Abstract
With increased corporate liquidity and debt repayment pressure, CSR's “insurance” role has received more attention. We conducted a comprehensive empirical analysis of 4988 listed companies in the Chinese context during 2011–2020. Our research has three findings: First, the initial increase in CSR will lead to a rise in default risk. However, once the CSR level exceeds a specific threshold, the default risk decreases as the CSR rises. We tested the robustness of the results by replacing the explanatory and the explained variables and taking into account the lag time effect, which proved the reliability of our research conclusions. Second, the mediation analysis shows financing constraints play an important mediating role in this inverted U-shaped relationship. On the left side of the U-shape, CSR performance intensifies financing constraints, while on the right side, increasing CSR reduces financing constraints. Finally, we confirm heterogeneity in the impact of CSR on the default risk of different enterprises' ownership and size. Our study complements the current literature on the effects of CSR on default risk. We are making policymakers and stakeholders aware of the importance of mandatory CSR disclosure.
Keywords: Corporate social responsibility, Default risk, Financing constraints, Inverted U-Shape
1. Introduction
Corporate social responsibility (CSR) is a strategic issue of long-term concern for companies and an essential tool to attract potential investors and maintain stakeholder relationships [1,2]. Research has revealed that CSR enhances companies' foresight, allowing them to predict the future more effectively and promptly respond to external changes to avoid adverse outcomes [3]. In recent years, CSR's “insurance” role has been increasingly emphasized with increased corporate liquidity and debt servicing pressures. The dominant view is that CSR performance contributes to default risk reduction. Lee and Faff [4] reveal that companies demonstrating social responsibility exhibit a reduced idiosyncratic risk attributed to a thriving market portfolio. Goss and Roberts [5] show that a significant decline in firm risk can be attributed to improved CSR performance. Mishra and Modi [6]reveal that a substantial reduction in firm risk can be attributed to positive CSR performance. Kim et al. [7]emphasize that organizations demonstrating a robust social responsibility generally exhibit higher levels of information disclosure. This prevents the concealment of negative news, reducing the likelihood of sudden stock price drops.
On the contrary, some scholars argue that CSR exacerbates default risk due to the existence of managerial trenches [8]. Farah et al. [9]reveal that the influence of CSR on systemic risk demonstrates an increasing trend before descending, and it is only when CSR performance reaches a specific level that it can effectively lower systemic risk. These conflicting findings prevent us from conducting a solitary assessment of the impact of CSR on default risk, regardless of whether the correlations are linear, positive, or negative. This paper aims to deepen this controversial research and address the question: Does CSR affect default risk? How does the transmission mechanism work, and does the impact of CSR on default risk vary among different corporations?
We propose that the impact of CSR on default risk demonstrates an inverted U-shaped curve, marked by a preliminary escalation in risk, followed by a reduction. Initially, stakeholders tend to overlook the constructive signals transmitted by corporations through CSR initiatives [9]; an escalation in CSR leads to an increase in default risk. However, when CSR reaches a threshold, the firm's default risk mitigates as CSR increases. This is because more costly CSR activities make it difficult for competitors to imitate and widen the gap between firms and competitors. Also, financial constraints serve as a mediating factor in this inverted U-shaped association.
First, inexpensive CSR expenditures can undermine a company's profitability and reputation. Such expenditures consume resources, exacerbate financial constraints, and ultimately increase the default risk. After reaching a certain level, CSR brings a higher reputation and profitability to the firm, obtains abundant cash flow, and decreases the financing constraint, decreasing the default risk.
We use the CSMAR database and CNRDS database to verify our conjectures. The sample comprises Chinese firms from 2011 to 2020, totalling 26,980 firm-year observations. We construct an aggregate social responsibility score by combining firms' performance on philanthropy, corporate governance, diversity, employee relations, environment, and products. We use Merton's DD model to measure firms' default risk. We conduct panel regression analyses, considering time, industry fixed effects, and other factors influencing default risk.
