Litigation Risk and Corporate Social Responsibility—Evidence from a Poverty Alleviation Campaign in China

Abstract

This paper investigates the impact of an external uncertain factor, litigation risk, on corporate participation in Targeted Poverty Alleviation (TPA) activities. It proposes and explores three possible mechanisms, namely, restore legitimacy, send positive signals, and maintain corporate reputation, for corporations to manage their litigation risk via the participation in TPA. Using a sample of Chinese listed firms from 2016 to 2020, it shows that for corporations with high legitimacy pressure, high stakeholder concern, and strong reputation protection motive, litigation risk increases corporate investment in TPA. After litigation cases arise, corporations can manage their litigation risk through participation in TPA, thereby restoring legitimacy, sending positive signals, and maintaining corporate reputation. Furthermore, participation in TPA can also moderate the negative impact of litigation risk on enterprise value. The results remain significant after robustness tests on endogeneity, variable and measurement errors, and firm fixed effects. This paper is insightful for future studies relating to the economic consequences of litigation risk. Concurrently, by exploring the role of China’s legal environment in promoting the effect of corporate participation in TPA, this paper not only expands the scope of factors influencing corporate TPA inputs, but also provides policy implications for the formulation of China’s upcoming Rural Revitalization Strategy.

1. Introduction

Poverty is a global problem that not only impairs the psychological, emotional, and behavioral health of adolescents [1], increases disease prevalence and the mortality rate [2]; it also closely relates to violence, murder, and other criminal activities that contribute to social instability, acting as a “stumbling block” for economic and social development [3]. Therefore, poverty reduction is an important part of governance globally. Since the founding of the People’s Republic of China, anti-poverty has always been an important agenda for the country’s economic and social development [4]. As of 23 November 2020, all 832 poverty-stricken counties in China and 98.99 million people have been lifted out of poverty. Corporations have played a key role in poverty alleviation following the China Securities Regulatory Commission’s issuance of the “Recommendations on utilizing the role of the Capital Market to serve the National Poverty Alleviation Strategy” in September 2016, which supports and encourages listed companies to fulfill their social responsibilities [5]. Since then, corporations have begun to participate in China’s Targeted Poverty Alleviation (TPA) through various means such as donating cash or resources, employing the poor, and establishing partnership projects with impoverished areas. In 2018 alone, 1170 corporations participated in TPA, with investments totaling 55.2 billion yuan. Corporate participation in government-led poverty reduction efforts provides a unique research background for this paper, whereas previous studies have mainly focused on corporate-initiated Corporate Social Responsibility (CSR) activities.

The eradication of poverty concerns social justice and equity. Walze (1984) [6] argues that insufficient income prevents citizens from integrating into society. Rawls (1999) [7] proposes that social justice requires a system that enables the poorest members of society to achieve prosperity. Egalitarians generally believe that if one person was worse off than another for no fault of their own or choices, it would be unfair. Since poverty is often caused by war and lack of access to employment opportunities and infrastructure, it is not the fault or choice of the poor, and it is thus only fair that the poor receive help and get access to sufficient resources [8]. Therefore, the TPA projects initiated by the government have important theoretical significance for social fairness and justice.

Why are profit-oriented companies willing to participate in TPA? Most of the existing research interprets this question from the perspective of the personal traits of management. For example, Chang et al. (2021) [9] studied the poverty alleviation behavior of corporations from the perspective of management’s self-interest, altruistic, and risk management motives, and found that four types of corporations are more inclined to participate in TPA, namely: (1) State-owned enterprises with executives hired by the government; (2) corporations whose executives face a higher risk of corruption; (3) corporations under the governance of local government officials who have strong desires for promotions; (4) corporations whose executives have personal experiences of poverty. As discussed, existing research on the motivation of Chinese corporations to participate in TPA is still in its infancy, and most of them focus on the personal traits of the management, leaving a large gap in the exploration of corporation-level factors. The purpose of this paper is thus to provide evidence to narrow the gap by taking corporate litigation risks as an entry point.

With the advancement of China’s economic restructuring and the steps toward a rule-of-law society [10], the probability of Chinese corporations being subject to civil lawsuits continues to increase. Especially since the implementation of the new Accounting Standards for Business Enterprises in 2007, the number of lawsuits and the amount involved have both continued to increase: from 346 lawsuits in 2007, involving 1.36 billion yuan, to 3012 lawsuits in 2017, involving 207 million yuan, an increase of more than tenfold in the amount involved [11]. Litigation risks not only incur direct costs in legal proceedings (such as litigation fees, attorney fees, and loss compensation, etc.), but may also cause reputational damage to the litigated company [12]. The increase in litigation cases thus made Chinese corporations gradually become more aware of the importance of litigation risk management. Corporations involved in civil proceedings may be able to manage their litigation risk and protect corporate value by actively engaging in CSR to restore legitimacy, send positive signals, and maintain corporate reputation.

In the existing studies, very few literatures have studied CSR as a tool for managing litigation risks [13]. Fu (2017) [14] discussed the management of litigation risks from the perspective of charitable donations, but donations in the past were initiated by corporations, while TPA is initiated by the government. In a developing market such as China, the government oversees the allocation of important resources and has the administrative adjudication power in corporate litigation. Currently, poverty alleviation is the core political task of Chinese governments at all levels. Therefore, when corporations fulfill social responsibilities for TPA, it not only accumulates moral capital for them but also contributes to the national poverty alleviation goal. In return, the government, as the main stakeholder of the corporation, believes that in case any negative events happen to the corporation, it is the result of management mistake or pure bad luck, rather than a deliberate wrongdoing, thereby reducing the frequency and severity of government sanctions against corporations [15]. Previous research also suggests that Chinese firms can use social responsibility as an effective way to build political connections with politicians, so firms have an even stronger incentive to obey government orders and thus contribute to poverty alleviation [16]. In addition, by combining their own strengths and local quality resources, corporations can improve the infrastructure of impoverished areas, promote local industrial development, achieve “supporting aspirations and wisdom” of the needy, injecting both external and internal motivation into poverty-stricken areas, paving the way for long-term independent and sustainable development. This effectively avoids the potential phenomenon of falling back into poverty, a common pitfall under the traditional approach to poverty reduction depending on limited charitable donations. Therefore, this paper’s approach is based on the use of TPA as a tool for Chinese corporations to manage litigation risks.

This paper constructs four indicators to measure corporations’ TPA inputs, including “Total Poverty Alleviation”, “Poverty Alleviation Investment”, “Poverty Alleviation Resources”, and “Population lifted out of poverty”, based on the legitimacy theory, signal theory, and risk management theory, along with manual data collection on TPA efforts disclosed in the annual reports and social responsibility reports of A-share listed companies in China from 2016 to 2020. This paper also measures the litigation risks of corporations by tracking the existence of litigation cases, the number of litigation cases, and the amount of money involved in litigation to study the impact of litigation risks on corporations’ TPA inputs. The primary findings of this paper suggest litigation risks as a source of motivation for corporations to increase their investments in TPA. Furthermore, this paper finds that the positive relationship between litigation risks and investments in TPA is mainly found in the samples of state-owned enterprises, larger and long-established corporations, indicating that the behavior of managing litigation risks through participation in TPA is mainly found in corporations with high legitimacy pressure, high stakeholder concern, and strong reputation-protection motives. In addition, this paper further investigates the role of participation in TPA on the economic consequences of corporate litigation risks and finds that TPA helps mitigate the negative impact of litigation risks on the future financial performance of firms, thus providing evidence that TPA helps protect corporate value. Finally, this paper further analyzes industrial TPA as a dependent variable, demonstrating that corporations are more willing to participate in industrial poverty alleviation to manage litigation risks as compared to other TPA projects.

