Politics-Finance Nexus Under Pandemic Shock Based on DID Model:

From the Modern Slavery Perspective

Xiangdong Chen

1, *, #

and Jincheng Zuo

2, #

School of Humanities and Social Sciences, Xi’an Jiaotong-Liverpool University, China

International Business School Suzhou, Xi’an Jiaotong-Liverpool University, China

# These authors contributed equally to this work.

Keywords: Empirical Data Analysis, DID model, Politics-Finance Nexus, Public Management.

Abstract: The COVID-19 pandemic has created uncertainty in financial conditions around the world and generated risk-

based vulnerabilities for households and societies. Workers in severely affected areas may face worse work

conditions and imbalances due to the pandemic. Possible potential labour exploitation in those regions is a

product and manifestation of this status quo. The existing research is still on the macro impact of the pan-

demic. This article attempts to analyze the latent unfair treatment that employees may receive by constructing

modern slavery as a theoretical guide for analysis. There are 1760 sample observations of seven areas, running

from 2016 through 2020. We used Shanghai as the treatment group it was hit hard by COVID-19 in 2020.

The other six provinces, as the control group, each had less than 200 confirmed cases by comparison. Since

the emergence of the pandemic is an exogenous shock, we are interested in the alterations in modern slavery

before and after the experiment, the DID Model is applied for measuring the impact effect of the experimental

group in this study. The parallel trend test was used to evaluate the validity of the DID model, and regression

was utilized to determine the influence of pandemic shock.

1 INTRODUCTION

COVID-19 has spread globally and is considered the

most serious health crisis in the 21st century (Yu,

2020). In reaction to the pandemic, all governments

have implemented lockdowns and isolation, causing

local and international cooperation problems. The

pandemic and lockdown are new financial catastro-

phes that will raise worldwide poverty. According to

World Bank data, during these blockades, the manu-

facturing industry was closed and the global supply

chain was disturbed (Vidya, 2020). There are signifi-

cant and instant influences caused by lockdowns on

economic activities in China (Chen, 2020).

China has been studied by management and pub-

lic health researchers, but not from a micro-political

economics standpoint. While the pandemic damaged

the domestic economy and finance, worldwide trade

and collaboration, many individuals lost their jobs

temporarily or permanently, which leads to increasing

inequality (Shibata, 2020). During the pandemic,

many unequal treatments faced by labour have been

reported. However, existing theories struggle to ex-

plain how economic shifts affect workers during

COVID-19. Facing uninterrupted scattering of out-

breaks, remote work and home office are usually al-

ternatives to clocking in, so the exploitation and threat

of unemployment are more likely to occur. This study

explores the political economy of Shanghai employ-

ees in 2020 when the pandemic broke out by adopting

a novel yet emerging method, modern slavery, and

the Difference-in-Difference (DID) model to support

our findings.

This article will firstly construct modern slavery

through a critical analysis of current debates and em-

pirical evidence from the Marxist perspective, and ex-

plore the impact of the pandemic on modern slavery.

Second, this paper employee DID model to quantita-

tively analyse whether modern slavery occurs during

severe pandemic shock. The paper uses data from

CSMAR. 2016-2020 is the sample's span. 1080

Shang-hai observations and 680 from the other six

provinces total 1760. Since the pandemic is an exog-

enous shock, we are interested in "modern slavery"

before and after the experiment; consequently, the

DID Model is ideal for analyzing the experimental

group's impact effect in this study. This research em-

ployed the parallel trend test to evaluate the DID

Chen, X. and Zuo, J.

Politics-Finance Nexus Under Pandemic Shock Based on DID Model: From the Modern Slavery Perspective.

DOI: 10.5220/0011737100003607

In Proceedings of the 1st International Conference on Public Management, Digital Economy and Internet Technology (ICPDI 2022), pages 367-372

ISBN: 978-989-758-620-0

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

367

model's validity and regression to determine the im-

pact of the pandemic shock.

2 THEORIZING MODERN

SLAVERY

Although the global living conditions have improved

due to economic, trade, and political globalization,

and the poverty rate has fallen sharply, many negative

effects have been brought (Chen, 2010). These influ-

ences are mainly on workers, including poor working

conditions and unsafe working environments

(Selwyn, 2019). With the importance of human rights

and the critical attitude of global supply chains, more

scholars are focusing on the term modern slavery to

help vulnerable groups affected by imbalance and in-

equality through laws, social supervision, and other

measures. Modern slavery has been explored more

frequently in political economy, history, society, and

law, owing to the International Labour Organization

(ILO) and the Walk Free Foundation (Benstead,

2020). It has become a vague concept of extreme ex-

ploitation of labour, by taking advantages of the vul-

nerabilities and weaknesses of employees, including

not only threats and violent behaviour, but also ma-

nipulating victims to claim that they are voluntary

(Machura, 2019).