Initially, we present evidence suggesting that the relationship between CSR and default risk exhibits an inverted U-shaped pattern. The primary term coefficient, associated with CSR and regarding the default distance, demonstrates a notably negative trend at the 10 % level. In contrast, the secondary term coefficient displays a significantly positive pattern at the 1 % level. It shows that the influence of CSR on the default distance is U-shaped, and the effect on default risk is inverted U-shaped. Second, we demonstrate the mediating effect of financial constraints within this inverse U-shaped correlation. The fulfilment of CSR in the left portion of the inverted U-shaped augments financing constraints, whereas increased CSR mitigates firms' financial constraints in the right portion. This finding implies that financial constraints act as a mediating mechanism that enables CSR to influence default risk. Finally, our heterogeneous analysis reveals that the coefficients of CSR's primary and secondary components are negligible in NSOEs; however, the positive trend of the secondary component's coefficient in CSR is noticeable within state-owned enterprises. The research indicates that CSR is more prominent in mitigating default risk across SOEs. Additionally, the coefficients of CSR's quadratic expressions show a statistically significant positive slope for companies of varying sizes, but the effect is more significant for SMEs.
To ensure the credibility of our research, we perform a series of robust checks. First, we substitute the measurement of the independent variable CSR. In light of an earlier study [9], we adopt ESG rating as a proxy for CSR. The inverse U-shaped correlation between ESG and default risk indicates a significant deviation at the 5 % significance margin. Second, we consider the lagged effect of CSR affecting default risk. The statistical significance of the inverse U-shaped correlation between CSR and default risk is established at the 1 % threshold when considering one and two-year lags. Finally, we use DD_Bhsh as a proxy variable for DD. The study reveals an inverse U-shaped relationship between CSR and default risk at the 1 % level of statistical significance.
Our results reinforce the study of the relationship between CSR and default risk, making two crucial contributions. First, we expand on the influencing factors of default risk. Research on the factors influencing default risk mainly focuses on stock liquidity [10], return rate, and market volatility [11]. Our study confirmed that CSR is also an essential factor affecting default risk and found that the impact of CSR on default risk is not a simple linear relationship. With different levels of CSR, the impact direction of CSR on default risk is also different. Second, we expand the mediating mechanism by which CSR affects default risk. In the past few studies on the relationship between CSR and default risk, scholars paid more attention to heterogeneity [1], persistence [12], and region [13], and there needed to be more discussion on the impact mechanism. Our research proves that lower financing constraints are essential for CSR to reduce default risk. In the whole U-shaped process of CSR's influence on default risk, financing constraints play an important mediating role. On the left side of the U-shape, an increase in CSR leads to an increase in financing constraints, leading to a rise in default risk. On the right side of the U-shape, an increase in CSR leads to a reduction in financing constraints, leading to a decline in default risk. Overall, our study complements the current literature on the impact of CSR on default risk. We are making policymakers and stakeholders aware of the importance of mandatory CSR disclosure.
The remaining portion of this article is structured around the following framework: The second section provides a concise overview of the literature pertinent to CSR, default risk, and financing constraints, formulating the two hypotheses underpinning the paper. The third section explains the methodology and data collection processes in detail. Part IV presents the empirical findings. Part V summarizes the conclusions of the paper and gives specific managerial recommendations.
2. Literature review and hypotheses formation
2.1. CSR and default risk
The existing literature distinguishes two ways CSR affects default risk: the risk mitigation view and the over-investment view. The initial perspective is the risk mitigation view, grounded in stakeholder theory [14], which argues that CSR can act as a risk mitigator. The second viewpoint is the over-investment theory, rooted in shareholder and agency theory [15]. This theory suggests that CSR could escalate a company's default risk.
Freeman's [14] stakeholder theory suggests that firms maintain strong ties with various stakeholders by fulfilling their social responsibilities, obtaining a stable cash flow, and reducing the default risk. As a reward for companies that take the initiative to fulfil their social responsibility, stakeholders will support the company through improved reputation, increased loyalty, and lower financing costs. In turn, the goodwill created through the fulfilment of CSR provides an insurance-like effect on the future cash flow of the enterprise, making the enterprise more resilient in the event of risks [16,17].
Most academics have embraced the risk mitigation view regarding the CSR and default risk relationship. They assert that the two exhibit a negative correlation. Spicer [18] kick-started the investigation into the impact of CSR on systemic risk, providing preliminary empirical evidence to support the notion that fulfilling social responsibility minimizes risk. Herremans et al. [19] [][19][]found that companies with a robust reputation for social responsibility in the U.S. manufacturing sector tend to have higher stock returns and lower investment risks, benefiting investors. Albuquerque et al. [20]maintain that the buffering effect of CSR on systemic risk is demonstrated through an increase in corporate value.