The possible theoretical contributions of this paper are as follows: (1) It enriches the study of CSR. Firstly, it enriches the connotation of CSR. Existing literature mainly studies CSR from the perspectives of environment, charitable donations, and social responsibility information disclosure, but the social responsibility of corporations participating in poverty alleviation has not been fully explored. This paper outlines the responsibilities of Chinese listed companies in the government-initiated TPA and corporations’ investments in TPA based on the level of litigation risks faced by them. Secondly, it goes beyond the typical research framework, which mainly focuses on the influence of management’s personal motivation on corporate investments in TPA, to investigate the impact of litigation risk on corporate decisions to be involved in TPA from the perspective of corporate risk. In addition, unlike existing studies, which mainly focus on the ex-ante risk management and the protection on corporate value arising from prior social responsibility inputs when negative events occur [17,18], this paper takes on an ex-post risk management approach and finds that firms can reduce the negative impact on corporate value caused by litigation risks after the emergence of it by participating in TPA. Therefore, the findings of this paper broaden the application scope of the social responsibility risk management theory proposed by Godfrey (2005) [19] and Godfrey et al. (2009) [20] from the perspective of ex-post risk management. (2) Based on existing studies [21,22], it provides empirical evidence on the economic consequences of litigation risk, as well as strategic support for firms’ practice in managing litigation risks. This paper finds that corporations with high litigation risks can protect their value by responding to policy calls and participating in TPA, and this approach is more common among state-owned enterprises, large corporations, and long-established corporations. This finding helps corporate management to effectively manage litigation risks and make decisions that are conducive to maximizing corporate value by weighing the dynamic relationship between the marginal cost of litigation risk, the marginal cost of TPA, and the marginal benefit based on their own characteristics. (3) This paper enriches research on the effectiveness of TPA policies. Poverty eradication is a key governance goal pursued by many international organizations and governments around the world. Designing effective poverty alleviation policies is also an important topic to scholars. Griesse (2001) [19] found that the Brazilian government’s poverty alleviation policy had little effect due to corruption and this paper finds that China’s increasingly improving legal environment has a positive impact in both getting corporations to be involved in TPA and towards achieving the national goal of poverty alleviation.

The rest of the paper is organized as follows: Section 2 outlines the theoretical analysis and research hypothesis; Section 3 presents the research design; Section 4 covers the analysis of the empirical results; Section 5 is the conclusion of the study.

2. Theoretical Analysis and Research Hypothesis

In recent years, with the compilation and adoption of the “Civil Code of the People’s Republic of China”, the socialist legal system with Chinese characteristics has been continuously improved, and litigation has become an important way for natural and legal persons to resolve disputes in China. In this context, the incidence of corporate litigation risks has been increasing, with a total of 3012 litigation cases involving RMB 20.7 billion for A-share listed companies in China in 2017 [23]. Frequent litigation cases with large amounts involved not only directly impact the current business performance of corporations but may also cause huge reputational damage to corporations and affect their long-term sustainable operations. Existing studies have found that firms can manage their litigation risks by ways such as establishing political ties with officials and increasing information disclosure [24], and auditors may also respond to the litigation risks of audited firms by increasing audit fees. This paper speculates that in the context of China Securities Regulatory Commission’s “supporting and encouraging listed companies to serve the national poverty alleviation strategy”, companies may also use participation in TPA as an important tool to manage their own litigation risks.

This paper will use the theory of legitimacy, signaling, and risk management to analyze the role of corporate participation in TPA.

First, restoring legitimacy. Legitimacy is the general judgment of stakeholders on whether a corporate behavior is desirable and appropriate [25]. Stakeholders’ legitimacy requirements for corporations can be divided into two categories: pragmatic legitimacy and moral legitimacy. Among them, pragmatic legitimacy is based on the stakeholders’ own interests and reflects their direct utility calculations; moral legitimacy is based on social ethics and reflects the stakeholders’ evaluation of the corporation’s activities in terms of social welfare [26]. Since litigation cases will bring direct compensation costs and indirect reputation damage to corporations, being involved in litigation cases can damage both the pragmatic legitimacy and moral legitimacy of corporations, which will cause the operation of corporation to be under the scrutiny of stakeholders. Participation in TPA thus becomes an important means for corporations to restore their legitimacy. On the one hand, with poverty alleviation now the core political task of Chinese governments at all levels, corporate participation in TPA will help local governments achieve poverty alleviation goals, which then translates to lower amounts and possibilities of them being fined after litigation cases arise, thereby reducing the value loss for investors and other stakeholders, and to some extent preserving the firm’s pragmatic legitimacy. On the other hand, poverty alleviation, as an institutional rather than instrumental social responsibility act, aims to increase social welfare, so participation in TPA helps companies maintain moral legitimacy [27]. Hence, corporate participation in TPA is beneficial in the restoration of both pragmatic and moral legitimacy damaged by litigation disputes.

Second, sending positive signals. Litigation risks that are not properly handled may become the trigger for a series of subsequent operational risks for the corporation. James and Wooten (2006) [28] find that only a small fraction of the high costs associated with litigation risks come directly from litigation risk, with the majority coming from operational or financing risks arising from improper responses to litigation risks. For example, if a firm is involved in a debt lawsuit and is awarded a high amount of damages, its operational and financial risks will rise sharply: creditors may demand early repayment for financial security reasons, suppliers may reduce the firm’s commercial credit limit or simply demand cash transactions, investors may no longer inject capital or demand higher returns, and all of these consequences may jeopardize the corporation’s survival and growth [29]. At this point, to reduce the negative chain effects that may be triggered by litigation risks, it is reasonable for companies to send positive signals to the outside world. Participation in TPA is an effective signaling tool that corporations can choose. Firstly, in the context of a national movement for poverty alleviation, the disclosure of information about a corporation’s participation in TPA is likely to attract the attention of key stakeholders such as investors and creditors. Secondly, TPA and other CSR activities can convey positive signals about the future financial performance, operational performance, and corporate value of corporations to the outside world [30], thus achieving the goal of improving key stakeholders’ confidence to maintain sustainable and healthy business operations.

Finally, protecting corporate reputation. Damage to corporate reputation is one of the significant hazards of litigation risks [31]. Previous studies have proven that corporate reputation has value and must be protected for it to contribute to a firm’s competitive advantage in the marketplace, and that threats to corporate reputation or mismanagement of reputation may create hidden risks at multiple strategic levels of the firm, such as finance, marketing, or human resources [32]. At the same time, unlike other ordinary business capital, reputational capital lacks a market that can be relied upon to insure and reduce risk exposure. Therefore, corporations have no choice but to cushion reputational damage from litigation risks through some corporate actions. Godfrey (2005) [16] found that engaging in CSR activities such as charitable donations can help firms build positive moral capital among their stakeholders, thus helping them to moderate negative stakeholder judgments and sanctions against the company when adverse situations arise. In the context of poverty alleviation, corporations can build and accumulate moral capital by fulfilling their social responsibilities through various “quasi-charitable” approaches [33], such as donating cash or resources and hiring the poor, which will help corporations protect their reputation and cushion possible damages caused by litigation cases. Based on the above analysis, this paper puts forward the hypothesis:

H1. 

Litigation risks increase the participation of corporations in TPA.

3. Research Design

3.1. Research Design and Variable Definition

This paper constructs the following model to investigate the relationship between corporate litigation risks and investments in TPA [34].

Model 1:

$$\begin{array}{cc}AntiPovert{y}_{i,t+1}& ={\beta }_0+{\beta }_{1}LITI\_Ris{k}_{i,t}+{\beta }_{2}SO{E}_{i,t}+{\beta }_{3}CAS{H}_{i,t}+{\beta }_{4}SIZ{E}_{i,t}\hfill \\ & +{\beta }_{5}RO{A}_{i,t}+{\beta }_{6}LE{V}_{i,t}+{\beta }_{7}DUA{L}_{i,t}+{\beta }_{8}B{M}_{i,t}+{\beta }_{9}AG{E}_{i,t}\\ & +{\beta }_{10}I{O}_{i,t}+{\beta }_{11}TOP{10}_{i,t}+{\beta }_{12}TO{P}_{i,t}+{\beta }_{13}DONATE\_OTHE{R}_{i,t}\\ & +{\beta }_{14}PGD{I}_{i,t}+INDUSTR{Y}_{i,t}+YEA{R}_{i,t}+{\epsilon }_{i,t}\end{array}$$

The explained variable of Model 1 is the corporation’s TPA inputs (AntiPoverty). Since corporations can participate in TPA in various ways such as direct resources donation, establishing partnership projects, and employing the poor, this paper refers to the practice of Chang et al. (2021) [9] and selects the three most basic factors as disclosed by corporations in their TPA involvements, including cash donation, resources donation, and helping registered poor people out of poverty. Based on the three factors, this paper constructs four variables to measure a corporation’s TPA inputs: (1) the natural logarithm of the total amount invested by corporation in either cash or resources in TPA (DONATE); (2) the natural logarithm of the total amount invested by corporation in cash in TPA (MONEY); (3) the natural logarithm of the total amount invested by corporation in resources in TPA (RESOURCE); (4) the natural logarithm of the number of people lifted out of poverty due to corporate participation in TPA (PEOPLE). In the main test, this paper examines the impact of litigation risks on the above four TPA input indicators, respectively; in further tests, this paper only reports the main results with DONATE as the dependent variable. To better portray the statistical information of corporation’s participation in TPA, this paper also constructs a dummy variable DONATE_DUMMY, which takes a value of 1 if corporations participate in TPA and 0 otherwise.

