Data Analysis of Economic Policy Uncertainty and the Number of
Enterprise Employees based on Panel Regression Model -Taking
China's A-share Listed Companies as an Example
Yifan Zhou
*
School of Mathematics and Statistics, Jiangsu Normal University, Shanghai Road, Xuzhou, China
Keywords: Economic Policy Uncertainty, Number of Employees, Panel Regression Model, Text Analysis.
Abstract: To make a systematic analysis of the uncertainty of economic policy and then to propose effective
countermeasures has been an important subject of business management for many years. This paper selects
the asset data of China’s A-share listed companies from 2011 to 2020 and the economic policy uncertainty
index EPU formulated by Baker to create a panel regression model, focusing on studying the impact of
economic policy uncertainty on the number of employees, and trying to find out the factors that inhibit the
impacts of economic policy uncertainty on employment. EPU is an uncertainty index constructed by Baker
based on keywords in the South China Evening News, using technical means such as big data crawlers and
text analysis. The data results show that economic policy uncertainty is negatively correlated with the number
of employees. It is further found that enterprises with large financing constraints and non-state-owned
enterprises are more affected by economic policy uncertainty. Finally, based on this conclusion, suggestions
and countermeasures are made to relevant policy makers.
1 INTRODUCTION
Since the development of the Economic Policy
Uncertainty Index, scholars worldwide have
conducted research on the index and multiple aspects
of economic performance, especially on the
correlation between the index and the aspects of
macroeconomic growth. The Economic Policy
Uncertainty Index was developed by three
researchers including Scott R. Baker of Stanford
University, and is mainly used to measure the
economic conditions and policy uncertainty of major
economies in the world (Hao Xiaoyan, 2018). Their
research results point out that there is a clear
correlation between the EPU index and macro
indicators such as China's macroeconomic growth
rate and employment rate.
The analysis of the correlation between economic
policy uncertainty and investment has always been a
hot topic in economic research and study. The
research mainly focuses on the impact of economic
policy uncertainty on corporate fixed asset
investment (Li Fengyu, 2015) (Han Guogao, 2016),
innovation and R&D investment (Chen Juanjuan,
2021). Meanwhile, lots of research focuses on the
moderating effect of other variables such as
investment efficiency (Rao Pingui, 2017),
entrepreneurial subjective factors (Han Guogao,
2016), cash holdings (Wang Yizhong, 2017). The
impact of economic policy uncertainty on enterprise
investment and business environment will inevitably
lead to an impact on the ability of enterprises to
absorb employment. Qian Xueya (2018) found that
economic policy uncertainty has a significant
negative impact on the employment rate. Xin Daleng
(2018) found that when economic policy uncertainty
increases, manufacturing jobs will decrease
significantly. Foreign scholars Saud Asaad Al-
Thaqeb et al. (2019) found that the economic policy
uncertainty of their country is negatively correlated
with the productive investment of enterprises and
employment.
It is the common goal of China and all world
economies to maintain sustainable economic
development, improve the external business
environment of enterprises, and allow employees as
the main body of enterprises to receive decent wages
to improve well-being. Existing studies on economic
policy uncertainty pay more attentions to its impact
on corporate investment behavior, and most studies
406
Zhou, Y.
Data Analysis of Economic Policy Uncertainty and the Number of Enterprise Employees based on Panel Regression Model -Taking China’s A-share Listed Companies as an Example.
DOI: 10.5220/0011738600003607
In Proceedings of the 1st International Conference on Public Management, Digital Economy and Internet Technology (ICPDI 2022), pages 406-412
ISBN: 978-989-758-620-0
Copyright
c
2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
on the relationship between policy uncertainty and
employment are conducted at the national macro
level. This paper attempts to explore the substantial
impact of economic policy uncertainty on
employment in a bottom-up method by analyzing the
changes in the scale of employees in China's A-share
listed companies. At the same time, it also aims to put
forward constructive suggestions on maintaining job
security, which is one of the most important factors
for people’s well-being.
2
RESEARCH HYPOTHESES
2.1 Economic Policy Uncertainty and
Scale of Enterprise Employees
Many studies have shown that economic policy
uncertainty will increase the difficulty of business
operations. The bigger the uncertainty is, the more
difficulties to obtain funds from outside and the less
internal willingness to actively invest. Enterprises
often adopt defensive strategies to reduce
expenditures and control costs. As big part of the
operation cost, headcounts are in high likelihood to be
cut or frozen. Based on this, this paper proposes the
first research hypothesis:
H1: Economic policy uncertainty is negatively
related to the number of corporate employees.