As the definition of modern slavery is not clear,

scholars tried to analyze and define this term from a

historical perspective, and offer an understanding be-

tween ‘new’ and ‘old’ (Manzo, 2005). Manzo first

agreed on the Marxist theory of slavery as a special

form of exploitation of unpaid labour, and claimed the

fact that workers have lost their freedom and right to

choose due to violence, then conclude slavery as un-

paid forced labour (Manzo, 2005). At the same time,

by examining the international laws related to human

rights and slavery, he defined the new modern slavery

as control without considering autonomy, and con-

sisting of the use of violence and threats, making la-

bour their personal freedom and rights of choice, and

the forced unpaid work (Manzo, 2005). Manzo's def-

inition and theory have largely helped construct the

concept of modern slavery. The origins of modern

slavery are explained by political economy: the pro-

duction and trade of commodities involve the consid-

eration of labour costs in the global value chain,

which makes employers more inclined to choose

modern slavery; at the same time, the expansion of

global capitalism promotes unequal conditions of ex-

change, and it ultimately leads to the widespread ex-

istence of modern slavery (Manzo, 2005).

In forced labour, however, meagre salaries are

possible, but they should not be viewed as a normal

and liberal relationship between employers and em-

ployees. His study offers a restricted definition of

modern slavery. From a post-colonial perspective,

this definition is European-centric, focused solely on

human trafficking and child labour. His method

doesn't applicable to sweatshops and Marxist con-

cepts of surplus value, where individuals make a

modest wage and are exploited and poor. Therefore,

wages in the relationship between employers and em-

ployees should be emphasized.

Wages help distinguish between contract labour

and slavery in the 19th and 20th centuries (Jones,

2019). Symbolic wages that can't meet basic demands

and commoditized labour are indications of modern

slavery in transnational labour (Jones, 2019). Histor-

ically, workers after signing a labour contract work

according to the employer's guidelines within a cer-

tain contract duration, and the labour contract is

merely a tool to legalize exploitation, especially when

employees are not supported by their own country's

laws since they work overseas (Jones, 2019). Free-

dom and non-freedom, or slavery and non-slavery,

cannot be totally separated in history (Jones, 2019).

This is a continuum where employees have been en-

slaved.

Overall, modern slavery can be narrowly under-

stood as the exploitation of people whose personal

freedom is deprived within the supply chain from the

extraction of raw materials to the final customer to

provide services or products, or those who are forced

to work without payment (Stefan, 2015); it could also

be broadly understood as the process where people

are forced to work or whose human rights have been

violated (Lucas, 2020). However, current discussions

are usually limited to research on global supply

chains, such as human trafficking, differences in

working environment and wages caused by imbal-

ance and inequality between suppliers and transna-

tional companies, and sweatshops, while domestic

situations are ignored. In fact, modern slavery within

the countries is also common, especially in develop-

ing countries, such as the 996 working system of

high-tech companies in China, which is an inhumane

working system (12 hours a day, 6 days a week) and

affects many people’s health including sudden death

and cancer (Wang, 2020).

Due to the pandemic's increased demand, modern

slavery audits have been neglected (Trautrims, 2020).

Medical supply pressure, including masks and gloves,

placed the slavery audit on hold (Feinmann, 2020).

Moreover, cooperation partners and business institu-

tions in developed economies often choose to infringe

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

368

on the interests of vulnerable labourers (Trautrims,

2020). Pandemic implications of modern slavery in-

clude labour displacement, changes in migratory

worker rules, and an increase in homeless trafficking

(Lucas, 2020). In addition to the perspective of eco-

nomic politics, the difficulties posed by modern slav-

ery have been selectively or intentionally filtered out

relevant, and institutions and companies cannot re-

ceive any information (Lucas, 2020). These vulnera-

ble groups are hit most by the pandemic.

This paper argues that a broader modern slavery

system should be defined, which can be linked to the

global poverty chain, that is, any employer restricts

employees’ normal life activities, and employees’

wages are not proportional to the working content and

intensity of the work, and exploitation in an unhealthy

working environment. Under this alternative defini-

tion, the content of modern slavery can not only in-

clude the dark side brought about by the global value

chain, but also exposes the domestic problems of cap-

italist globalization. Based on this alternative under-

standing of modern slavery, a DID model has been

formulated to support our arguments.