The experimental variable in this paper is the corporation’s litigation risks (LITI_Risk). This paper defines three measures of litigation risks by referring to the common practice of existing studies (e.g., [14]): (1) the size of litigation risks (LITI_NUM), the natural logarithm of the number of times being sued; (2) the cost of litigation risks (LITI_SIZE), the amount being sued divided by total assets; (3) the presence of litigation risks (LITI_DUMMY), which takes a value of 1 if the firm is sued and 0 otherwise. In the main test, this paper tests the impact of the above three types of litigation risks indicators on the corporation’s TPA inputs, respectively; in further tests, this paper only reports the main results with LITI_NUM as the explanatory variable.

In the model, i denotes the corporation and t denotes the year. By examining the impact of a corporation’s litigation risks in year t on its TPA inputs in year t + 1, not only can the endogeneity problem between litigation risks and TPA decisions be mitigated to some extent, but this also can provide empirical evidence on how firms manage litigation risks by participating in TPA after litigation cases occur. The focus of this paper is on the coefficient of litigation risks (β1), which measures the impact of litigation risks on corporation’s investments in TPA, and its sign and significance are the focus of this paper. In addition, this paper adds the following control variables to the model: (1) drawing on existing studies related to charitable giving, the following control variables that may affect corporate charitable giving decisions are added to the model [35]: nature of corporate ownership (SOE), cash holding level (CASH), firm size (SIZE), profitability level (ROA), financial leverage (LEV), separation of ownership and management (DUAL), growth opportunity (BM), corporation age (AGE), institutional shareholding (IO), equity concentration (TOP10), majority shareholder shareholding ratio (TOP), year dummy variable (YEAR), and industry dummy variable (INDUSTRY); (2) considering that there may be a substitution relationship between TPA investments and corporate charitable giving, that is, companies participating in TPA may reduce the amount of other types of charitable donations, this paper defines ordinary charitable donations as other charitable donations besides corporate TPA investment and adds the natural logarithm of the total number of other charitable donations (DONATE_OTHER) to the control variable; (3) considering that corporations’ TPA inputs are influenced by the local rural economic level, this paper uses the local rural Per Capita Disposable Income (PGDI) at the provincial level where the corporations are located to measure the local rural economic level, and adds it as a control variable. The specific definitions of the above-mentioned variables are shown in

Table 1. Definition of Variables.

Variable Type	Variable	Variable Definition and Calculation Method
Dependent Variables	DONATE	The natural logarithm of the total amount invested by corporation in either cash or resources in TPA
	DONATE_DUMMY	Indication of corporate participation in TPA with 1 as participate and 0 otherwise
	MONEY	The natural logarithm of the total amount invested by corporation in cash in TPA
	RESOURCE	The natural logarithm of the total amount invested by corporation in resources in TPA
	PEOPLE	The natural logarithm of the number of people lifted out of poverty due to corporate participation in TPA
Experimental Variables	LITI_NUM	The natural logarithm of the number of times being sued
	LITI_SIZE	The amount being sued divided by total assets
	LITI_DUMMY	Indication of presence of litigation risks with 1 as firm being sued and 0 otherwise
Control Variables	SOE	Indication of corporate ownership with 1 representing State-owned enterprise and 0 otherwise
	CASH	The natural logarithm of corporation’s cash and cash equivalents
	SIZE	The natural logarithm of corporation’s total assets
	ROA	Corporate net profit divided by total assets
	LEV	Corporate total liabilities divided by total assets
	DUAL	Indication of the separation of ownership and management with 1 representing combined Chairman–CEO role and 0 otherwise
	BM	Book value of net assets divided by its market value
	AGE	Number of years the company has been listed
	IO	Number of shares held by institutional investors as a percentage of total share capital
	TOP10	Number of shares held by top ten shareholders as a percentage of total share capital
	TOP	Number of shares held by the largest shareholder as a percentage of total share capital
	DONATE_OTHER	The natural logarithm of corporation’s total number of other non-TPA charitable donations
	PGDI	The natural logarithm of rural per capita disposable income at the provincial level where the corporation is located
	INDUSTRY	Industry dummy variable: take the first code of the CSRC’s 2012 “Industry Classification Guidelines for Listed Companies”, 1 indicating for the industry and 0 otherwise
	YEAR	Year dummy variable, 1 if it belongs to the year and 0 otherwise

.

3.2. Sample Selection and Data Sources

In September 2016, the China Securities Regulatory Commission’s issued the “Recommendations on utilizing the role of the Capital Market to serve the National Poverty Alleviation Strategy”, advocating the participation of corporations in TPA. In December 2016, the Shanghai and Shenzhen Stock Exchanges, respectively, issued the “Notice on Further Improving the Information Disclosure of Poverty Alleviation Work of Listed Companies” and the “Notice Good Information Disclosure of Poverty Alleviation Work of Listed Companies”, which comprehensively refined the information disclosure requirements for listed companies’ participation in social responsibility work related to poverty alleviation, while stipulating that since 2016, listed companies should disclose detailed information about their participation in TPA in their annual reports. By the end of 2020, all 832 impoverished counties in China have been lifted out of poverty. Therefore, this paper takes all A-share listed companies from 2016 to 2020 as the initial sample and selects them according to the following criteria: 1. Exclude financial industry samples, because the format of financial statements in the financial industry differs from other industries; 2. Exclude samples with missing variables; 3. Exclude ST-listed companies, finally obtaining 10,105 company annual observations for the study. The TPA data used in this paper were obtained by manually sorting out the corporations’ annual reports and social responsibility reports. All other data involved in the study were obtained from the China Stock Market & Accounting Research Database (CSMAR). In addition, to avoid the influence of outliers on the empirical results, this paper performs winsorize processing on all continuous variables at the upper and lower 1% level.

Table 2. Listed companies’ participation and investment in TPA.

Year	Total Number of Listed Companies	Percentage of Listed Companies that Made Monetary Donations	Percentage of Companies that Made Resources Donations	Percentage of Companies that Made either Monetary or Resources Donations	Total Average Investment Amount (Ten Thousand Yuan)	Total Median Investment Amount (Ten Thousand Yuan)	Average Monetary Donations (Ten Thousand Yuan)	Median Monetary Donations (Ten Thousand Yuan)	Average Resources Donations (Ten Thousand Yuan)	Median Resources Donations (Ten Thousand Yuan)	Average Number of People Lifted out of Poverty (Hundred)	Median Number of People Lifted out of Poverty (Hundred)
2016	2790	16.70	8.39	17.35	1055.94	66.27	948.88	60.00	83.12	5.01	877.66	128.00
2017	3157	21.63	10.04	22.33	1519.66	70.70	1518.77	67.10	97.56	9.00	1657.70	120.50
2018	3243	27.14	12.64	28.06	1431.11	87.52	1409.02	76.26	93.00	10.00	2280.07	145.00
2019	3447	28.05	12.45	29.13	1548.17	80.00	1511.95	73.93	94.74	10.16	2042.23	166.00
2020	3955	27.41	12.26	28.37	1317.84	87.24	1257.61	74.40	115.98	15.40	2088.72	145.00

Table reports the distribution of firms making donations to the poverty alleviation campaign by year.

reports the participation and investment of listed companies in TPA. In 2016, 16.70% of the 2790 listed companies made monetary donations and 8.39% made resources donations. The average monetary donation was 948,800 yuan and 831,200 yuan for resources donation. The median values for monetary and resources donations were much lower, at 600,000 yuan and 50,100 yuan, respectively, meaning that some companies donated significantly more than others. In total, 17.35% of listed companies participated in either monetary or material donations, with an average donation of 10,559,400 yuan (median = 662,700 yuan). Besides, each listed company on average helped to lift 87,766 people out of poverty in 2016, indicating that listed companies’ contributions have a tangible impact on poverty reduction.

