio of the number of independent directors to board directors CSMAR Database
Organ The proportion of shares held by external institutional investors CSMAR Database
GDP The natural logarithm of the Per capita GDP of each province EPS Database
FDI The ratio of foreign direct investment to local GDP EPS Database
FIN The ratio of deposits and loans of financial institutions to local GDP EPS Database
Note: These variables are total factor productivity (TFP), centralized allocation of decision rights (Cen), company size (Size), asset-liability ratio (Lev), company age
(Age), return of assets (Roa), cash flow (Cash), R&D intensity (Rd), independent director ratio (Indep), institutional shareholding ratio (Organ), regional economic level
(GDP), foreign direct investment (FDI), and financial development level (FIN).
https://doi.org/10.1371/journal.pone.0319063.t001
Table 2. Descriptive statistics for key variables.
Variables Obs Mean S.D. Minimum Median Maximum
TFP 19141 7.360 0.885 5.503 7.250 9.874
Cen 19141 -0.005 0.224 -0.629 0.005 0.463
Size 19141 22.139 1.258 19.155 21.948 26.150
Lev 19141 0.411 0.201 0.049 0.402 0.895
Age 19141 2.145 0.719 0.475 2.210 3.341
Roa 19141 0.040 0.059 -0.251 0.038 0.213
Cash 19141 0.005 0.088 -0.249 0.003 0.488
Rd 19141 0.025 0.024 0.000 0.021 0.145
Indep 19141 0.375 0.053 0.333 0.333 0.571
Organ 19141 0.428 0.248 0.003 0.448 0.918
GDP 19141 11.103 0.465 9.995 11.102 12.009
FDI 19141 0.024 0.013 0.001 0.022 0.096
FIN 19141 0.009 0.007 0.001 0.006 0.033
Note: This table presents summary statistics. The sample includes 19141 firm-year observations from 2010 to 2021. These variables are total factor productivity (TFP),
centralized allocation of decision rights (Cen), company size (Size), asset-liability ratio (Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D intensity
(Rd), independent director ratio (Indep), institutional shareholding ratio (Organ), regional economic level (GDP), foreign direct investment (FDI), and financial devel-
opment level (FIN). Detailed definitions of these variables are given in Table 1.
https://doi.org/10.1371/journal.pone.0319063.t002
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 7 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
great differences in the organizational management modes of enterprise groups. Because the
median of Cen is greater than 0, more than half of the enterprise groups choose the manage-
ment mode of centralized allocation of decision rights. Meanwhile, the descriptive statistical
results of control variables are consistent with previous literature [23,25].
4.2. Baseline regression results
Table 3 presents the estimation results of the baseline model. The main variable of focus in
the model (1) is the centralized allocation of decision rights (Cen). Column (1) presents the
Table 3. The centralized allocation of decision rights affects enterprise transformation and upgrading.
Variables (1) (2) (3)
TFP TFP TFP
Cen 0.112*** 0.122*** 0.119***
(0.00) (0.00) (0.00)
Size 0.476*** 0.461*** 0.466***
(0.00) (0.00) (0.00)
Lev 0.775*** 0.833*** 0.739***
(0.00) (0.00) (0.00)
Age 0.043*** 0.051*** 0.046***
(0.00) (0.00) (0.00)
Roa 2.186*** 2.268*** 2.045***
(0.00) (0.00) (0.00)
Cash -0.049 -0.077** -0.078**
(0.20) (0.04) (0.03)
Rd 0.197 -0.402 0.591
(0.58) (0.28) (0.13)
Indep -0.092 -0.152 -0.097
(0.54) (0.30) (0.47)
Organ 0.013 0.048 0.037
(0.72) (0.19) (0.29)
GDP 0.169*** 0.148***
(0.00) (0.00)
FDI 1.100*1.501**
(0.08) (0.02)
FIN 1.129 0.767
(0.38) (0.54)
cons -3.654*** -5.263*** -5.130***
(0.00) (0.00) (0.00)
N 19141 19141 19141
Ind/Year NO NO YES
AR2 0.632 0.641 0.674
Note: The table shows that the explained variable is total factor productivity (TFP), the explanatory variable is cen-
tralized allocation of decision rights (Cen), and the other control variables are company size (Size), asset-liability ratio
(Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D intensity (Rd), independent director ratio
(Indep), institutional shareholding ratio (Organ), regional economic level (GDP), foreign direct investment (FDI),
and financial development level (FIN). The control variables in the regression of column (1) only contain the control
variables at the company level. Then, on this basis, column (2) continues to add control variables at the regional level.
Finally, the year and industry fixed effects are further added in column (3). Robust standard errors are presented in
parentheses. ***, **, and * denote statistical significance at the 1%, 5%, and 10% levels, respectively.
https://doi.org/10.1371/journal.pone.0319063.t003
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 8 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
basic effects of the centralized allocation of decision rights on enterprise transformation and
upgrading while only controlling for company-level control variables. The coefficient on Cen
is 0.112 and significantly positive, suggesting that the centralized allocation of decision rights
promotes enterprise transformation and upgrading. Based on column (1), the control vari-
ables at the regional level are added, and the results are shown in column (2). The results show
that the coefficient on Cen is 0.122 and significantly positive. Based on column (2), column
(3) further adds the year-fixed effects and industry-fixed effects. The coefficient on Cen is still
significantly positive at the 1% level. Collectively, these findings evidence that the centralized
allocation of decision rights can facilitate enterprise transformation and upgrading, providing
empirical support for the hypothesis.
4.3. Robustness checks
4.3.1. Instrumental variable approach. To further mitigate the endogeneity issue, we
employ 2SLS with instrumental variables for the test. Here, we mainly choose the geographical
distance between parent-subsidiary companies. This is primarily because the farther the
geographical distance between parent-subsidiary companies, the more serious the information
asymmetry within the group [37], and the higher the cost of information communication
and transmission, which increases the cost of group centralization and weakens the group’s
willingness to centralize [15]. Therefore, we could believe that the geographical distance
between parent-subsidiary companies will affect the group’s centralized allocation of decision
rights, but will not directly affect enterprise transformation and upgrading. Further, we
perform the 2SLS regressions as follows:
Cen
it it it it
PGDControlsYearInd
,,
,,
=
ΓΓ Γ
01 2
+ ++++
∑∑ε (3)
TFPCen PGDControlsYearInd
it it it
it
,,
,,
_=
ΖΖ Ζ
01 2
+ ++++
∑∑ε (4)
where the model (3) is the first-stage estimation, and model (4) is the second-stage regression.
PGDi,t represents the instrumental variable; Cen_PGDi,t is the fitted value of the first-stage
estimation; other variables are the same as in model (1).