3 HYPOTHESIS, EMPIRICAL

MODEL AND DATA

DESCRIPTION

Through the exploration of modern slavery, the indi-

cators related to wages are expected to provide a basis

for further quantitative analysis of the impact of the

pandemic on modern slavery. Thus, this paper pro-

poses the following two hypotheses:

Hypothesis 1: COVID-19 increases the gap be-

tween "directors, supervisors, executives" pay and

employee pay.

Hypothesis 2: COVID-19 has an adverse impact

on average employee salaries.

This paper uses the Difference-in-Difference ap-

proach to test the hypotheses above and examine the

impact of the COVID-19 outbreak as a quasi-natural

experiment on the extent of 'modern slavery' among

employees in selected areas of China. We use inter-

provincial data for selected areas of China and distin-

guish between areas severely affected by the outbreak

(treatment group) and areas not severely affected

(control group) based on the cumulative number of

confirmed cases.

According to Yang and Wang (Yang, 2021),

COVID-19 started in China on 20 January 2020, so

we used Shanghai as the treatment group because its

cumulative confirmed cases reached 1,500 by the end

of 2020. The control group included Guizhou, Jilin,

Gansu, Qinghai, Ningxia, and Hainan provinces as

they each had less than 200 cumulative confirmed

cases in 2020. The model specification is shown as

follows:

𝑴𝑺

𝒊𝒕

= 𝜶 + 𝜷𝑹𝒆𝒈𝒊𝒐𝒏

𝒕

× 𝑪𝒐𝒗𝒊𝒅

𝒊𝒕

+ 𝜸𝑹𝒆𝒈𝒊𝒐𝒏

𝒕

+𝜹𝑪𝒐𝒗𝒊𝒅

𝒊𝒕

+ 𝜽𝑪𝒐𝒏𝒕𝒓𝒐𝒍

𝒊𝒕

+ 𝝁

𝒊

+ 𝜺

𝒊𝒕





Equation (1) is a DID estimation model that takes into

account individual fixed effects. 𝑀𝑆



is a measure of

the degree of 'modern slavery', which consists of two

indicators: a) ratio of the compensation of directors,

supervisors and executives to the salaries of employ-

ees (abbreviated as S/E); b) average employee salary

(AvPay). 𝑅𝑒𝑔𝑖𝑜𝑛



× 𝐶𝑜𝑣𝑖𝑑



is an interaction term.

𝑅𝑒𝑔𝑖𝑜𝑛



is a dummy variable for the treatment group,

corresponding to a value of 1 for Shanghai and 0 for

other regions. 𝐶𝑜𝑣𝑖𝑑



is a dummy variable for the

treatment effect period, as COVID-19 broke out in

early 2020, and it takes the value of 1 after 2020 and

0 before. 𝐶𝑜𝑛𝑡𝑟𝑜𝑙



indicates control variables, in-

cluding a) Ownership, with state-owned enterprises

taking a value of 1 and private enterprises taking a

value of 0. Foreign enterprises and joint ventures are

not considered; b) Firm size, expressed as the natural

logarithm of the firm's total assets (LnAssets). 𝜇



an individual fixed effect and 𝜀



is a random error

term.

The data used in this paper are obtained from the

CSMAR database. The duration of the sample spans

from 2016 to 2020. There are 1080 sample observa-

tions of Shanghai and 680 observations for the other

six provinces, giving a total of 1760 observations. In

order to eliminate the dimensional relationship be-

tween variables and enhance comparability, the data

used in this paper have been standardized. To mitigate

the influence of outliers, variables are winsorized at

the 1% and 99% levels.

4 EMPIRICAL RESULTS

4.1 Descriptive Statistics

The objective of descriptive statistical analysis is to

reduce a set of complicated data to a few descriptive

values. It is a general description of the existing data

set that facilitates an understanding of the data's cen-

tral and dispersed patterns.

Politics-Finance Nexus Under Pandemic Shock Based on DID Model: From the Modern Slavery Perspective

369

Table 1: Descriptive Statistics.

(1) (2) (3) (4) (5)

VARIABLES N mean s

min max

Yea

1,760 2,018 1.349 2,016 2,020

SE 1,760 52.90 26.67 0 94.51

AvPa

1,760 69.75 25.61 9.573 100

Regions 1,760 1.614 0.487 1 2

LnAssets 1,760 22.42 1.584 18.29 29.92

Owne

1,760 0.433 0.496 0 1

Figure 1: Parallel Trend for S/E and AvPay before and after 2020.