2.2 The Moderating Effect of
Financing Constraints
Financing ability varies significantly with the scale of
the enterprise, the nature of the ownership of the
enterprise and the level of financial development in
the region where it is located. In China, private
enterprises, especially small and medium-sized
enterprises, are with much more difficulties to obtain
financing support than large state-owned enterprises
(Zou Yao, 2015). From a risk perspective, companies
with financing difficulties often choose to downsize
their business or cancel investment when facing the
challenges. Therefore, the total employment is
downsized or frozen accordingly. Based on this, this
paper proposes the second research hypothesis:
H2: The higher the degree of corporate financing
constraints is, the higher the negative correlation is
between economic policy uncertainty and the number
of corporate employees.
2.3 Moderating Effect of Ownership
Concentration
Research shows that sufficient voting rights can
ensure the company's shareholder’s high participation
in company's operations. The higher level they are
involved in the business, the more the company stick
to the value, mission and vision which are in large
degree aligned to owner’s individual pursuit.
When companies can more consider and follow
long-term goals, companies tend to pay more
attentions to sustainable development, establish more
people-oriented values, and increase their own
investment in human development costs. The
importance of stability on human resources are
usually given high weight by those enterprises on
business long term strategy. Based on this, this paper
proposes the third research hypothesis:
H3: The higher level the ownership concentration
is, the smaller the negative correlation is between
economic policy uncertainty and employee size.
2.4 The Effect of Equity Nature
In this paper, listed companies are divided into state-
owned enterprises and non-state-owned enterprises
according to the nature of equity. The existing state-
owned enterprises in China have relatively large scale
of assets and number of employees. The state-owned
enterprises are one of the fundamental forces to the
national economy and people's livelihood. Under the
circumstance of high economic policy uncertainty,
state-owned enterprises have the higher ability and
responsibility of achieving the goals not only on
economy and but also on social employment stability.
Comparing to the other type of enterprise, state-
owned enterprises are in general with good conditions
on financing. Based on this, this paper proposes the
fourth hypothesis:
H4: Compared with non-state-owned enterprises,
the number of employees in state-owned enterprises
is less affected by economic policy uncertainty.
3 RESEARCH DESIGN
3.1 Sample Selection and Data Sources
This paper selects the data of listed companies in
Shanghai and Shenzhen A-share companies from
2011 to 2020, and draws on other research (Xu
Yekun, 2020) to process the selected data as follows:
(1) exclude ST and ST* companies; (2) exclude
Data Analysis of Economic Policy Uncertainty and the Number of Enterprise Employees based on Panel Regression Model -Taking China’s
A-share Listed Companies as an Example
407
financial enterprises; (3) eliminate corporate data
with missing or obviously wrong main variable
information, and perform 1% abbreviated processing
for all continuous variables. This method refers to the
practice of Chen Juanjuan et al. (2021) and reduces
the impact of extreme values on the regression results
by shortening the tail.
It ends up with 13760 sample observations by
using this methodology. The above enterprise data all
come from the CSMAR database. The economic
policy uncertainty index selects the EPU index
formulated and published by Baker et al. The index
data is downloaded from the PU website
(http://www.policyuncertainty.com).
3.2 Definition and Measurement of
Variables
3.2.1 Explained Variables
Number of employees of the enterprise (STAFF).
Because the number of employees of listed
companies in China varies greatly, this paper adopts
the method of calculating the natural logarithm of the
number of employees to measure the number of
employees in the enterprise.
3.2.2 Explanatory Variables
(1) Economic Policy Uncertainty (EPU). This index
(EPU) is constructed by Baker et al. based on the
news index of two major newspapers in China. This
paper refers to Qi Jianhong et al. (2020) to calculate
the arithmetic mean of the monthly EPU index to
obtain the annual EPU index and adopts the one-
period lag EPU index as an indicator of economic
uncertainty for robustness testing.
(2) Financing constraints (SA). This variable (SA)
is the corporate financing constraint index, which is
collected from the Cathay Pacific database. The
higher the SA index, the greater the corporate
financing constraint.
(3) Equity concentration (H). This variable (H) is
the sum of the squares of the shareholding ratios of
the top 5 major shareholders of the company. The
larger the h index, the higher the ownership
concentration.
3.2.3 Control Variables
This paper refers to previous studies to determine
macro-level control variables and enterprise-level
control variables respectively. The macro control
variable is the per capita GDP of the province where
each enterprise is located; the enterprise control
variable includes enterprise scale (SIZE), enterprise
leverage ratio (LEV), enterprise return on assets
(ROA), and enterprise scale is measured by the
logarithm of the total enterprise assets. Corporate
leverage is measured by the ratio of total liabilities to
total assets and return on assets is measured by the
ratio of after-tax net profit to total assets.