Columns (1) and (2) in Appendix 1 show the results of the instrumental variable
approach. In the first-stage estimations in column (1), the instrumental variable PGD is
statistically negatively correlated with Cen, suggesting that the instrumental variable meets
the relevance criterion. Moreover, the F-statistics of the first-stage regression is 362.384 and
significantly positive, thereby rejecting the weak instrumental variables hypothesis. Column
(2) shows the second-stage results of TFP as the dependent variables. The coefficient on
Cen_PGD is 0.595 and significantly positive, confirming the robustness of the aforemen-
tioned basic results.
4.3.2. Propensity score matching (PSM). Although we argue that there is a positive
association between the centralized allocation of decision rights and enterprise transformation
and upgrading, the benchmark findings may suffer from self-selection bias. That is, enterprise
groups with specific features may be more inclined to adopt centralized management and
prefer to carry out transformation and upgrading. To ease this concern, we re-regress the
equation (1) employing the propensity score matching method.
First, our indicator variable measuring the group’s decision rights, Cen_H, equals one if
the value of Cen is greater than zero, showing that the group’s decision rights are centralized;
otherwise, Cen_H equals zero, indicating that the group’s decision rights are decentralized.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 9 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Second, we regress our indicator variable Cen_H, employing a logit model to estimate the
propensity score for a group to have a centralized allocation of decision rights. Afterwards,
the treatment group sample (Cen_H = 1) is matched to the control group sample (Cen_H
= 0), which has the closest propensity score. We require the caliper to be 0.05 and perform
the matching with replacement. Appendix 2 details the results of the equilibrium hypothesis
tests. It suggests the matching is overall well balanced as the differences across covariates are
generally insignificant in all characteristics. Column (3) in Appendix 1 reports the regression
results of model (1) with the matched sample. The coefficient on Cen is 0.053 and significantly
positive, which is in line with the baseline findings in Table 3.
4.3.3. Heckman selection model. Since the propensity score matching method can only
alleviate the self-selection bias problem of observable variables, we further use the Heckman
selection model to solve the sample selection problem of unmeasurable variables [38].
In the first stage, we employ the probit model to estimate the probability that a group
adopts centralized management, measured as the indicator variable, as noted in section
4.3.2. Referring to previous literature [25], we control the following variables: company size
(Size), asset-liability ratio (Lev), firm age (Age), return of assets (Roa), cash flow (Cash),
R&D intensity (Rd), independent director ratio (Indep), institutional shareholding ratio
(Organ), The ratio of net fixed assets to total operating income (Zbcc), The ratio of the total
assets of the parent company’s statements to the total assets of the consolidated statements
(Asset_ratio), The ratio of the shareholding ratio of the 2nd to 5th largest shareholder to
that of the largest shareholder (Balance), and managerial ownership (Mshare). Following
Quan et al. [2], we adopt the industry annual average with the centralized allocation of
decision rights (MCen) as the instrumental variable. We argue that enterprises in the same
industry are likely to have incentives to adopt a similar allocation of decision rights, such an
industry-level mean variable is likely to be positively correlated with Cen, but less likely to
affect enterprise transformation and upgrading.
The first-stage regression results (Appendix 3) indicate that the coefficient on MCen is sig-
nificantly positive. The Inverse Mills Ratio (IMR) generated in the first stage was subsequently
incorporated into the second-stage regression to control for sample selection bias. Column
(4) in Appendix 1 shows the second-stage results. The coefficient on Cen is 0.196 and remains
significantly positive. Consequently, the above results indicate that our findings are unlikely
due to sample selection bias.
In addition, we make a supplementary test of the Heckman selection model. First, we
use the industry annual average with centralized allocation of decision rights (MCen) as the
instrumental variable in the first stage of the Heckman selection model, and the control vari-
ables are consistent with the benchmark model (1). The regression results are shown in col-
umns (1) and (2) of Appendix 4. Second, following the instrumental variable in Section 4.3.1,
we use the geographical distance of the parent-subsidiary company (PGD) as an instrumental
variable in the first stage of the Heckman selection model, and the control variables are con-
sistent with the benchmark model (1). The regression results are shown in columns (3) and
(4) of Appendix 4. Third, we delete the instrumental variable in the first stage regression of the
Heckman selection model and then estimate the model. The regression results are shown in
columns (5) and (6) of Appendix 4. These results are consistent with the expectations of the
article.
4.3.4. Using Cen with a lag of one year. Considering that the impact of groups’
centralized allocation of decision rights on enterprise transformation and upgrading may
have a certain time lag, thus we re-run our model (1) using Cen with a lag of one year
(L.Cen). Column (5) in Appendix 1 shows the results. The coefficient on L.Cen is 0.151 and
significantly positive, consistent with the above conclusions.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 10 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
4.3.5. Alternative measure for explained variables. The level of innovation reflects
a firm’s core competitiveness and is of great significance in promoting enterprise
transformation and upgrading [9]. Hence, we use the level of innovation (Patent), measured
as the natural logarithm of the number of patent applications, as an alternative variable for
enterprise transformation and upgrading. Column (6) in Appendix 1 presents the results. The
coefficient on Cen is 0.645 and significantly positive. Unsurprisingly, the result is in line with
prior findings.
5. Further analysis
5.1. Mechanism test
By now, we have confirmed that the group’s centralized allocation of decision rights signifi-
cantly promotes enterprise transformation and upgrading. Further, we explore the underlying
mechanism behind the basic results in this section. According to the above analysis, human
capital level and capital allocation efficiency are two important ways to influence enterprise
transformation and upgrading [27], and the centralized allocation of decision rights can
improve the corporate human capital level and capital allocation efficiency. To verify these
mechanisms, we empirically test the above two channels.
First, referring to the research of Chemmanur et al. [39], we use the proportion of employ-
ees with bachelor’s degrees and above in the total number of employees to measure the
human capital level (HCL). Second, referring to Richardson’s research [40]. We construct
the following model (5) and estimate its residual, using the absolute value of the residual to
measure enterprise inefficient investment (Absinv). Among them, the explained variable
enterprise investment (Investi,t) is the difference between the cash paid for the construction
of fixed assets, intangible assets and other long-term assets and the net cash recovered from
the disposal of fixed assets, intangible assets and other long-term assets, and divided by the
total assets at the beginning of the period; the explanatory variables include company growth
(Growthi,t-1), cash flow (Cashi,t-1), asset-liability ratio (Levi,t-1), company age (Agei,t-1), annual
stock return (Reti,t-1), company size (Sizei,t-1) and enterprise investment (Investi,t-1), while con-
trolling the year and industry fixed effects.
Invest Growth Cash LevAge
it it it it it,,,, ,
=+ +++
µµ µµµ
01 12 13 14−−
−−
−1
++ ++++
µµ
µε
516171
RetSizeInvestYearInd
it it it it,,
,,
−− −∑∑ (5)
Table 4 presents the results. The explained variable is HCL in Panel A, and the control
variables are taken from previous studies [41]. Specifically, we include company size (Size),
asset-liability ratio (Lev), firm age (Age), cash flow (Cash), return of assets (Roa), independent
director ratio (Indep), institutional shareholding ratio (Organ), and cash paid for long-term
assets scaled by total assets (Capex). The coefficient on Cen is 2.357 and significantly positive,
thereby providing evidence to support that the centralized allocation of decision rights facili-
tates enterprise transformation and upgrading by improving the human capital level.