The over-investment view [21] challenges Freeman's stakeholder theory, proposing that excessive resource allocation to CSR could be wasteful, leading to increased fluctuations in cash flow. This perspective aligns with shareholder and agency theory, implying that CSR initiatives might divert crucial resources to sectors unrelated to operational activities and that managerial interests might take precedence over shareholders.
A few studies support the view that CSR is an over-investment for the default risk. Godfrey [22] claimed that when CSR investment crosses the optimal point, investment in society and the environment creates an additional burden for the firm without adding more to the firm's bottom line, which increases the risk of bankruptcy. Cespa and Cestone [23] also argued that managers engage in social activities to gain stakeholders' support. Engaging in such social endeavours enhances the chances of the organization becoming susceptible to risks and substantially harming its operational efficiency. In summary, these two conflicting perspectives create a research vacuum for academics to further investigate the effects of CSR on default risk.
This paper argues that when CSR scores are low, stakeholders ignore the positive signals that firms send by fulfilling their social responsibilities. As a result, if CSR fails to reach a certain level, it creates a higher default risk. This supports the overinvestment view that CSR positively affects default risk [15]. Conversely, when CSR scores are high, firms develop more substantial competitiveness that outweighs the cost of fulfilling CSR, ultimately reducing default risk. This research supports the notion that CSR can have a detrimental effect on reducing default risk [14]. Integrating the over-investment and risk mitigation views results in an inverted U-shaped curve. The theoretical analysis above leads us to the first hypothesis of this paper.
Hypothesis 1
CSR can have an inverted U-shaped effect on default risk.
2.2. Financing constraints
It has been shown [2] that CSR can reduce the probability of default by alleviating financing constraints. A high level of financing constraints will inhibit the support of external channels for corporate funding, increasing the likelihood of corporate default. In contrast, a low level of financing constraints will promote the support of external channels for corporate funding, reducing the probability of corporate default [24]. Before CSR reaches the threshold, fulfilling social responsibility will increase the enterprise's financing constraints because poor social responsibility performance will not enhance the company's reputation and profitability. However, this will also undermine the company's profitability and reputation due to resource diversion, leading to these enterprises only seeking to borrow money from other channels with higher financing costs. This situation exacerbates the firm's financial constraints and thus increases the firm's default risk [24]. This will lead to increased financing constraints, which in turn will increase the risk of corporate default. However, when the social responsibility score reaches a certain threshold, it will bring the enterprise a higher reputation and profitability, leading to abundant cash flow and lower financial distress and financing constraints. The above arguments show that the advantage of CSR in alleviating financing constraints works only when the social responsibility score is high and that poor social responsibility performance can exacerbate financing constraints. The above discussion prompts us to propose the second hypothesis of this paper.
Hypothesis 2
Financial constraints act as a mediating mechanism that influences default risk through CSR.
3. Data and method
3.1. Sample selection
The CSR data used in this study were from the CNRDS database, and the financial data were from the CSMAR database, with a time range from 2011 to 2020. The selection of sampling is based on the following criteria: (1) Excluding ST listed companies; (2) Excluding financial listed companies; (3) Excluding listed companies with a large number of missing data. First, we collected CSR scores from the CNRDS database that included company characteristics but did not consider industry differences. In addition to the above three screening criteria, we consider whether the selected company has an overall and individual CSR score for the year. Second, we collect financial data from the CSMAR database, including the dependent variable default risk, the mediating variable financing constraints, and a series of control variables. Finally, we matched the data obtained from the two databases by year and ticker symbol, and the final dataset contained 26,980 observations.
3.2. Variables
3.2.1. Dependent variable
Default risk. Merton's [25] Default Distance (DD) is a market-based measure of a firm's default risk. Default occurs when a company's liabilities exceed its assets; conversely, when assets are more significant than liabilities, the firm is less likely to default. Default Distance (DD) serves as an inverse indicator, where a longer Distance to Default (DD) signifies a reduction in the default risk faced by the firm. It has been demonstrated [[26], [27], [28]] that this method provides a better estimate of default risk than accounting methods.