3.3 Empirical Model
In order to study the impact of economic policy
uncertainty on the number of corporate employees,
this paper uses the following model to test the
assumptions proposed above.
STAFF
i,t
=α
0
+α
1
EPU
t
+α
2
Z
i,t
+μ
i
+γ
i
+ε
i,t
(1)
Model (1) is the basic model to test the research
focus of this paper: the correlation test between
economic policy uncertainty and the number of
corporate employees, where i represents an
individual, namely a listed company, t represents the
year, α
0
represents a constant term, and μ
i
represents
a fixed term effect, γ
i
stands for time effect, ε
i,t
stands
for random error, and Z
i,t
stands for a series of control
variables, namely firm size, leverage ratio, financing
constraints, and per capita GDP. According to H1,
this paper expects the model (1) variable EPU
coefficient to be negative, that is, there is a negative
correlation between economic policy uncertainty and
the number of corporate employees.
And according to H2 and H3 respectively, namely
examining the moderating effects of corporate
financing constraints and equity concentration on the
number of employees from economic policy
uncertainty, the model is further adjusted to obtain:
STAFF
i,t
=
α
0
+β
1
EPU
t
+β
2
SA
i,t
+β
3
SA
i,t
*EPU
t
+β
4
Z
i,t
+μ
i
+γ
i
+ε
i,t
(2)
STAFF
i,t
=
α
0
+𝜂
1
EPU
t
+𝜂
2
H
i,t
+𝜂
3
H
i,t
*EPU
t
+𝜂
4
Z
i,t
+μ
i
+γ
i
+ε
i,t
(3)
In model (2), SA
i,t
*EPU
t
represents the interaction
term between economic policy uncertainty and
corporate financing constraints.
In model (3), H
i, t
*EPU
t
represents the interaction
term between economic policy uncertainty and
corporate ownership concentration.
Referring to the practice of Jiang Teng et al.
(2018), when the interaction term is significant, the
higher the coefficient of the interaction term, the
stronger the moderating effect.
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
408
Table 1: Benchmark regression results.
(1) (2) (3) (4)
lnY lnY lnY lnY
EPU -0.025
***
-0.021
***
-0.038
***
-0.038
***
(-18.513) (-11.136) (-9.387) (-4.710)
SIZE 0.423
***
0.432
***
0.447
***
0.447
***
(78.638) (72.198) (71.024) (26.911)
LEV 0.224
***
0.214
***
0.186
***
0.186
**
(6.957) (6.621) (5.708) (2.513)
ROA -0.165
**
-0.190
***
-0.182
**
-0.182
(-2.335) (-2.677) (-2.551) (-1.516)
lnK -0.075
***
0.095
***
0.095
(-3.274) (2.676) (1.381)
_cons -5.827
***
-5.282
***
-7.595
***
-7.595
***
(-34.868) (-22.397) (-18.057) (-8.197)
N 13760 13760 13760 13760
r2 0.439 0.439 0.445 0.445
F 2418.405 1938.385 763.371 126.413
model selection FE FE FE FE
time fixed effects NO NO YES YES
firm fixed effects NO NO NO YES
*** 1% ** 5% * 10%
4 EMPIRICAL ANALYSIS
4.1 Results of Regression Analysis
4.1.1 Economic Policy Uncertainty and
Number of Employees
Using the model (1) formulated in Section 3.3 to test,
the regression results of the impact of economic
policy uncertainty on the number of employees of
enterprises are shown in Table 1. The basic model of
per capita GDP, the data results show that the
economic policy uncertainty is negatively correlated
with the number of employees, the preliminary
verification H1, columns (2) (3) (4) are all added to
the macro control variables, columns (3) (4) is added
to the time fixed effect, and column (4) is added to the
firm fixed effect. As can be seen from the figure, after
adding control variables and fixed effects in turn,
economic policy uncertainty is still negatively
correlated with the number of employees, so
hypothesis 1 can be tested.