The explained variable is Absinv in Panel B, and the control variables are drawn from pre-
vious studies [42]. Specifically, we include company size (Size), asset-liability ratio (Lev), firm
age (Age), cash flow (Cash), return of assets (Roa), independent director ratio (Indep), board
size (Board), and the ratio of total assets to total income (Ci). The coefficient on Cen is -0.006
and significantly negative, thus providing empirical evidence that the centralized allocation
of decision rights facilitates enterprise transformation and upgrading by improving capital
allocation efficiency.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 11 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
5.2. Cross-sectional analysis
5.2.1. The matching degree of cash flow rights and control rights. In emerging
economies where investor protection is insufficient, the intra-conglomerate capital
market, which was originally designed to improve the efficiency of capital allocation, has
been partially alienated as a channel for transferring benefits to controlling shareholders.
Broadly speaking, only when the matching degree between cash flow rights and control
rights is high can the group restrain the parent company’s tunnelling motivation and
minimize the resource dissipation caused by the centralized allocation of decision rights
[43], thus giving full play to the governance advantages of the centralized allocation
of decision rights. If so, it could be expected that the positive impact of the centralized
allocation of decision rights on enterprise transformation and upgrading is more
pronounced for groups with a high matching between cash flow rights and control rights.
Following Chandler [44], we first standardize the group’s control rights (Cen) and cash
flow rights (CashPower), and then use the matching formula PPD = 1- | Cen – CashPower |
Table 4. Potential mechanisms.
Panel A human capital level Panel B capital allocation efficiency
Variables HCL Variables Absinv
Cen 2.357*** Cen -0.006***
(0.00) (0.00)
Size 0.321** Size 0.002***
(0.04) (0.00)
Lev -6.953*** Lev 0.016***
(0.00) (0.00)
Age 0.590** Age 0.001***
(0.01) (0.01)
Cash -0.857 Cash 0.016***
(0.60) (0.00)
Roa 11.142*** Roa 0.029***
(0.00) (0.00)
Indep 13.583*** Indep -0.002
(0.00) (0.80)
Organ 2.235*** Board -0.008***
(0.00) (0.00)
Capex -30.470*** Ci 0.002***
(0.00) (0.00)
cons 21.518*** cons -0.009
(0.00) (0.19)
N 19141 N 19141
Ind/Year YES Ind/Year YES
AR2 0.216 AR2 0.076
Note: Panel A and panel B of this table test the two mechanisms of human capital level and capital allocation efficien-
cy, respectively. The explained variables are human capital level (HCL) and inefficient investment (Absinv), the ex-
planatory variable is centralized allocation of decision rights (Cen), and the control variables are company size (Size),
asset-liability ratio (Lev), firm age (Age), cash flow (Cash), return of assets (Roa), independent director ratio (Indep),
board size (Board), institutional shareholding ratio (Organ), the ratio of total assets to total income (Ci) and cash paid
for long-term assets scaled by total assets (Capex). Robust standard errors are presented in parentheses. ***, **, and *
denote statistical significance at the 1%, 5%, and 10% levels, respectively.
https://doi.org/10.1371/journal.pone.0319063.t004
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 12 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
to obtain the matching degree of the group’s cash flow rights and control rights. Based
on the model (1), the interaction term of Cen and PPD and the PPD variable are added,
and the baseline model is re-estimated. Panel A in Table 5 presents the estimation results.
The coefficient on the interaction term is 0.180 and significantly positive, confirming
Table 5. Cross-sectional tests.
Panel A the matching degree of cash flow rights and con-
trol rights
Panel B the supervision of major shareholders Panel C the willingness of subsidiaries
to cooperate
Variables TFP Variables TFP Variables TFP
Cen 0.134*** Cen 0.113*** Cen 0.230***
(0.00) (0.00) (0.00)
Cen*PPD 0.180*** Cen*Top1 0.003** Cen*SOP 0.253***
(0.00) (0.02) (0.00)
PPD 0.043*** Top1 0.002*** SOP 0.167***
(0.00) (0.00) (0.00)
Size 0.465*** Size 0.465*** Size 0.459***
(0.00) (0.00) (0.00)
Lev 0.737*** Lev 0.741*** Lev 0.703***
(0.00) (0.00) (0.00)
Age 0.044*** Age 0.053*** Age 0.034***
(0.00) (0.00) (0.00)
Roa 2.054*** Roa 2.037*** Roa 2.008***
(0.00) (0.00) (0.00)
Cash -0.074*Cash -0.071*Cash -0.075*
(0.08) (0.10) (0.08)
Rd 0.594*** Rd 0.676*** Rd 0.638***
(0.00) (0.00) (0.00)
Indep -0.101 Indep -0.135** Indep -0.085
(0.14) (0.05) (0.21)
Organ 0.027 Organ -0.017 Organ 0.034**
(0.10) (0.35) (0.04)
GDP 0.150*** GDP 0.146*** GDP 0.139***
(0.00) (0.00) (0.00)
FDI 1.446*** FDI 1.450*** FDI 1.569***
(0.00) (0.00) (0.00)
FIN 0.859 FIN 0.803 FIN 1.132*
(0.14) (0.17) (0.05)
cons -5.111*** cons -5.119*** cons -4.887***
(0.00) (0.00) (0.00)
N 19141 N 19141 N 19141
Ind/Year YES Ind/Year YES Ind/Year YES
AR2 0.674 AR2 0.675 AR2 0.677
Note: The table panel A, panel B and panel C respectively test the three moderating effects of matching degree of cash flow rights and control rights, supervision of ma-
jor shareholders and willingness of subsidiaries to cooperate. The explained variable is total factor productivity (TFP), the explanatory variable is centralized allocation
of decision rights (Cen), the moderating variables are matching degree of cash flow rights and control rights (PPD), supervision of major shareholders (Top1), willing-
ness of subsidiaries to cooperate (SOP), and the control variables are company size (Size), asset-liability ratio (Lev), company age (Age), return of assets (Roa), cash flow
(Cash), R&D intensity (Rd), independent director ratio (Indep), institutional shareholding ratio (Organ), regional economic level (GDP), foreign direct investment
(FDI), and financial development level (FIN). Robust standard errors are presented in parentheses. ***, **, and * denote statistical significance at the 1%, 5%, and 10%
levels, respectively.
https://doi.org/10.1371/journal.pone.0319063.t005
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 13 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
our expectation that the positive effect of the centralized allocation of decision rights on
enterprise transformation and upgrading is more pronounced for groups with a high
matching between cash flow rights and control rights.