From 2017 to 2020, more listed companies have participated in TPA and made more contributions to poverty eradication. By 2020, the percentage of participating corporations increased to 27.41%. The average monetary and resources donation also increased to 13,178,400 yuan in 2020 (median = 872,400 yuan). Finally, the average number of people that corporations helped lift out of poverty also increased to 208,827 in 2020 (median = 14,500 yuan). Evidently, the TPA strategy has received widespread support from listed companies.

4. Analysis of Empirical Results

4.1. Descriptive Statistical Analysis

Table 3. Descriptive Statistics.

Variables	Mean	Standard Deviation	Minimum	Median	Maximum
DONATE	0.74	1.88	0.00	0.00	13.92
DONATE_DUMMY	0.16	0.37	0.00	0.00	1.00
MONEY	0.70	1.83	0.00	0.00	13.65
RESOURCE	0.19	0.84	0.00	0.00	13.82
PEOPLE	0.33	1.36	0.00	0.00	12.04
LITI_NUM	0.22	0.63	0.00	0.00	7.93
LITI_SIZE	2.43	6.13	0.00	0.00	27.23
LITI_DUMMY	0.16	0.37	0.00	0.00	1.00
SOE	0.33	0.47	0.00	0.00	1.00
CASH	20.17	1.37	17.21	20.07	24.16
SIZE	22.26	1.32	19.97	22.08	26.32
ROA	0.05	0.04	0.00	0.04	0.21
LEV	0.40	0.20	0.06	0.39	0.86
DUAL	0.30	0.46	0.00	0.00	1.00
BM	0.35	0.16	0.06	0.33	0.79
AGE	13.49	8.09	1.00	11.00	29.00
IO	0.42	0.26	0.00	0.44	0.96
TOP10	0.60	0.15	0.25	0.61	0.94
TOP	0.34	0.15	0.09	0.32	0.75
DONATE_OTHER	7.02	6.45	0.00	9.90	17.04
PGDI	9.76	0.33	9.02	9.75	10.41

Table reports the summary statistics of the variables. The sample includes firm-year observations of listed Chinese firms in 2016–2020.

provides descriptive statistics of the main variables. It can be seen that: 1.) In the research sample of this paper, only less than 16% of corporations participate in TPA (DONATE_DUMMY), indicating that there are still many corporate forces to be tapped in TPA; at the same time, the minimum and maximum values of the total amount of cash and resources invested by corporations in TPA (DONATE) are 0 and 13.92, respectively, reflecting the disparity of corporate efforts in TPA, which provides the prerequisite for this paper in investigating factors influencing corporations’ investments in TPA. 2.) The mean value of litigation risks (LITI_DUMMY) is 16%, indicating that each corporation in our sample has an average probability of 16% of being subject to civil litigation, which is consistent with the previous results (such as Qin et al., 2020), proving the urgency of corporation litigation risk management. The distribution characteristics of the remaining variables are generally consistent with previous literature.

Further, this paper also conducts a correlation analysis of the main variables, with

Table 4. Correlation coefficients.

Index	Variables	1	2	3	4	5	6	7	8	9	10
1	DONATE	1
2	MONEY	0.99 ***	1
3	RESOURCE	0.59 ***	0.50 ***	1
4	PEOPLE	0.61 ***	0.61 ***	0.43 ***	1
5	LITI_NUM	0.05 ***	0.05 ***	0.05 ***	0.05 ***	1
6	LITI_SIZE	0.06 ***	0.06 ***	0.05 ***	0.06 ***	0.82 ***	1
7	LITI_DUMMY	0.06 ***	0.05 ***	0.05 ***	0.05 ***	0.83 ***	0.92 ***	1
8	SOE	0.18 ***	0.17 ***	0.13 ***	0.21 ***	0.07 ***	0.07 ***	0.07 ***	1
9	CASH	0.27 ***	0.26 ***	0.21 ***	0.20 ***	0.04 ***	0.04 ***	0.02 ***	0.35 ***	1
10	SIZE	0.32 ***	0.31 ***	0.22 ***	0.25 ***	0.07 ***	0.07 ***	0.05 ***	0.39 ***	0.83 ***	1
11	ROA	−0.01 *	−0.01	−0.02 ***	−0.07 ***	−0.08 ***	−0.08 ***	−0.08 ***	−0.21 ***	−0.02 ***	−0.18 ***
12	LEV	0.15 ***	0.14 ***	0.11 ***	0.13 ***	0.11 ***	0.11 ***	0.11 ***	0.29 ***	0.35 ***	0.56 ***
13	DUAL	−0.08 ***	−0.08 ***	−0.05 ***	−0.09 ***	−0.04 ***	−0.04 ***	−0.04 ***	−0.31 ***	−0.16 ***	−0.21 ***
14	BM	0.05 ***	0.05 ***	0.02 **	0.05 ***	−0.05 ***	−0.06 ***	−0.07 ***	−0.01	0.05 ***	0.03 ***
15	AGE	0.14 ***	0.14 ***	0.11 ***	0.12 ***	0.15 ***	0.15 ***	0.16 ***	0.52 ***	0.37 ***	0.47 ***
16	IO	0.19 ***	0.19 ***	0.14 ***	0.17 ***	0.05 ***	0.05 ***	0.06 ***	0.42 ***	0.40 ***	0.42 ***
17	TOP10	0.07 ***	0.07 ***	0.05 ***	0.06 ***	−0.05 ***	−0.05 ***	−0.06 ***	−0.04 ***	0.12 ***	0.07 ***
18	TOP	0.09 ***	0.09 ***	0.06 ***	0.09 ***	−0.00	−0.01	−0.01	0.23 ***	0.18 ***	0.17 ***
19	DONATE_OTHER	0.32 ***	0.32 ***	0.19 ***	0.18 ***	0.09 ***	0.11 ***	0.11 ***	−0.07 ***	0.13 ***	0.18 ***
20	PGDI	−0.07 ***	−0.06 ***	−0.07 ***	−0.10 ***	−0.11 ***	−0.11 ***	−0.12 ***	−0.14 ***	0.03 ***	−0.00
Index	Variables	11	12	13	14	15	16	17	18	19	20
11	ROA	1
12	LEV	−0.41 ***	1
13	DUAL	0.11 ***	−0.15 ***	1
14	BM	0.00	−0.41 ***	0.00	1
15	AGE	−0.26 ***	0.36 ***	−0.27 ***	−0.12 ***	1
16	IO	0.01	0.22 ***	−0.19 ***	−0.05 ***	0.29 ***	1
17	TOP10	0.25 ***	−0.06 ***	0.07 ***	0.12 ***	−0.33 ***	0.440 ***	1
18	TOP	0.10 ***	0.07 ***	−0.03 ***	0.05 ***	−0.02 ***	0.472 ***	0.61 ***	1
19	DONATE_OTHER	0.02 ***	0.08 ***	−0.02 ***	0.03 ***	0.06 ***	0.071 ***	0.03 ***	−0.01	1
20	PGDI	0.07 ***	−0.06 ***	0.11 ***	0.15 ***	−0.18 ***	−0.05 ***	0.09 ***	0.01	0.07 ***	1

*, **, *** indicate significance level of 0.1/0.05/0.01, respectively.

presenting the contents of the correlation coefficient. The results show that: (1) the four TPA investment indicators and the three litigation risk indicators are highly correlated; (2) regarding the correlation between litigation risk and TPA input, all three litigation risk indicators are significantly positively correlated with TPA input indicators, thus providing preliminary empirical evidence for our hypothesis; (3) compared to firms with low litigation risks, firms with high litigation risks have worse profitability (ROA), higher asset–liability ratio (LEV), poorer growth (BM), and also less likely to make other types of non-TPA charitable donations (DONATION_OTHER) than firms with low litigation risks. At the same time, considering the different relationships between the company size (SIZE) and the three variables of litigation risks, it was necessary to control for this variability when setting up the model.