Table 1 shows the descriptive statistical analysis

results of the sample, including sample size, variable

mean, standard deviation and maximum value. This

study is most concerned with the explained variables,

namely the ratio of executive and employee salaries

(SE) and the average employee salary (AvPay). All

the data have been standardized. It is not difficult to

find that both SE and AvPay have great range differ-

ences. Therefore, to avoid the possibility of outliers

affecting the analysis results, this study adopts winso-

rization. In terms of the degree of data dispersion,

both of the two key variables have large standard de-

viations, which to a certain extent reflects that the in-

dicators related to employee wages ("modern slav-

ery" proxies) have fluctuated greatly between re-

gions, and before and after the pandemic. In addition,

the specific cause behind this is analyzed in the fol-

lowing sections with the DID model to show implica-

tions.

4.2 Parallel Trend Test

A prerequisite assumption for DID is that the trend of

change in the treatment and control groups should be

(almost) the same prior to the event. To do this, we

further examined changes in trends for the 4 years be-

fore the COVID-19 outbreak in 2020 up to the final

year of the sample. The analysis results are reported

in Figure 1. It appears that the trends in the treatment

and control groups before the outbreak were almost

identical and not significantly different. After 2020,

the 'modern slavery' indicator for the treatment group

increased/decreased significantly compared to the

control group. The sample, therefore, passes the par-

allel trend test required for double difference estima-

tion.

4.3 Baseline Regression

This section examines the pandemic’s impact on

"modern slavery" in China. Table 2 presents the re-

sults of examining the pandemic shock using equation

(1) including firm-level control variables.

The pandemic shock (Region×Covid) has a positive

effect on the ratio of directors', supervisors', and ex-

ecutives' pay to employee salaries (S/E) at the 90%

confidence level, meaning it widens the gap between

directors’ and executives' pay and employee pay in

pandemic-affected regions (Shanghai in this study),

although the effect is marginal. Moreover, the pan-

demic reduces the average income of employees in

badly afflicted areas compared to those that survived

the big blow on the 95% confidence level.

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

370

Table 2: Baseline Regression Results.

(1) (2)

VARIABLES S/E AvPa

ion×Covi

5.253* -5.730**

(3.106) (2.838)

ion 4.822*** 23.61***

(1.400) (1.197)

Covi

-2.032 0.238

(2.543) (2.261)

Ownershi

14.16*** 7.762***

(

1.337

)

(

1.083

)

LnAssets -2.946*** 3.871***

(

0.487

)

(

0.293

)

Constant 109.9*** -34.10***

(

10.61

)

(

6.391

)

Observations 1,726 1,743

R-square

0.079 0.306

Robust standard errors in parentheses

*** p<0.01, ** p<0.05, * p<0.1

5 DISCUSSION AND

REMEDIATING MODERN

SLAVERY

This paper corroborates two assumptions using data

from limited regions, leading to two credible results:

first, at the 90% confidence level, the pandemic shock

(RegionCovid) has a positive effect on the ratio of di-

rectors', supervisors', and executives' pay to employee

salaries (S/E), implying that it widens the discrepancy

between the senior executives' pay and employee pay

in pandemic-affected regions (Shanghai in this

study), though the effect is marginal; second, on a

95% confidence level, the pandemic decreases the av-

erage salary of employees in seriously affected areas

compared to those who survived the big hit. These

findings account for business size and property rights.

This is consistent with the results of our preliminary

parallel trend test. This result supports our previous

statement that modern slavery may have increased

during the pandemic, with employees struggling to

meet some of their material and health needs, and that

there may have been an increase in the price of some

necessities in hard-hit areas, further reducing employ-

ees' disposable income and adding to the evidence of

modern slavery. Possible factors include the difficulty

of auditing and examining enterprise employment

systems during the outbreak and the economic impact

of the pandemic forcing enterprises to decrease em-

ployment costs. The execution of policies to address

the pandemic has been cautious in boosting wage in-

come growth, and there is insufficient autonomy in

wage distribution at the grassroots level (Chang,

2021), which confirms the conclusions of this work.

This study shows the negative consequences of global

value chains and globalization of capital on disadvan-

taged employees through the lens of wages, making

modern slavery visible in modern society in the con-

text of the pandemic crackdown and drawing atten-

tion to it. Our experiment has the benefit of having

adequate sample data. However, because COVID-19

has spread throughout the country and even around

the world, it is difficult to pinpoint locations that are

"not substantially afflicted by the pandemic." Be-

cause of the effective management of China's central

and local governments, several places (six provinces

in this study) were not severely impacted during the

outbreak's early stages, and they were chosen as the

control group. Future research should focus on more

provinces or even country comparisons and, if acces-

sible, evaluate home working conditions and em-

ployee mental health.