5.2.2. The supervision of major shareholders. In China, large shareholders often have
actual control of the group, and their interests are bound to the interests of the group. They
not only have a strong incentive to promote enterprise transformation and upgrading to
achieve steady growth in shareholder wealth, but also have the ability to promote enterprise
transformation and upgrading by monitoring the self-interested behavior of subsidiaries. If so,
it could be conjectured that the positive effect of the centralized allocation of decision rights
on enterprise transformation and upgrading is more pronounced for groups with strong
supervision of major shareholders. Following Wang et al. [45], we employ the shareholding
ratio of the largest shareholder (Top1) to measure the supervision of major shareholders,
and the larger the value, the stronger the supervision of major shareholders. Based on model
(1), the interaction term of Cen and Top1 and the Top1 variable are added, and the baseline
model is re-estimated. Panel B in Table 5 presents the estimation results. The coefficient on
the interaction term is 0.003 and significantly positive, showing that the positive effect of the
centralized allocation of decision rights on enterprise transformation and upgrading is more
pronounced for groups with strong supervision of major shareholders, which is consistent
with the above conjecture.
5.2.3. The willingness of subsidiaries to cooperate. Whether the governance
advantages of the centralized allocation of decision rights can be effectively exerted
depends to a certain extent on the willingness of subsidiaries to cooperate. Generally
speaking, the stronger the willingness of subsidiaries to cooperate, the lower the inefficient
investment level at the subsidiary level can be, and the more conducive it is to establishing
and maintaining an internal capital market, thus enhancing the efficiency of resource
allocation and knowledge transfer within the group, and ultimately facilitating enterprise
transformation and upgrading. Following Zhu and Kong [25], we use subsidiary business
scale (SOP), measured as the difference between consolidated statement operating income
and parent company operating income divided by consolidated statement operating income,
as a proxy variable for the willingness of subsidiaries to cooperate. Based on the model (1),
the interaction term of Cen and SOP and the SOP variable are added, and the baseline model
is re-estimated. Panel C in Table 5 presents the estimation results. The coefficient on the
interaction term is 0.230 and significantly positive, suggesting that the positive effect of the
centralized allocation of decision rights on enterprise transformation and upgrading is more
pronounced for groups with a strong willingness of subsidiaries to cooperate, which is in line
with the above analysis.
5.2.4. Ownership type of the enterprise. The difference of enterprise ownership type
will directly affect the financing environment and governance mode of enterprises. The
centralized allocation of decision rights helps enterprises to better respond to national
policies and integrate various resources, form a unified strategic plan and action plan, and
promote the transformation and upgrading of enterprises. Compared with state-owned
enterprises, private enterprises are relatively limited in resource acquisition, and the
promotion effect of centralized allocation of decision rights on enterprise transformation
and upgrading will be reduced [46]. Foreign enterprises usually need to take into account
the global strategy and the needs of the local market, with a more standardized and
international management system, the decision-making process is often more complex.
This may weaken the promotion effect of centralized allocation of decision rights of foreign
enterprises on enterprise transformation and upgrading. In addition, there are differences
in the governance concepts and experiences of enterprises with different ownership types.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 14 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
State-owned enterprises are usually controlled by the government, and their decision-making
power is concentrated in higher-level management institutions or government departments
to a certain extent to ensure the consistency and implementation of policies. State-owned
enterprises may accumulate more experience about centralized decision-making to promote
change in the long-term development, while private enterprises and foreign enterprises
may need to pay more to adapt and adjust. This will lead to different effects of centralized
allocation of decision rights on enterprise transformation and upgrading under different
ownership types.
Therefore, this paper divides the samples into state-owned enterprise samples, private
enterprise samples and foreign enterprise samples, and tests the impact of centralized alloca-
tion of decision rights on enterprise transformation and upgrading. Table 6 panel A, panel B
and panel C are the test results of state-owned enterprise samples, private enterprise samples
and foreign enterprise samples respectively. The results show that the Cen coefficients are all
significantly positive. The regression coefficient of Cen in the state-owned enterprise sample
and private enterprise sample are significantly positive at the level of 1%, while the regression
coefficient of Cen in the foreign enterprise sample is only significantly positive at the level of
10%. Therefore, compared with foreign enterprise sample, the promotion effect of centralized
allocation of decision rights on enterprise transformation and upgrading is more significant
in the samples of state-owned enterprises and private enterprises. Furthermore, the regression
coefficient of Cen in state-owned enterprise sample is greater than that in private enterprise
sample. This shows that when other conditions remain unchanged, the conclusion of this
paper are applicable to enterprises with different ownership types, but the influence degree is
different.
6. Discussion
Different from previous studies, this paper systematically analyzes the promotion effect of
centralized allocation of decision rights on enterprise transformation and upgrading from
the perspective of human capital level and capital allocation efficiency. The existing literature
on the economic consequences of centralized allocation of decision rights mainly focuses
on agency problem [19], enterprise efficiency [20], enterprise decision-making cost [21],
resource mismatch [22] and enterprise innovation [23]. In particular, although this study is
similar to Lou and Zhu’s research [23] on centralized allocation of decision rights and enter-
prise innovation, there are also essential differences. Specifically, first, although there is a
correlation between the explained variable enterprise innovation studied by Lou and Zhu and
the explained variable enterprise transformation and upgrading in this study, there are also
differences. We usually think that enterprise innovation and enterprise transformation and
upgrading are important events for enterprises. Although the goals of enterprise innovation
and enterprise transformation and upgrading sometimes overlap, their core concerns and
implementation paths are different. Enterprise innovation focuses more on the development
of new products, new technologies and new processes, while enterprise transformation and
upgrading focuses more on the adjustment of the overall structure and strategic re-planning.
Enterprise innovation can be a means of enterprise transformation and upgrading, but the
means of enterprise transformation and upgrading are not limited to enterprise innova-
tion. On the contrary, there are many purposes of enterprise innovation, not necessarily for
enterprise transformation and upgrading. Second, Lou and Zhu’s research mentioned the
potential impact of centralized allocation of decision rights on enterprise innovation, but did
not discuss its specific mechanism in depth. We analyze in detail how the centralized allo-
cation of decision rights can promote the transformation and upgrading of enterprises by
improving the level of human capital and the efficiency of capital allocation. Furthermore, in
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 15 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
order to verify the validity of the theoretical analysis in this paper, we test the mediating effect
of human capital level and capital allocation efficiency. The results show that the regression
coefficient of Cen is significantly positive in the regression model of centralized allocation of
decision rights (Cen) to human capital level (HCL). In the regression model of centralized
allocation of decision rights (Cen) to enterprise inefficient investment (Absinv), the regres-
sion coefficient of Cen is significantly negative. This shows that the centralized allocation of
decision rights does improve the human capital level and the capital allocation efficiency, thus
promoting the enterprise transformation and upgrading.
We further enrich the research on the logical framework of the influence of centralized allo-
cation of decision rights on enterprise transformation and upgrading under different special
Table 6. Grouping test of enterprise ownership types.