4.2. Litigation Risks and TPA Inputs

Table 5. Litigation risk and TPA inputs.

Panel A Effect of Size of Litigation Risk on Corporation’s TPA Inputs
	(1)	(2)	(3)	(4)	(5)
VARIABLES	Expected Signs	DONATE	MONEY	RESOURCE	PEOPLE
LITI_NUM	+	0.12 ***	0.09 ***	0.07 ***	0.10 ***
		(3.86)	(3.26)	(4.21)	(4.08)
SOE	+	0.36 ***	0.34 ***	0.09 ***	0.42 ***
		(7.12)	(7.01)	(3.61)	(10.52)
CASH	+	0.05 **	0.04 *	0.06 ***	0.06
		(2.28)	(1.82)	(4.95)	(0.29)
SIZE	+	0.45 ***	0.45 ***	0.12 ***	0.27 ***
		(13.70)	(13.80)	(7.22)	(9.46)
ROA	+	0.80	0.82*	−0.25	−1.41 ***
		(1.64)	(1.71)	(−1.01)	(−3.91)
LEV	-	−0.71 ***	−0.69 ***	−0.20 ***	−0.41 ***
		(−4.89)	(−4.93)	(−2.68)	(−3.97)
DUAL	+	0.04	0.02	0.01	−0.03
		(0.98)	(0.70)	(0.68)	(−1.18)
BM	+	−0.78 ***	−0.72 ***	−0.32 ***	−0.33 ***
		(−5.04)	(−4.75)	(−3.96)	(−2.81)
AGE	+	−0.02 ***	−0.01 ***	−0.00 **	−0.01 ***
		(−4.24)	(−4.04)	(−2.15)	(−4.52)
IO	+	0.14	0.10	0.07 **	0.05
		(1.64)	(1.29)	(1.98)	(0.95)
TOP10	+	0.03	0.10	0.10	0.31 **
		(0.14)	(0.58)	(1.11)	(2.29)
TOP	+	0.15	0.13	0.01	0.06
		(0.99)	(0.85)	(0.10)	(0.50)
DONATE_OTHER	+	0.04 ***	0.04 ***	0.01 ***	0.01 ***
		(14.17)	(14.11)	(6.75)	(5.96)
PGDI	−	−0.80 ***	−0.72 ***	−0.30 ***	−0.53 ***
		(−13.55)	(−12.57)	(−9.76)	(−11.63)
Constant		−2.92 ***	−3.36 ***	−0.97 **	−0.41
		(−3.73)	(−4.36)	(−2.44)	(−0.62)
Year fixed effect		Yes	Yes	Yes	Yes
Industry fixed effect		Yes	Yes	Yes	Yes
Observations		13065	13065	13065	13065
Adj. R-square		0.26	0.25	0.13	0.18
Panel B Effect of Cost of Litigation Risk on Corporation’s TPA Inputs
	(1)	(2)	(3)	(4)	(5)
VARIABLES	Expected Signs	DONATE	MONEY	RESOURCE	PEOPLE
LITI_SIZE	+	0.013 ***	0.01 ***	0.01 ***	0.01 ***
		(4.24)	(3.74)	(3.25)	(4.20)
SOE	+	0.36 ***	0.35 ***	0.09 ***	0.42 ***
		(7.11)	(7.01)	(3.56)	(10.50)
CASH	+	0.05 **	0.04*	0.06 ***	0.01
		(2.29)	(1.83)	(4.94)	(0.30)
SIZE	+	0.45 ***	0.45 ***	0.12 ***	0.27 ***
		(13.69)	(13.79)	(7.21)	(9.46)
ROA	+	0.81*	0.82*	−0.27	−1.41 ***
		(1.66)	(1.73)	(−1.05)	(−3.90)
LEV	−	−0.71 ***	−0.69 ***	−0.20 ***	−0.42 ***
		(−4.92)	(−4.97)	(−2.64)	(−3.99)
DUAL	+	0.03	0.02	0.01	−0.03
		(0.96)	(0.69)	(0.68)	(−1.20)
BM	+	−0.79 ***	−0.72 ***	−0.32 ***	−0.33 ***
		(−5.06)	(−4.77)	(−3.97)	(−2.83)
AGE	+	−0.02 ***	−0.01 ***	−0.00 **	−0.01 ***
		(−4.27)	(−4.08)	(−2.04)	(−4.53)
IO	+	0.14 *	0.11	0.07 **	0.05
		(1.66)	(1.30)	(2.00)	(0.97)
TOP10	+	0.03	0.10	0.10	0.31 **
		(0.15)	(0.58)	(1.14)	(2.29)
TOP	+	0.15	0.13	0.09	0.06
		(1.00)	(0.86)	(0.10)	(0.52)
DONATE_OTHER	+	0.04 ***	0.04 ***	0.01 ***	0.01 ***
		(14.15)	(14.10)	(6.74)	(5.95)
PGDI	−	−0.78 ***	−0.72 ***	−0.30 ***	−0.53 ***
		(−13.56)	(−12.57)	(−9.86)	(−11.63)
Constant		−2.92 ***	−3.36 ***	−0.92 **	−0.39
		(−3.73)	(−4.37)	(−2.33)	(−0.60)
Year fixed effect		Yes	Yes	Yes	Yes
Industry fixed effect		Yes	Yes	Yes	Yes
Observations		13065	13065	13065	13065
Adj. R-square		0.26	0.25	0.13	0.18
Panel C Effect of Presence of Litigation Risk on Corporation’s TPA Inputs
	(1)	(2)	(3)	(4)	(5)
VARIABLES	Expected Signs	DONATE	MONEY	RESOURCE	PEOPLE
LITI_DUMMY	+	0.23 ***	0.20 ***	0.11 ***	0.19 ***
		(4.58)	(4.05)	(4.11)	(4.64)
SOE	+	0.36 ***	0.35 ***	0.09 ***	0.42 ***
		(7.09)	(6.99)	(3.55)	(10.48)
CASH	+	0.05 **	0.04 *	0.06 ***	0.07
		(2.32)	(1.86)	(4.98)	(0.33)
SIZE	+	0.45 ***	0.45 ***	0.12 ***	0.27 ***
		(13.75)	(13.85)	(7.26)	(9.50)
ROA	+	0.79	0.81 *	−0.27	−1.42 ***
		(1.63)	(1.70)	(−1.07)	(−3.94)
LEV	-	−0.71 ***	−0.70 ***	−0.20 ***	−0.42 ***
		(−4.94)	(−4.99)	(−2.68)	(−4.01)
DUAL	+	0.03	0.02	0.01	−0.03
		(0.95)	(0.67)	(0.66)	(−1.22)
BM	+	−0.79 ***	−0.72 ***	−0.32 ***	−0.33 ***
		(−5.06)	(−4.77)	(−3.97)	(−2.83)
AGE	+	−0.02 ***	−0.01 ***	−0.00 **	−0.01 ***
		(−4.32)	(−4.13)	(−2.13)	(−4.57)
IO	+	0.13	0.10	0.07*	0.05
		(1.59)	(1.24)	(1.92)	(0.88)
TOP10	+	0.02	0.10	0.10	0.31 **
		(0.14)	(0.57)	(1.12)	(2.29)
TOP	+	0.16	0.13	0.10	0.06
		(1.02)	(0.87)	(0.12)	(0.53)
DONATE_OTHER	+	0.04 ***	0.04 ***	0.01 ***	0.01 ***
		(14.17)	(14.11)	(6.76)	(5.97)
PGDI	-	−0.80 ***	−0.72 ***	−0.30 ***	−0.53 ***
		(−13.57)	(−12.57)	(−9.83)	(−11.63)
Constant		−2.97 ***	−3.41 ***	−0.96 **	−0.44
		(−3.80)	(−4.43)	(−2.43)	(−0.68)
Year fixed effect		Yes	Yes	Yes	Yes
Industry fixed effect		Yes	Yes	Yes	Yes
Observations		13065	13065	13065	13065
Adj. R-square		0.26	0.25	0.13	0.18