4.1.2 The Moderating Effect of Financing
Constraints
According to the model (2), the moderating effect of
financing constraints on the correlation between
economic policy uncertainty and the number of
employees is tested, and the SA index is used as a
measure of corporate financing constraints. In the
case that SA is negative, the closer SA is to 0, the
greater the financing constraints faced by enterprises
can be considered. Therefore, in order to test the
moderating effect of financing constraints, the
interaction term between financing constraints and
economic policy uncertainty index is added to the
model. The larger the coefficient of the interaction
term, the greater the moderating effect (Jiang Teng,
2018). The index in column (1) of Table 2 shows that
the coefficient of EPU is -0.038, which is
significantly negative at the 1% level. Meanwhile, the
interaction term of economic policy uncertainty and
corporate financing constraint (SA index) (SA*EPU)
coefficient is 0.016, which is significantly positive at
the 10% level, indicating that with the increase of
corporate financing constraints, economic policy
Data Analysis of Economic Policy Uncertainty and the Number of Enterprise Employees based on Panel Regression Model -Taking China’s
A-share Listed Companies as an Example
409
uncertainty has a greater negative effect on the
number of companies. So, hypothesis 2 can be tested.
4.1.3 The Moderating Effect of Ownership
Concentration
According to the model (3), the moderating effect of
ownership concentration on the correlation between
economic policy uncertainty and the number of
employees is tested. The moderating effect of
ownership concentration is shown in the second
column of Table 2. The coefficient of EPU is -0.036,
which is significantly negative at the level of 1%, but
the coefficient of interaction between ownership
concentration and economic policy uncertainty index
is -0.01, which is not significant, indicating that
ownership concentration has no moderating effect on
the correlation between economic policy uncertainty
and number of corporate employees, so hypothesis 3
cannot be tested.
4.1.4 The Effect of Equity Nature
According to the different nature of equity, this paper
divides the enterprise samples into two groups of
6350 state-owned enterprises and 7030 non-state-
owned enterprises. According to H4, this paper
assumes that the different nature of the company's
equity will affect the company's business strategy
when facing increasing economic policy uncertainty
and consequently enterprise take varies of approaches
on employment. From the regression data in Table 3,
it can be seen that among state-owned enterprises and
non-state-owned enterprises, the coefficients of EPU
are -0.030 and -0.051 respectively, and they are
significantly negatively correlated at the 1% level.
Therefore, economic policy uncertainty has an
inhibitory effect on the number of employees in both
types of enterprises, however non-state-owned
enterprises are more affected by economic policy
uncertainty. So, hypothesis 4 can be tested.
Table 2: Moderating effect test.
(1) (2)
lnY lnY
EPU -0.038
***
-0.036
***
(-3.282) (-4.221)
SA index -0.084
(-0.574)
EPU*SA 0.016
*
(1.809)
EPU*H -0.010
(-0.488)
H 0.078
(0.474)
SIZE 0.445
***
0.447
***
(26.745) (26.941)
LEV 0.171
**
0.184
**
(2.280) (2.490)
ROA -0.179 -0.182
(-1.488) (-1.534)
lnK 0.092 0.095
(1.332) (1.387)
_cons -7.495
***
-7.614
***
(-8.082) (-8.276)
N 13760 13760
r2 0.446 0.445
F 109.685 110.316
time fixed effects YES YES
firm fixed effects YES YES
*** 1% ** 5% * 10%
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
410
Table 3: The effect of equity nature.
state-owne
d
non-state-owne
d
enter
p
rise enter
p
rise
lnY lnY
EPU -0.030
***
-0.051
***
(-3.020) (-3.665)
SIZE 0.429
***
0.464
***
(
15.072
)
(
21.731
)
LEV 0.054 0.212
**
(
0.437
)
(
2.034
)
ROA -0.273 -0.168
(-1.146) (-1.263)
ln
K
0.001 0.219
*
(
0.015
)
(
1.785
)
_
cons -5.952
***
-9.464
***
N 6350 7030
r2 0.349 0.501
F 35.431 88.647
time fixed effects YES YES
firm fixed effects YES YES
*** 1% ** 5% * 10%
4.2 Robustness Check
This paper refers to previous studies to solve the
endogeneity problem by using lagged variables.
Using the variable of economic policy uncertainty
with a lag of one period, and using the method of
negative binomial regression data, the results show
that after excluding the endogeneity problem, the
economic policy uncertainty and the number of
employees still show a significant negative
correlation. Secondly, this paper uses the method of
expanding the sample size to test the robustness of the
overall regression results. The sample in this paper
comes from the data of listed companies in China’s
A-shares from 2011 to 2020. In order to expand the
sample size, this paper adds the data of B-share listed
companies in China. It shows that even if the sample
size is expanded, economic policy uncertainty is still
significantly negatively correlated with the number of
employees, so the basic hypothesis of H1 is tested
again.
5 CONCLUSIONS
The analysis above draws the following conclusions:
Firstly, economic policy uncertainty has a significant
negative correlation with the number of corporate
employees i.e., the higher the EPU is, the lower the
scale of enterprise employment goes. Secondly, the
bigger the financings are constrained, the more
negative correlation goes between the number of
employees and the uncertainty of economic policy.