Variables Panel A state-owned enterprises Panel B private enterprises Panel C foreign enterprises
TFP TFP TFP
Cen 0.152*** 0.094*** 0.118*
(0.00) (0.00) (0.07)
Size 0.481*** 0.465*** 0.441***
(0.00) (0.00) (0.00)
Lev 0.631*** 0.784*** 0.760***
(0.00) (0.00) (0.00)
Age 0.114*** -0.001 0.021
(0.00) (0.88) (0.36)
Roa 2.318*** 1.886*** 1.831***
(0.00) (0.00) (0.00)
Cash -0.084 -0.074 -0.013
(0.41) (0.13) (0.92)
Rd 1.943*** -0.270 1.705***
(0.00) (0.29) (0.00)
Indep -0.060 -0.099 -0.165
(0.61) (0.28) (0.43)
Organ 0.053 0.041* 0.096*
(0.26) (0.05) (0.10)
GDP 0.124*** 0.167*** 0.040
(0.00) (0.00) (0.37)
FDI 1.822*** 0.948** 3.794***
(0.00) (0.03) (0.00)
Fin 0.189 1.460*-2.248
(0.86) (0.05) (0.23)
cons -5.356*** -5.222*** -3.406***
(0.00) (0.00) (0.00)
N 6827 10906 1408
Ind/Year YES YES YES
AR2 0.685 0.606 0.655
Note: The table panel A, panel B and panel C are the test results of three groups of samples: state-owned enterprise, private enterprise and foreign enterprise. The ex-
plained variable is total factor productivity (TFP), the explanatory variable is centralized allocation of decision rights (Cen), and the other control variables are company
size (Size), asset-liability ratio (Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D intensity (Rd), independent director ratio (Indep), institutional
shareholding ratio (Organ), regional economic level (GDP), foreign direct investment (FDI), and financial development level (FIN). Robust standard errors are present-
ed in parentheses. ***, **, and * denote statistical significance at the 1%, 5%, and 10% levels, respectively.
https://doi.org/10.1371/journal.pone.0319063.t006
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 16 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
scenarios, such as the matching degree of cash flow rights and control rights, the supervision
power of major shareholders and the willingness of subsidiaries to cooperate. From the parent-
subsidiary level, we consider the preconditions for centralized allocation of decision rights
of enterprise groups to give full play to their governance advantages, mainly focusing on the
parent company’s tunneling motivation, the supervisory role of major shareholders, and the
willingness of subsidiaries to cooperate with parent companies to complete major strategic
tasks. The low hollowing-out motivation of the parent company means that the parent com-
pany is unlikely to seek personal gain by transferring resources or encroaching on the interests
of subsidiaries, which is very important for the governance of group enterprises. The strong
supervision function of major shareholders means that the internal governance structure of
the enterprise is relatively sound, and the major shareholders can play an effective supervi-
sion function to ensure that the management follows the principles of transparency, legality
and benefit to the overall interests of the group in the decision-making process. The strong
willingness of subsidiaries to cooperate with the parent company means that there is a good
communication and cooperation relationship between the parent company and subsidiaries,
and subsidiaries are willing to support the strategic direction of the parent company, which is
conducive to enhancing the overall interests of the group enterprises. The research results show
that when the parent company’s tunneling motivation is low, the major shareholder’s supervi-
sion function is strong, and the subsidiary company’s willingness to cooperate with the parent
company is strong, the centralized allocation of decision rights of the enterprise group is more
likely to exert the governance advantage of focusing on major events, that is, the centralized
allocation of decision rights has a stronger promotion effect on enterprise transformation and
upgrading. In addition, considering that different types of enterprise ownership will have an
impact on the governance effect of centralized allocation of decision rights, we further exam-
ine the heterogeneous effects of three types of ownership of state-owned enterprises, private
enterprises and foreign enterprises on centralized allocation of decision rights and enterprise
transformation and upgrading. The research results show that when other conditions remain
unchanged, the conclusion of this paper are applicable to enterprises with different ownership
types, but the influence degree is different. This not only emphasizes the applicability of the
research conclusions, but also highlights the practical differences under different ownership
types, which is of great significance for the promotion of theory and practical application.
The main deficiency of this paper lies in the measurement method of centralized allocation
of decision rights in empirical tests. The allocation of decision rights is one of the core con-
tents of enterprise internal governance mechanism, which involves the vertical distribution of
decision rights between levels and the horizontal distribution between unit entities. The mea-
surement method in this paper may not fully and accurately reflect the allocation of enterprise
decision rights. In the future, we can continue to study the internal power allocation relation-
ship of enterprise groups, explore more rigorous measurement indicators of decision rights
allocation, and find more valuable research conclusions.
7. Conclusions
How to promote enterprise transformation and upgrading has increasingly attracted wide-
spread academic attention. This paper advances academic understanding of whether and how
the centralized allocation of decision rights within a group in emerging countries affects enter-
prise transformation and upgrading. Our findings evidence that the centralized allocation of
decision rights has a positive effect on enterprise transformation and upgrading. Further anal-
ysis shows that enhanced human capital level and capital allocation efficiency are two mecha-
nisms through which centralized allocation of decision rights affects enterprise transformation
and upgrading. In addition, we find that the positive relation is more evident for groups with a
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 17 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
high matching between cash flow rights and control rights, groups with strong supervision of
major shareholders, and groups with a strong willingness of subsidiaries to cooperate.
Overall, this paper provides timely implications for enterprise groups and governments that
intend to promote enterprise transformation and upgrading. To promote transformation and
upgrading, enterprise groups should consider centralized management, and the government
should appropriately guide enterprise groups to adopt a centralized management system and
create convenient conditions for them. These findings have implications for other emerging econ-
omies, particularly those with large conglomerates and under the pressure of transformation and
upgrading. Furthermore, when implementing centralized management, enterprise groups ought
to account for particular circumstances. For example, when the matching degree of cash flow
rights and control rights is high, the supervision of major shareholders is strong, and the willing-
ness of subsidiaries to cooperate is strong, the centralized management may bring a better effect.