This table reports the results from regressions estimating model 1. The sample includes firm-year observations for Chinese listed firms in 2016–2020. Variables are defined in Table 1. Standard errors are adjusted for the clustering effect at the firm level. *, **, *** indicate significance level of 0.1/0.05/0.01, respectively; number in parentheses is the t value.

reports the multivariate regression estimation results of litigation risks and TPA inputs. Panel A, Panel B, and Panel C, respectively, show the effects of the size of a firm’s litigation risks, the cost of litigation risks, and the presence of litigation risks on its TPA decisions. The coefficients of the corporate TPA input indicators are all positively significant, indicating that a corporation’s litigation risks in the current year increases its TPA inputs in the following year, which verifies the hypothesis of this paper. In addition, this paper finds that the impact of litigation risks on corporate charitable giving decisions is economically meaningful: taking columns (1) to (4) of Panel A as an example, the coefficients of LITI_NUM indicate that for every 1% increase in the number of litigation cases of a corporation, its TPA resources and cash input, cash input, resources input, and the number of people lifted out of poverty in the following year will increase by 0.12%, 0.09%, 0.07%, and 0.10%, respectively. Furthermore, the effects of the control variables in the regression results on firms’ TPA inputs are consistent with previous studies: state-owned enterprises are more likely to participate in TPA; large, profitable, and long-established corporations have higher TPA inputs; the number of poverty-stricken counties in the province where the company is located is also positively related to its investments in TPA.

4.3. Economic Consequences of Corporation’s Use of Participation in Poverty Alleviation to Manage Litigation Risks

In this section, this paper examines the economic consequences of corporations’ use of participation in poverty alleviation to manage litigation risk. Litigation risk is detrimental to corporate value because it not only triggers direct costs in legal proceedings (e.g., litigation costs, attorney fees, and damages for loss of litigation), but may also inflict reputational damage on the litigated firm. If corporations were to manage their litigation risk by participating in TPA, it should serve as an insurance for the firm’s value. To test this speculation, this paper constructs the dummy variable DONATE_DUMMY to measure whether the corporation participates in TPA, which takes a value of 1 if the corporation participates in TPA and 0 otherwise. Meanwhile, the dummy variable LITI_DUMMY is used to reflect whether the corporation has litigation risk, and the following model is constructed [36].

Model 2:

$$\begin{array}{cc}TOBIN{Q}_{i,t+1}=& {\beta }_0+{\beta }_{1}LITI\_DUMM{Y}_{i,t}\times DONATE\_DUMM{Y}_{i,t}\hfill \\ & +{\beta }_{2}DONATE\_DUMM{Y}_{i,t}+{\beta }_{3}LITI\_DUMM{Y}_{i,t}+{\beta }_{4}SO{E}_{i,t}\\ & +{\beta }_{5}SIZ{E}_{i,t}+{\beta }_{6}RO{A}_{i,t}+{\beta }_{7}LE{V}_{i,t}+{\beta }_{8}B{M}_{i,t}+{\beta }_{9}I{O}_{i,t}\\ & +{\beta }_{10}TO{P}_{i,t}+{\beta }_{11}POVERT{Y}_{i,t}+INDUSTR{Y}_{i,t}+YEA{R}_{i,t}+{\epsilon }_{i,t}\end{array}$$

The dependent variable $TOBIN{Q}_{i,t+1}$ = (market value of equity + market value of net debt)/(total assets at year-end − net intangible assets − net goodwill) and the coefficient of the cross product term β1 measures the economic consequences of litigation risk in precision poverty management.

Table 6. Litigation risks, TPA inputs and future financial performance.

	(1)	(2)
VARIABLES	TOBINQ	TOBINQ
LITI_DUMMY × DONATE	0.045 ***
	(4.20)
LITI_DUMMY × DONATE_DUMMY		0.20 ***
		(3.35)
DONATE	−0.00 **
	(−2.39)
DONATE_DUMMY		−0.07 **
		(−2.05)
LITI_DUMMY	−0.063*	−0.06 *
	(−1.77)	(−1.70)
SOE	−0.01	−0.09 **
	(−1.40)	(−2.05)
ROA	−0.09 **	7.97 ***
	(−2.12)	(11.80)
CASH	7.97 ***	−0.00
	(11.81)	(−0.07)
SIZE	−0.00	−0.19 ***
	(−0.07)	(−6.47)
LEV	−0.19 ***	−3.79 ***
	(−6.38)	(−19.31)
DUAL	−3.79 ***	0.20 ***
	(−19.14)	(4.94)
BM	0.20 ***	−6.29 ***
	(4.96)	(−32.34)
AGE	−6.29 ***	−0.00
	(−32.02)	(−1.35)
IO	−0.00	0.16 *
	(−1.31)	(1.75)
TOP	0.16 *	−1.27 ***
	(1.74)	(−8.97)
TOP10	−1.27 ***	2.32 ***
	(−8.98)	(16.11)
PGDI	2.32 ***	−0.06
	(16.11)	(−1.32)
Constant	10.05 ***	10.05 ***
	(14.96)	(15.11)
Year fixed effect	Yes	Yes
Industry fixed effect	Yes	Yes
Observations	11772	11772
Adj. R-square	0.49	0.49

This table reports the results from regressions estimating model 2. The sample includes firm-year observations for Chinese listed firms in 2016–2020. Variables are defined in Table 1. Standard errors are adjusted for the clustering effect at the firm level. *, **, *** indicate significance level of 0.1/0.05/0.01, respectively; number in parentheses is the t value.

reports the regression results of Model 2. The results are consistent with previous research [37], where the LITI_DUMMY coefficient is negative and significant, indicating that current litigation risk significantly damages the future financial performance of the corporation. More importantly, the coefficient of the cross-product term is significantly positive, indicating that participating in TPA weakens the negative correlation between litigation risks and financial performance, thus proving the protective effect of participating in TPA on corporate value.

4.4. Robustness Test

To test the robustness of the above results, this section conducts robustness tests from two aspects: endogeneity and measurement bias [38]. Firstly, there may be endogeneity between corporate litigation risks and TPA input decisions. Specifically, there may be certain omitted variables that are difficult to observe and measure (e.g., corporate culture) which affect both the probability of encountering civil litigation and the incentive to participate in TPA, thus skewing the relationship between litigation risks and poverty alleviation inputs. To rule out this possibility, this paper uses the instrumental variable approach for robustness testing. LITI_NUM_MEAN is an instrumental variable of the target firm’s size of litigation risks (LITI_NUM), and it is calculated as follows:

Formula (1):

$$LITI\_NUM\_MEAN=\left| LITI\_NUM\_MEA{N}_{i,t}-{\displaystyle \sum }_{j\ne i}^{N}LITI\_NUM\_MEA{N}_{j,t}/\left(N-1\right) \right|$$

(1)

In the Formula (1), t is the year, i is the target firm, j is the firm in i’s industry, and N is the total number of firms in i’s industry. In this paper, the test is conducted by the two-stage least squares (2SLS) method, with columns (1) and (2) of

Table 7. Robustness Test.