REFERENCES

Benstead, A. V, Hendry, L. C. and Stevenson, M. (2020)

‘Detecting and remediating modern slavery in supply

chains: a targeted audit approach’, Production Planning

& Control, pp. 1–22. doi: 10.1080/09537287.

2020.1795290.

Chen, Q. et al. (2020) ‘Economic Effects of Lockdown in

China’, in Impact of COVID-19 on Asian Economies

and Policy Responses. WORLD SCIENTIFIC, pp. 3–

10. doi: doi:10.1142/9789811229381_0001.

Politics-Finance Nexus Under Pandemic Shock Based on DID Model: From the Modern Slavery Perspective

371

Chen, S. and Ravallion, M. (2010) ‘The Developing World

is Poorer than We Thought, But No Less Successful in

the Fight Against Poverty*’, The Quarterly Journal of

Economics, 125(4), pp. 1577–1625. doi: 10.1162/

qjec.2010.125.4.1577.

Chang, F., Xiao, T. and Liu, J. (2021) ‘Effect evaluation of

wage and income distribution policies in response to the

pandemic (Part Two)’, China's Human Resources So-

cial Security, pp. 48-49.

Feinmann, F. (2020) ‘The Scandal of Modern Slavery in the

Trade of Masks and Gloves’, BMJ (Clinical Research

ed.). doi: 10.1136/bmj.m1676.

Jones, G. (2019) ‘The construction of a slave identity: an

examination of the dual identity of indentured labourers

across the Western Pacific’, Labor History, 60(5), pp.

540–557. doi: 10.1080/0023656X.2019.1584272.

Lucas, B. and Landman, T. (2020) ‘Social listening, mod-

ern slavery, and COVID-19’, Journal of Risk Research,

pp. 1–21. doi: 10.1080/13669877.2020.1864009.

Machura, S. et al. (2019) ‘Recognizing Modern Slavery’,

Journal of Human Trafficking, 5(3), pp. 201–219. doi:

10.1080/23322705.2018.1471863.

Manzo, K. (2005) ‘Modern slavery, global capitalism &

deproletarianisation in West Africa’, Review of African

Political Economy, 32(106), pp. 521–534. doi:

10.1080/03056240500467013.

Shibata, I. (2020) ‘The distributional impact of recessions:

The global financial crisis and the COVID-19 pandemic

recession’, Journal of Economics and Business, p.

105971. doi: https://doi.org/10.1016/j.jeconbus.2020

.105971.

Selwyn, B. (2019) ‘Poverty chains and global capitalism’,

Competition and Change, 23(1), pp. 71–97. doi:

10.1177/1024529418809067.

Stefan, G., Alexander, T. and Zoe, T. (2015) ‘Modern slav-

ery challenges to supply chain management’, Supply

Chain Management: An International Journal, 20(5),

pp. 485–494. doi: 10.1108/SCM-02-2015-0046.

Trautrims, A. et al. (2020) ‘Survival at the expense of the

weakest? Managing modern slavery risks in supply

chains during COVID-19’, Journal of Risk Research,

23(7–8), pp. 1067–1072. doi: 10.1080/13669877

.2020.1772347.

Vidya, C. T. and Prabheesh, K. P. (2020) ‘Implications of

COVID-19 Pandemic on the Global Trade Networks’,

Emerging Markets Finance and Trade, 56(10), pp.

2408–2421. doi: 10.1080/1540496X.2020.1785426.

Wang, J. J. (2020) ‘How managers use culture and controls

to impose a “996” work regime in China that constitutes

modern slavery’, Accounting and Finance, 60, pp.

4331–4359. doi: 10.1111/acfi.12682.

Yang, Z and Wang, S (2021) ‘Systemic Financial Risk Con-

tagion of Global Stock Market Under Public Health

Emergency: Empirical Evidence from COVID19 Pan-

demic’, Economic Research Journal, 8, pp. 22-38

Yu, D. E. C., Razon, L. F. and Tan, R. R. (2020) ‘Can global

pharmaceutical supply chains scale up sustainably for

the COVID-19 crisis?’, Resources, Conservation and

Recycling

, 159, p. 104868. doi: https://doi.org/10.1016

/j.resconrec.2020.104868.

ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology

372