Appendix 1
Robustness tests
Variables Instrumental variable PSM Heckman Cen with a lag
of one year
Alternative
measure
(1) (2) (3) (4) (5) (6)
Cen TFP TFP TFP TFP Patent
PGD -0.565***
(0.00)
Cen_PGD 0.595***
(0.00)
Cen 0.053*** 0.196*** 0.645***
(0.00) (0.00) (0.00)
IMR 7.000***
(0.00)
L.Cen 0.151***
(0.00)
Size -0.015*** 0.475*** 0.469*** 0.372*** 0.466*** 0.263***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Lev 0.174*** 0.657*** 0.747*** 1.520*** 0.740*** -0.286***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Age -0.068*** 0.078*** 0.037*** -0.230*** 0.049*** -0.079***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Roa 0.395*** 1.854*** 2.087*** 3.871*** 2.049*** -0.430***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Cash -0.022 -0.056 -0.005 -0.145*** -0.061 -0.296***
(0.23) (0.21) (0.94) (0.00) (0.18) (0.00)
Rd -0.133*0.707*** 0.189 -0.851*** 0.553*** 10.072***
(0.09) (0.00) (0.45) (0.00) (0.00) (0.00)
Indep -0.178*** -0.026 -0.175*-0.904*** -0.074 0.080
(0.00) (0.73) (0.06) (0.00) (0.29) (0.55)
Organ 0.070*** -0.002 0.036 0.380*** 0.033** -0.121***
(0.00) (0.93) (0.11) (0.00) (0.05) (0.00)
GDP -0.044*** 0.168*** 0.150*** 0.129*** 0.152*** 0.014
(0.00) (0.00) (0.00) (0.00) (0.00) (0.51)
FDI 0.439*** 1.279*** 1.168*** 1.278*** 1.469*** 0.309
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 18 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Variables Instrumental variable PSM Heckman Cen with a lag
of one year
Alternative
measure
(1) (2) (3) (4) (5) (6)
Cen TFP TFP TFP TFP Patent
(0.00) (0.00) (0.01) (0.00) (0.00) (0.59)
FIN 0.404 0.417 0.531 1.188** 0.707 -0.740
(0.10) (0.49) (0.50) (0.03) (0.23) (0.49)
cons 0.903*** -5.657*** -5.242*** -8.047*** -5.167*** -5.113***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
N 18873 18873 10120 19141 18236 19141
Ind/Year YES YES YES YES YES YES
AR2 0.084 0.660 0.684 0.696 0.674 0.205
First-stage
F-stat
/362.384*** ////
Note: Appendix 1 shows the robustness test results. The variables include total factor productivity (TFP), centralized
allocation of decision rights (Cen), geographical distance of parent-subsidiary company (PGD), explanatory variables
after instrumental variable estimation (Cen_PGD), inverse Mills Ratio (IMR), centralized allocation of decision rights
with a lag of one year (L.Cen), company size (Size), asset-liability ratio (Lev), company age (Age), return of assets
(Roa), cash flow (Cash), R&D intensity (Rd), independent director ratio (Indep), institutional shareholding ratio
(Organ), regional economic level (GDP), foreign direct investment (FDI), and financial development level (FIN). In
addition, because there are missing values when calculating the instrumental variable (PGD), the total samples in
columns (1) and (2) are reduced. Robust standard errors are presented in parentheses. ***, **, and * denote statistical
significance at the 1%, 5%, and 10% levels, respectively.
Appendix 2
Balanced hypothesis testing
Variables Sample Mean Difference T-Value(p-Value) Deviation(%)
Treatment Group Control Group %bias %reduct
Size Before matching 22.028 22.254 -12.52(0.000) -18.1
After matching 22.028 22.052 -1.49(0.162) -2 89.2
Lev Before matching 0.406 0.416 -3.22(0.000) -4.7
After matching 0.406 0.410 -1.40(0.160) -0.2 56.8
Age Before matching 2.040 2.254 -20.74(0.000) -30
After matching 2.040 2.045 -0.45(0.654) -0.7 97.8
Roa Before matching 0.044 0.036 9.60(0.000) 13.9
After matching 0.044 0.044 0.08(0.939) 0.1 99.2
Cash Before matching 0.005 0.006 -0.95(0.340) -1.4
After matching 0.005 0.004 0.75(0.451) 1.1 18.9
Rd Before matching 0.026 0.025 4.09(0.000) 5.9
After matching 0.026 0.026 1.45(0.148) 2.1 64.7
Indep Before matching 0.372 0.377 -6.09(0.000) -8.8
After matching 0.372 0.372 -0.04(0.967) -0.1 99.4
Organ Before matching 0.434 0.434 3.33(0.001) 4.8
After matching 0.434 0.440 -1.64(0.102) -2.4 50.7
GDP Before matching 11.085 11.122 -5.52(0.000) -0.8
After matching 11.085 11.084 0.08(0.936) 0.1 98.6
FDI Before matching 0.024 0.024 -2.35(0.019) -3.4
After matching 0.024 0.024 -0.86(0.391) -1.3 63
FIN Before matching 0.009 0.008 2.48(0.013) 3.6
After matching 0.009 0.009 -1.03(0.302) -1.5 57.8
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 19 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Note: Appendix 2 details the results of the equilibrium hypothesis tests. The variables include company size (Size),
asset-liability ratio (Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D intensity (Rd), indepen-
dent director ratio (Indep), institutional shareholding ratio (Organ), regional economic level (GDP), foreign direct
investment (FDI), and financial development level (FIN).
Appendix 3
Heckman selection model
Panel A First Stage Panel B Second Stage
Variables Cen Variables TFP
MCen 0.854*** Cen 0.196***
(0.00) (0.00)
Size -0.022*** IMR 7.000***
(0.00) (0.00)
Lev 0.199*** Size 0.372***
(0.00) (0.00)
Age -0.058*** Lev 1.520***
(0.00) (0.00)
Roa 0.466*** Age -0.230***
(0.00) (0.00)
Cash -0.014 Roa 3.871***
(0.49) (0.00)
Rd -0.178** Cash -0.145***
(0.03) (0.00)
Indep -0.199*** Rd -0.851***
(0.00) (0.00)
Organ 0.083*** Indep -0.904***
(0.00) (0.00)
Zbcc 0.040*** Organ 0.380***
(0.00) (0.00)
Asset_ratio 0.054*** GDP 0.129***
(0.00) (0.00)
Balance -0.024*** FDI 1.278***
(0.00) (0.00)
Mshare 0.000** FIN 1.188**
(0.01) (0.03)
cons 0.492*** cons -8.047***
(0.00) (0.00)
N 19140 N 19141
Ind/Year YES Ind/Year YES
AR2 0.077 AR2 0.696
Note: Appendix 3 Panel A and Panel B are the first and second stages of the Heckman selection model. The variables
include total factor productivity (TFP), centralized allocation of decision rights (Cen), industry annual average with
centralized allocation of decision rights (MCen), inverse Mills Ratio (IMR), company size (Size), asset-liability ratio
(Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D intensity (Rd), independent director ratio
(Indep), institutional shareholding ratio (Organ), regional economic level (GDP), foreign direct investment (FDI),
and financial development level (FIN). The ratio of net fixed assets to total operating income (Zbcc), The ratio of
the total assets of the parent company’s statements to the total assets of the consolidated statements (Asset_ratio),
The ratio of the shareholding ratio of the 2nd to 5th largest shareholder to that of the largest shareholder (Balance),
managerial ownership (Mshare). Robust standard errors are presented in parentheses. ***, **, and * denote statistical