	Endogeneity: Instrumental Variable Approach		Measurement Bias: Dummy Variable Construction	GMM Estimate	Truncation Process
	(1)	(2)	(3)	(4)	(5)
VARIABLES	PREDICTED LITI_NUM	DONATE	DONATE_DUMMY	DONATE	DONATE
LITI_NUM_MEAN	0.72 ***
	(27.29)
PREDICTED LITI_NUM		2.84 ***
		(19.43)
L.DONATE				0.805 ***
				(34.50)
LITI_NUM			0.08 **	0.050 *	0.11 ***
			(2.40)	(1.69)	(3.41)
SOE	−0.03 *	0.49 ***	0.86 ***	0.22 ***	0.35 ***
	(−1.90)	(7.81)	(12.31)	(4.35)	(6.70)
CASH	−0.00	−0.00	0.02	0.01	0.04 *
	(−0.19)	(−0.11)	(0.59)	(0.18)	(1.72)
SIZE	−0.01	0.51 ***	0.40 ***	0.25 ***	0.361 ***
	(−1.05)	(12.67)	(8.45)	(6.82)	(10.42)
ROA	−0.50 ***	3.24 ***	1.52 **	0.81	0.59
	(−2.98)	(4.86)	(2.02)	(1.57)	(1.09)
LEV	0.22 ***	−1.43 ***	−0.95 ***	−0.29 **	−0.56 ***
	(4.73)	(−7.36)	(−3.97)	(−2.23)	(−3.56)
DUAL	0.01	0.01	0.07	−0.00	0.03
	(0.88)	(0.17)	(1.12)	(−0.01)	(0.71)
BM	0.03	−0.46 **	−0.53 **	−0.33 **	−0.65 ***
	(0.71)	(−2.49)	(−2.11)	(−2.37)	(−3.97)
AGE	0.01 ***	−0.03 ***	−0.00	−0.00	−0.07 *
	(7.11)	(−7.80)	(−0.01)	(−0.47)	(−1.66)
IO	0.01	0.17	0.47 ***	0.01	0.12
	(0.24)	(1.32)	(3.08)	(0.17)	(1.44)
TOP10	0.05	0.09	−0.12	0.19	0.14
	(0.84)	(0.40)	(−0.44)	(1.09)	(0.77)
TOP	−0.04	0.20	0.13	0.35 **	−0.11
	(−0.72)	(0.97)	(0.54)	(2.05)	(−0.70)
DONATE_OTHER	0.00	0.05 ***	0.09 ***	0.03 ***	0.04 ***
	(0.16)	(11.54)	(16.62)	(7.34)	(14.18)
PGDI	−0.50 ***	−0.34 ***	−1.72 ***	−0.65 ***	−0.79 ***
	(−7.44)	(−7.65)	(16.62)	(−12.09)	(−13.17)
Constant	1.06 ***	−3.52 ***	4.89 ***	0.85	−0.93
	(5.01)	(−4.041)	(4.87)	(1.25)	(−1.11)
Year fixed effect	Yes	Yes	Yes	Yes	Yes
Industry fixed effect	Yes	Yes	Yes	No	Yes
Observations	13065	13065	13065	12783	11558
Adj. R-square	0.08	0.092			0.23
Pseudo R-square			0.186
AR (1)				0.00
AR (2)				0.12
Hansen				0.85

This table column 1–2 reports the results from regressions estimating model 3, Columns 1 and 2 of Table 7 report the regression results estimating model 3, Column 3 reports the results of model 1 by replacing the dependent variable with DONATE_DUMMY. Column 4 reports the results of the GMM estimation method, and column 5 reports the results of model 1 by truncating all continuous variables at the upper and lower 1% levels. The sample includes Chinese listed firms from 2016–2020. Variables are defined in Table 1. Standard errors are adjusted for the clustering effect at the firm level. *, **, *** indicate significance level of 0.1/0.05/0.01, respectively; number in parentheses is the t value.

reporting the results of the relevant tests. Evidently, the average number of litigation cases in the industry has a significant positive impact on the number of litigation cases of the target firms (LITI_NUM_MEAN), thus proving the effectiveness of the instrumental variable: the predicted level of corporate litigation risks has a significant positive impact on the company’s investments in TPA (PREDICTED LITI_NUM), proving that the results of this paper pass the instrumental variables test.

In addition, considering the possibility that some corporations may misrepresent the amount of investments in TPA, which may cause measurement bias in the dependent variable of this paper, in column (3) of Table 7, this paper uses the dummy variable DONATE_DUMMY to measure whether corporations participate in TPA and test whether it is affected by corporation litigation risks through logit regression. Specifically, DONATE_DUMMY takes the value of 1 if the firm participates in TPA and 0 otherwise. In column (3), the coefficient of LITI_NUM remains positively significant, demonstrating that the results of this paper are still robust after taking the measurement error of TPA into consideration. Arellano and Bover (1995) [39] and Blundell and Bond (1998) [40] proposed the Generalized Method of Moments Estimation, which was further enhanced by using the lags of the differential variables as instrumental variables for the horizontal values, increasing the available instrumental variables, i.e., using both the horizontal differential equation and the difference equation in the estimation process. The enterprise-level TPA inputs are somewhat persistent, that is, serially correlated. To address this issue [41], this paper further used systematic GMM regressions to test the robustness of the previous findings. The test statistics presented in column 4 of Table 7 show that the instrumental variables are valid and satisfy the conditions for the use of systematic GMM. The regression results in column 4 of Table 7 show that the coefficient of LITI_NUM in regression is significantly positive at 10% level. This indicates that after considering the characteristic of serial correlation of TPA inputs (controlling for L.DONATE and its resulting endogeneity), the incentive effect of litigation risk on corporate TPA inputs still exists, hence proving the previous conclusion robust. Finally, this paper truncated all continuous variables at the upper and lower 1% levels and the results remained stable.

5. Further Analyses

5.1. Different Type of TPA

The China Securities Regulatory Commission requires listed companies to classify their TPA projects into nine categories: industrial development, employment transfer, relocation, education, health, ecological protection, basic security, social poverty alleviation, and other projects. This paper further analyzes the industrial poverty alleviation among the TPA projects based on the following reasons. Firstly, the business experience of listed companies gives them a greater advantage in industrial poverty alleviation projects as TPA participation can be referenced as new growth points for their own development. Secondly, compared with other poverty alleviation methods, corporations engaged in industrial poverty alleviation combine their participation in TPA with the company’s own business development, and participate in TPA by virtue of their own investments, brand, technology, management, and other capabilities, which has a greater impact on the company’s business development.

From the above analysis, industrial poverty alleviation is more of an investment, which not only responds to the call of the state but also expands corporation’s own business and increases its corporate value. In this paper, the natural logarithm (AntiPoverty_Ind) of the amount invested by corporations in industrial poverty alleviation is used to measure industrial poverty alleviation, and the natural logarithm (AntiPoverty_Oth) of the total amount of poverty alleviation after deducting the amount of industrial poverty alleviation is used to measure the participation of corporation in other TPA projects [42].

Table 8. Litigation risks and industrial poverty alleviation inputs.

	(1)	(2)
VARIABLES	AntiPoverty_Ind	AntiPoverty_Oth
LITI_ DUMMY	0.13 ***	0.09
	(3.39)	(1.27)
SOE	0.29 ***	0.10 *
	(7.74)	(1.72)
CASH	0.03 **	0.02
	(1.99)	(0.65)
SIZE	0.29 ***	0.16 ***
	(11.22)	(4.90)
ROA	−0.85 **	1.91 ***
	(−2.41)	(3.65)
LEV	−0.38 ***	−0.35 **
	(−3.51)	(−2.40)
DUAL	0.00	0.04
	(0.16)	(1.06)
BM	−0.39 ***	−0.38 ***
	(−3.39)	(−2.65)
AGE	−0.01 ***	−0.00
	(−4.81)	(−0.06)
IO	−0.04	0.17 *
	(−0.63)	(1.81)
TOP10	0.29 **	−0.21
	(2.15)	(−1.17)
TOP	−0.01	0.17
	(−0.07)	(0.93)
DONATE_OTHER	0.01 ***	0.03 ***
	(4.35)	(10.99)
PGDI	−0.34 ***	−0.49 ***
	(−7.77)	(−7.44)
Constant	−3.34 ***	0.81
	(−5.31)	(1.05)
Year fixed effect	Yes	Yes
Industry fixed effect	Yes	Yes
Observations	13065	13065
Adj. R-square	0.16	0.11

*, **, *** indicate significance level of 0.1/0.05/0.01, respectively; number in parentheses is the t value.

reports the multivariate regression results of litigation risks and industrial poverty alleviation inputs. The coefficient of LITI_DUMMY is significantly positive in the AntiPoverty_Ind column while not significant in the AntiPoverty_Oth column, further validating the paper’s view that corporation are more willing to participate in industrial poverty alleviation for litigation risk management compared to other poverty alleviation projects.