Thirdly, the level of ownership concentration has no
significant effect on the negative correlation between
the uncertainty of economic policy and the number of
enterprise employees. Fourthly, differences in equity
ownership can also effectively moderate the negative
correlation between economic policy uncertainty and
the number of employees. The state-owned
enterprises are less affected by economic policy
uncertainty than non-state-owned enterprises.
Based on the above analysis, in order to minimize
impact of the economic policy uncertainty on
employments, this study propose to the policy makers
the followings
1. Strengthen the communication mechanism
between the government and enterprises in order to
reduce the uncertainty of economic policies. Establish
green channels for effective and efficient
communication. Dynamically track the uncertainty of
economic policies and set up warning mechanisms.
Policy makers can take countermeasures before the
policy uncertainty index reaches the specified limit
and thus to reduce the EPU
2. Strengthen the balance of the financial market
development. In China there is still a big gap on the
degree of financial market development between the
coastal areas and the central & western regions.
Provide more support to the less developed regions
and small-medium sized enterprises. Consequently,
the national economic development gets more
balanced and employment gets more stabilized.
3. Strengthen the research at the macro level on
the impact of enterprise ownership difference on
Data Analysis of Economic Policy Uncertainty and the Number of Enterprise Employees based on Panel Regression Model -Taking China’s
A-share Listed Companies as an Example
411
employment and economic development. With the
rapid changes of the world, the uncertainties will
continue to bring challenges to enterprises. The
government should examine and redefine the role of
state-owned enterprises in safeguarding the country's
core interests, stabilizing the economic foundation,
and maintaining employment.
REFERENCES
Al‐Thaqeb S A, Algharabali B G, Alabdulghafour K T. The
pandemic and economic policy uncertainty. IntJ Fin
Econ. 2020;1–11.
Chen Juanjuan, Zhao Hongyan, Yang Xiaoli. Economic
Policy Uncertainty, Financing Constraints and
Enterprise Innovation [J]. Forecast, 2021, 40(02): 55-60.
Hao Xiaoyan, Han Yijun, Shi Zizhong. Analysis of the
Impact of Economic Policy Uncertainty on China's Grain
Trade [J]. Economic Issues Exploration, 2018(3):10.
Han Guogao, Hu Wenming. Macroeconomic Uncertainty,
Entrepreneur Confidence and Fixed Asset Investment: A
Systematic GMM Method Based on my country's Inter-
provincial Dynamic Panel Data [J]. Finance and
Economics, 2016(03): 79-89.
Jiang Teng, Zhang Yongji, Zhao Xiaoli. Economic Policy
Uncertainty and Corporate Debt Financing [J].
Management Review, 2018, 30(3):11.
Li Fengyu, Yang Mozhu. Will Economic Policy
Uncertainty Constrain Corporate Investment? An
Empirical Study Based on China's Economic Policy
Uncertainty Index [J]. Financial Research, 2015(04):
115-129.
Qian Xueya, Jiang Zhuoyu, Hu Qiong. Research on the
Influence of Social Insurance Contributions on
Enterprise Employment Wages and Scale [J]. Statistical
Research, 2018, 35(12): 68-79. DOI: 10.19343/j.cnki.
11-1302/c.2018.12.006.
Qi Jianhong, Yin Da, Liu Hui. How Does Economic Policy
Uncertainty Affect Enterprise Export Decisions? --
Based on the Perspective of Export Frequency [J].
Financial Research, 2020(5):19.
Rao Pingui, Yue Heng, Jiang Guohua. Research on
Economic Policy Uncertainty and Enterprise
Investment Behavior [J]. World Economy, 2017,
40(02): 27-51.
Wang Yizhong, Yuan Jun. The mechanism and effect of
macroeconomic risks affecting the company's cash
holdings [J]. Finance and Economics, 2017(09): 56-64.
DOI: 10.13762/j.cnki.cjlc.2017.09. 005.
Xin Daleng. Economic Policy Uncertainty and
Employment of Industrial Enterprises [J]. Industrial
Economics Research, 2018(05): 89-100.DOI:
10.13269/j.cnki.ier.2018.05.008.
Xu Yekun, Wang Yuanfang, An Suxia. Replacement of
Local Officials and Staff Allocation in Enterprises [J].
Comparison of Economic and Social Systems,
2020(2):11.
Zou Yao. Research on the Transformation and Upgrading
of Small and Medium Private Enterprises——Taking
Wenzhou Rongde Valve Co., Ltd. as an example [D].
Zhejiang University of Technology, 2015.
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
412