significance at the 1%, 5%, and 10% levels, respectively.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 20 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Appendix 4
Heckman selection model supplementary results
Variables (1) (2) Variables (3) (4) Variables (5) Variables (6)
Cen TFP Cen TFP Cen TFP
MCen 0.831*** PGD -0.565*** Size -0.022*** Cen 0.196***
(0.00) (0.00) (0.00) (0.00)
Cen 0.123*** Cen 0.111*** Lev 0.200*** IMR 9.022***
(0.00) (0.00) (0.00) (0.00)
IMR 1.779*** IMR -0.730*** Age -0.058*** Size 0.345***
(0.00) (0.00) (0.00) (0.00)
Size -0.020*** 0.443*** Size -0.015*** 0.475*** Roa 0.475*** Lev 1.750***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Lev 0.170*** 0.934*** Lev 0.174*** 0.660*** Cash -0.013 Age -0.311***
(0.00) (0.00) (0.00) (0.00) (0.52) (0.00)
Age -0.065*** -0.028 Age -0.068*** 0.077*** Rd -0.179** Roa 4.395***
(0.00) (0.24) (0.00) (0.00) (0.03) (0.00)
Roa 0.401*** 2.513*** Roa 0.395*** 1.859*** Indep -0.202*** Cash -0.164***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
Cash -0.017 -0.095** Cash -0.022 -0.057 Organ 0.084*** Rd -1.273***
(0.40) (0.03) (0.26) (0.19) (0.00) (0.00)
Rd -0.270*** 0.287 Rd -0.133 0.702*** Zbcc 0.040*** Indep -1.139***
(0.00) (0.18) (0.11) (0.00) (0.00) (0.00)
Indep -0.171*** -0.295*** Indep -0.178*** -0.029 Asset_ratio 0.053*** Organ 0.482***
(0.00) (0.00) (0.00) (0.68) (0.00) (0.00)
Organ 0.077*** 0.124*** Organ 0.070*** -0.000 Balance -0.024*** GDP 0.121***
(0.00) (0.00) (0.00) (1.00) (0.00) (0.00)
GDP -0.046*** 0.097*** GDP -0.044*** 0.167*** Mshare 0.000*** FDI 1.376***
(0.00) (0.00) (0.00) (0.00) (0.01) (0.00)
FDI 0.502*** 2.036*** FDI 0.439*** 1.289*** FIN 1.278**
(0.00) (0.00) (0.00) (0.00) (0.02)
FIN 0.706*** 1.583** FIN 0.404 0.428
(0.01) (0.01) (0.11) (0.47)
cons 1.024*** -5.392*** cons 0.963*** -4.972*** cons 0.497*** cons -8.857***
(0.00) (0.00) (0.00) (0.00) (0.00) (0.00)
N 19141 19141 N 18873 18873 N 19141 N 19141
Ind/Year YES YES Ind/Year YES YES Ind/Year YES Ind/Year YES
AR2 0.069 0.674 AR2 0.084 0.674 AR2 0.074 AR2 0.698
Note: Appendix 4 is the Heckman selection model supplementary results. First, we use the industry annual average with centralized allocation of decision rights (MCen)
as the instrumental variable in the first stage of the Heckman selection model, and the control variables are consistent with the benchmark model (1). The regression
results are shown in columns (1) and (2). Second, we use the geographical distance of the parent-subsidiary company (PGD) as an instrumental variable in the first
stage of the Heckman selection model, and the control variables are consistent with the benchmark model (1). The regression results are shown in columns (3) and (4).
Third, we delete the instrumental variable in the first stage regression of the Heckman selection model, and then estimate the model. The regression results are shown
in columns (5) and (6). The variables include total factor productivity (TFP), centralized allocation of decision rights (Cen), industry annual average with centralized
allocation of decision rights (MCen), geographical distance of parent-subsidiary company (PGD), inverse Mills Ratio (IMR), company size (Size), asset-liability ratio
(Lev), company age (Age), return of assets (Roa), cash flow (Cash), R&D intensity (Rd), independent director ratio (Indep), institutional shareholding ratio (Organ),
regional economic level (GDP), foreign direct investment (FDI), and financial development level (FIN). The ratio of net fixed assets to total operating income (Zbcc),
The ratio of the total assets of the parent company’s statements to the total assets of the consolidated statements (Asset_ratio), The ratio of the shareholding ratio of the
2nd to 5th largest shareholder to that of the largest shareholder (Balance), managerial ownership (Mshare). Robust standard errors are presented in parentheses. ***, **,
and * denote statistical significance at the 1%, 5%, and 10% levels, respectively.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 21 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
Author contributions
Conceptualization: Qingliu Tang.
Data curation: Qingliu Tang.
Writing – original draft: Qingliu Tang.
Writing – review & editing: Qingliu Tang, hongfeng sun.
References
1. Xie B, Yang C, Song W, Song L, Wang H. The impact of environmental regulation on capacity utiliza-
tion of China’s manufacturing industry: An empirical research based on the sector level. Ecological
Indicators. 2023;148:110085. https://doi.org/10.1016/j.ecolind.2023.110085
2. Quan X, Ke Y, Qian Y, Zhang Y. CEO foreign experience and green innovation: Evidence from China.
Journal of Business Ethics. 2023;182(1):535–57.
3. Huang C, Chang X, Wang Y, Li N. Do major customers encourage innovative sustainable develop-
ment? Empirical evidence from corporate green innovation in China. Bus Strat Env. 2022;32(1):163–
84. https://doi.org/10.1002/bse.3123
4. Zhang Q, Yu Z, Kong D. The real effect of legal institutions: Environmental courts and rm envi-
ronmental protection expenditure. Journal of Environmental Economics and Management.
2019;98:102254. https://doi.org/10.1016/j.jeem.2019.102254
5. Mohapatra B, Tripathy S, Singhal D, Saha R. Signicance of digital technology in manufactur-
ing sectors: Examination of key factors during Covid-19. Research in Transportation Economics.
2022;93:101134. https://doi.org/10.1016/j.retrec.2021.101134
6. Wen H, Lee C-C, Zhou F. Green credit policy, credit allocation efficiency and upgrade of energy-intensive
enterprises. Energy Economics. 2021;94:105099. https://doi.org/10.1016/j.eneco.2021.105099
7. Karagianni S, Pempetzoglou M, Saraidaris A. Tax burden distribution and GDP growth: Non-linear
causality considerations in the USA. International Review of Economics & Finance. 2012;21(1):186–
94. https://doi.org/10.1016/j.iref.2011.06.002
8. Copus A, Skuras D, Tsegenidi K. Innovation and Peripherality: An Empirical Comparative Study of
SMEs in Six European Union Member Countries. Economic Geography. 2008;84(1):51–82. https://doi.
org/10.1111/j.1944-8287.2008.tb00391.x
9. Ghandour FA, Swartz P, Grenek HM, Roberts EB. E-Business Transformation via Alliance
Clusters. Technology Analysis & Strategic Management. 2004;16(4):435–55. https://doi.
org/10.1080/0953732042000295775
10. Jaffe AB, Palmer K. Environmental regulation and innovation: a panel data study. Review of econom-
ics and statistics. 1997;79(4):1–9.