5.2. Conditional Analysis

In the previous hypothesis formulation section, this paper sorts out three possible mechanisms by which corporations use participation in poverty alleviation to manage their litigation risks: restoring legitimacy, sending positive signals, and maintaining corporate reputation. To further test these three influential mechanisms, this paper refers to existing research result [43,44] and selects the nature of corporation ownership (SOE), firm size (SIZE), and firm age (AGE) as the respective proxy measures of legitimacy pressure, stakeholder concern, and reputation protection motives. Firstly, natural political affiliation gives SOEs an advantage in terms of legitimacy, but also puts more pressure on firms to reestablish legitimacy when litigation risks occur, so SOEs are more likely to restore their legitimacy damaged by litigation disputes by participating in TPA than non-SOEs. Secondly, large corporations are more likely to receive attention from stakeholders such as investors, regulators, and information intermediaries [45]. Thus, improper handling of litigation risks is more likely to trigger operational and financing risks for large firms. Therefore, large corporations are more motivated to adopt means (i.e., participate in TPA) to send positive signals to the outside world after litigation risks occur. Finally, longer established firms tend to have a better reputation among investors as compared to newly established firms, and reputational capital gives longer established firms a source of competitive advantage at multiple strategic levels, such as finance, marketing, or human resources. Hence, such firms have a stronger incentive to protect their reputation and when faced with litigation risk, they are more motivated to protect their reputation by participating in TPA.

This paper examines the above three types of mechanisms by constructing cross-product terms of the above three types of variables and the presence of litigation risk (LITI_ DUMMY) and adding the cross-product terms to model (1).

Table 9. Effect of litigation risk on mechanisms of TPA inputs.

	(1)	(2)	(3)
VARIABLES	DONATE	DONATE	DONATE
LITI_ DUMMY × SOE 1	0.49 ***
	(4.97)
LITI_ DUMMY × SIZE		0.34 ***
		(8.09)
LITI_ DUMMY × AGE			0.02 ***
			(3.81)
LITI_ DUMMY	0.04	−7.37 ***	−0.11
	(0.72)	(−8.02)	(−1.13)
SOE	0.25 ***	0.35 ***	0.36 ***
	(4.71)	(6.99)	(7.06)
CASH	0.06 **	0.06 **	0.06 **
	(2.34)	(2.38)	(2.34)
SIZE	0.46 ***	0.38 ***	0.46 ***
	(13.88)	(11.26)	(13.95)
ROA	0.764	0.87*	0.72
	(1.56)	(1.78)	(1.47)
LEV	−0.72 ***	−0.70 ***	−0.74 **
	(−4.98)	(−4.86)	(−5.14)
DUAL	0.04	0.04	0.03
	(0.97)	(1.00)	(0.91)
BM	−0.78 ***	−0.74 ***	−0.79 ***
	(−5.04)	(−4.81)	(−5.08)
AGE	−0.02 ***	−0.01 ***	−0.02 ***
	(−4.24)	(−3.84)	(−5.47)
IO	0.14 *	0.15 *	0.13
	(1.66)	(1.85)	(1.60)
TOP10	0.02	−0.04	0.01
	(0.10)	(−0.24)	(0.06)
TOP	0.17	0.18	0.18
	(1.10)	(1.17)	(1.18)
DONATE_OTHER	0.04 ***	0.04 ***	0.04 ***
	(13.99)	(14.00)	(14.07)
PGDI	−0.80 ***	−0.79 ***	−0.80 ***
	(−13.66)	(−13.52)	(−13.59)
Constant	−2.99 ***	−1.42 *	−3.05 ***
	(−3.82)	(−1.77)	(−3.90)
Year fixed effect	Yes	Yes	Yes
Industry fixed effect	Yes	Yes	Yes
Observations	13065	13065	13065
Adj. R-square	0.26	0.27	0.26
Wald test
LITI_ DUMMY + LITI_ DUMMY × SOE = 0	0.00 ***
LITI_ DUMMY + LITI_ DUMMY × SIZE = 0		0.01 **
LITI_ DUMMY + LITI_ DUMMY × AGE = 0			0.06 *

This table reports the results from regressions estimating model 1 with by LITI_ DUMMY × SOE, LITI_ DUMMY × SIZE and LITI_ DUMMY × AGE. The sample includes firm-year observations for Chinese listed firms in 2016–2020. Variables are defined in table. Standard errors are adjusted for the clustering effect at the firm level. ¹ This paper grouped the variables SOE, SIZE, and AGE. Group-level testing revealed that results were more significant in the SOE group, as well as among larger and long-established corporations. *, **, *** indicate significance level of 0.1/0.05/0.01, respectively; number in parentheses is the t value.

reports the regression results of the above method. Evidently, the coefficients of the cross-product terms are significantly positive in columns (1) to (3), indicating that the impact of litigation risk on corporate TPA inputs is mainly found in state-owned enterprises, larger firms, and longer established firms. This finding not only provides more cross-sectional evidence on the relationship between litigation risk and TPA, but also proves the three mechanisms for companies to use their participation in TPA to manage their litigation risks: restoring legitimacy, sending positive signals, and maintaining corporate reputation.

6. Conclusions

Poverty eradication is a key goal pursued by governments and international organizations around the world, and an important topic that has been explored by scholars for a long time. Due to the externalities of poverty, most previous studies have examined the effects of macro policies on poverty eradication from the perspective of government governance. This paper, however, takes advantage of the unique context of government encouragement of corporate participation in poverty alleviation in the context of TPA in China to investigate which firms are more likely to respond to the government’s call and take on the social responsibility of TPA. The results of this paper suggest that litigation risk is an important motivation to promote corporations’ participation in TPA, and this result remains robust to multiple alternative measures and instrumental variables tests. At the same time, this paper explores the mechanisms by which litigation risks increases corporations’ participation in TPA and finds that corporations with high litigation risks use participation in TPA as an important risk management tool for them to restore their legitimacy, send positive signals, and protect their corporate reputation. Finally, the results of this paper show that participation in TPA helps moderate the negative impact of litigation risks on firms’ future financial performance, demonstrating the positive value of TPA in terms of reducing the value loss caused by litigation risks.

The theoretical significance of this paper is as follows: (1) This paper finds that corporate TPA can play a role in restoring legitimacy, sending positive signals, protecting corporate reputation and ultimately, corporate value after litigation cases arise. These findings expand and complement existing theories of ex-ante social responsibility risk management. (2) The results of this paper enrich the literature on the economic consequences of corporate litigation risk and provide effective coping strategies for firms to deal with litigation risks. (3) The findings of this paper illustrate the role of the optimization of the legal system environment in promoting the enthusiasm of corporations for TPA and achieving the overall national poverty alleviation goal.

The practical significance of this paper is as follows: (1) The findings of this paper provide a reference for society and regulators to better understand the motives of Chinese listed companies’ participation in TPA, which may not be altruism alone, but also to mitigate the possible negative consequences of litigation risks. For this reason, society and regulators should be more cautious about the true motives behind corporate involvement in TPA. (2) A favorable external environment (natural and social) is the “third kind of capital” of enterprises. Hence, the external governance mechanisms involving stakeholders in the accumulation and maintenance of such capital needs to be given due attention by enterprises. Based on the conclusions of this paper, it is recommended that companies should be fully aware of the legal litigation environment in which they operate in, so that they can manage and control the litigation risks they face. (3) 2020 is the final year of China’s poverty eradication strategy, and after eight years, nearly 100 million rural poor people in China have been completely lifted out of poverty under the current standard, and all 832 poverty-stricken counties have been removed from the list. However, there are still new tasks and new goals of consolidating and expanding the results of poverty eradication and promoting the rural revitalization strategy. These new tasks and new goals not only require corporations that have participated in TPA to stay on, but also need to attract more corporations to actively participate and support TPA initiatives. Therefore, the research findings of this paper have very important reference values and policy inspirations for the implementation of the rural revitalization strategy that is about to start. At the same time, the findings of this paper also have certain implications for the governance goals of poverty eradication in other countries and international organizations around the world.