11. Reichstein C, Sandkuhl K, Härting R. How companies can benet from enterprise architecture man-
agement in the digital transformation process-an extended research model. Enterprise Modelling and
Information Systems Architectures. 2019;14(1):1–22.
12. Cyert RM, March JG. A behavioral theory of the rm. Systems Research & Behavioral Science.
1963;2(4):81–95.
13. Winter SG. The satiscing principle in capability learning. Strategic management journal.
2000;21(10):981–96.
14. Fama EF, Jensen MC. Separation of Ownership and Control. The Journal of Law and Economics.
1983;26(2):301–25. https://doi.org/10.1086/467037
15. Christie AA, Joye MP, Watts RL. Decentralization of the rm: theory and evidence. Journal of Corpo-
rate Finance. 2003;9(1):3–36. https://doi.org/10.1016/s0929-1199(01)00036-0
16. Scharfstein DS, Stein JC. The Dark Side of Internal Capital Markets: Divisional Rent‐Seek-
ing and Inefficient Investment. The Journal of Finance. 2000;55(6):2537–64. https://doi.
org/10.1111/0022-1082.00299
1 7. Hutton AP, Marcus AJ, Tehranian H. Opaque nancial reports, R2, and crash risk⋆. Journal of Finan-
cial Economics. 2009;94(1):67–86. https://doi.org/10.1016/j.jneco.2008.10.003
18. Aktas N, Andreou PC, Karasamani I, Philip D. CEO Duality, Agency Costs, and Inter-
nal Capital Allocation Efficiency. British J of Management. 2018;30(2):473–93. https://doi.
org/10.1111/1467-8551.12277
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 22 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
19. Ambos B, Kunisch S, Leicht-Deobald U, Schulte Steinberg A. Unravelling agency relations inside the
MNC: The roles of socialization, goal conicts and second principals in headquarters-
subsidiary relationships. Journal of World Business. 2019;54(2):67–81. https://doi.org/10.1016/j.
jwb.2018.10.001
20. Mintzberg H, Waters JA. Tracking Strategy in an Entrepreneurial Firm. Academy of Management
Journal. 1982;25(3):465–99. https://doi.org/10.2307/256075
21. Keupp MM, Palmié M, Gassmann O. Achieving subsidiary integration in international innovation by
managerial “tools”. Management International Review. 2011;51:13–39.
22. George R, Kabir R. Business groups and prot redistribution: A boon or bane for rms?. Journal of
Business Research. 2008;61(9):1004–14.
23. Lou Z, Zhu M. Decision rights allocation and innovation: Evidence from China’s listed business
groups. Finance Research Letters. 2021;39:101572.
24. Stulz RM. The limits of nancial globalization. The Journal of Finance. 2005;60(4):1595–638.
25. Zhu L, Kong D. Does government transparency shape rm decentralization? Evidence from a
natural experiment in China. Emerging Markets Review. 2022;53:100915. https://doi.org/10.1016/j.
ememar.2022.100915
26. Kentchen Jr. DJ, Thomas JB, Snow CC. Organizational congurations and performance: A Compar-
ison of theoretical approaches. Academy of Management Journal. 1993;36(6):1278–313. https://doi.
org/10.2307/256812
2 7. Che Y, Zhang L. Human Capital, Technology Adoption and Firm Performance: Impacts of China’s
Higher Education Expansion in the Late 1990s. The Economic Journal. 2017;128(614):2282–320.
https://doi.org/10.1111/ecoj.12524
28. Williamson OE. Hierarchies, markets and power in the economy: an economic perspective. Industrial
and Corporate Change. 1995;4(1):21–49.
29. Colombo MG, Delmastro M. Delegation of Authority In Business Organizations:
An Empirical Test. The J Industrial Economics. 2004;52(1):53–80. https://doi.
org/10.1111/j.0022-1821.2004.00216.x
30. Shin H-H, Park YS. Financing constraints and internal capital markets: Evidence from
Korean `chaebols’. Journal of Corporate Finance. 1999;5(2):169–91. https://doi.org/10.1016/
s0929-1199(99)00002-4
31. Kim Y, Lui SS. The impacts of external network and business group on innovation: Do the types of
innovation matter?. Journal of Business Research. 2015;68(9):1964–73.
32. Saliola F, Zanfei A. Multinational rms, global value chains and the organization of knowledge transfer.
Research Policy. 2009;38(2):369–81.
33. Hsieh T, Yeh R, Chen Y. Business group characteristics and affiliated rm innovation: The case of
Taiwan. Industrial Marketing Management. 2010;39(4):560–70.
34. Stein JC. Internal capital markets and the competition for corporate resources. The Journal of
Finance. 1997;52(1):111–33.
35. Bebenroth R, Froese FJ. Consequences of expatriate top manager replacement on foreign subsidi-
ary performance. Journal of International Management. 2020;26(2):100730. https://doi.org/10.1016/j.
intman.2019.100730
36. Olley S, Pakes A. The dynamics of productivity in the telecommunications equipment industry. Journal
of the Econometric Society. 1996;64(6):1263–97.
3 7. Giroud X. Proximity and Investment: Evidence from Plant-Level Data *. The Quarterly Journal of Eco-
nomics. 2013;128(2):861–915. https://doi.org/10.1093/qje/qjs073
38. Heckman J. Shadow prices, market wages, and labor supply. Journal of the Econometric Society.
1974;42:679-94.
39. Chemmanur TJ, Kong L, Krishnan K, Yu Q. Top Management Human Capital, Inventor Mobility,
and Corporate Innovation. J Financ Quant Anal. 2018;54(6):2383–422. https://doi.org/10.1017/
s0022109018001497
40. Richardson S. Over-investment of free cash ow. Rev Acc Stud. 2006;11(2–3):159–89. https://doi.
org/10.1007/s11142-006-9012-1
41. Liu Q, Zhao C. Tax policy incentive and enterprise human capital upgrading. Economic Research.
2020;55(04):70–85.
42. Faccio M, Marchica M, Mura R. CEO gender, corporate risk-taking, and the efficiency of capital allo-
cation. Journal of corporate nance. 2016;39:193–209.
PLOS ONE | https://doi.org/10.1371/journal.pone.0319063 March 17, 2025 23 / 23
PLOS ONE Centralized allocation of decision rights and enterprise transformation and upgrading
43. Ben-Amar W, André P. Separation of ownership from control and acquiring rm performance: The
case of family ownership in Canada. Journal of Business Finance & Accounting. 2006;33(3):517–43.
44. Chandler AD. Strategy and structure: Chapters in the history of the industrial empire. Cambridge
Mass.1962;5(1):12-48.
45. Wang L, Qi J, Zhuang H. Monitoring or collusion? Multiple large shareholders and corporate ESG
performance: Evidence from China. Finance Research Letters. 2023;53:103673.
46. Liang H, Ren B, Sun SL. An anatomy of state control in the globalization of state-owned enterprises. J
Int Bus Stud. 2014;46(2):223–40. https://doi.org/10.1057/jibs.2014.35