3.2.2. Independent variable
CSR. The CSR score consists of six areas: “Philanthropy,” “Corporate Governance,” “Diversity,” “Employee Relations,” “Environment,” and “Products.” This paper measures each of the six dimensions from the perspectives of “Strengths” and “Concerns,” where “Strengths” refers to voluntary behaviours of enterprises that aim to improve social responsibility, which can be called positive social responsibility; “Concerns” refers to behaviours that will negatively affect the legitimate claims of stakeholders in the long run, which can be called negative social responsibility [29]. This article compiles the strengths and concerns associated with the sixfold dimensions, estimating both positive and negative social responsibility. Subsequently, we deduct the negative social responsibility from the positive one, resulting in the aggregate social responsibility score.
3.2.3. Mediating variables
Our study uses the SA index as a surrogate indicator to represent firms' financing constraints [30]. As mentioned in Ref. [31], a company's financial constraints are considered more stringent when the SA index registers a negative value, mainly when its absolute magnitude is greater.
3.2.4. Control variables
Drawing on previous research [32,33], we consider various factors that could influence default risk. The initial variable is the firm's size (SIZE), represented by the natural logarithm of its total assets. Generally speaking, an increase in a firm's size typically results in a more stable cash flow and a reduction in default risk [33]. Hence, we predict a negative correlation between the firm's size and default probability. Secondly, leverage (LEV) pertains to a company's capacity to fulfil its financial responsibilities. Increased leverage raises a company's economic vulnerability and is anticipated to influence default risk positively [34,35]. Third, we consider liquidity (LIQ) to measure a corporation's ability to meet short-term liabilities. Companies with solid liquidity ratios display a lower default risk than those with weaker liquidity ratios. Our control for liquidity is realized through the liquidity ratio. Fourthly, Tobin's Q (TOBINQ) is a measurement tool to assess a corporation's operational efficiency and expansion possibilities. This measure is derived by dividing the market value by the total assets. It is forecasted that there will be a negative association between this factor and default risk. This is because an improved market performance leads to a higher degree of stability for the company [34]. Lastly, we account for tangible assets (TANG), derived by dividing tangible assets into total assets. Some scholars have found that tangible assets negatively affect firm value [36]. Therefore, it is predicted that the presence of tangible assets will benefit default risk [34] (see Table 1).
Table 1.
Variables definitions.
| Name | Measurement | Source |
|---|---|---|
| CSR | subtract positive and negative social responsibility to obtain the total social responsibility score | CNRDS |
| DD | based on the average annual distance between stock defaults | CSMAR |
| SA | it is calculated by the size of the enterprise and the operating years | CSMAR |
| SIZE | logarithm of total assets | CSMAR |
| LEV | the ratio of total liabilities to total assets | CSMAR |
| LIQ | the ratio of current assets to current liabilities | CSMAR |
| TOBINQ | the ratio of market value to total assets | CSMAR |
| TANG | the ratio of total tangible assets to total assets | CSMAR |
3.2.5. Estimation model
The following model is used to estimate the impact of CSR on default risk:
| Default riski,t = β0+β1 CSRi,t+β2 CSR2i,t+ β3Controls i,t + Year + Industry+εi,t | (1) |
To examine the mediating influence of financial constraints, we adopt the U-shaped mediation effect testing method to construct the following assessment model for the transmission test mechanism:
| SA = α0+α1 CSRi,t+α2 CSR2i,t+ α3Controls i,t + Year + Industry+εi,t | (2) |
| Default riski,t = σ0+σ1 CSRi,t+σ2 CSR2i,t+ σ3 SA + σ4Controls i,t + Year + Industry+εi,t | (3) |
Default Distance (DD) is a metric to quantify the default risk associated with the dependent variable. CSR is obtained by subtracting the Company i's CSR positive and negative scores in year t. The control variables in Section 3.2.4 are controls. The Global Industry Classification Standard (GICS) industry code serves as the basis for the industry effect, while the year effect refers to the sample period (2011–2020).
4. Empirical results
4.1. Descriptive statistics
Table 2 presents the descriptive statistics concerning CSR, default risk, and control variables. The average CSR score is 13.09, with the lowest value at 0 and the highest at 36. This suggests a significant disparity in CSR scores among corporations, yet the overall score could be higher. The average score of default distance is 7.92, the minimum is 1.55, and the maximum is 27.97, which is slightly higher than the previous study results [33]. The average size of a company is $2.221 billion, with $2.62 of current assets to cover $1 of current liabilities (LIQ), 44 % of capital structure debt (LEV), growth opportunities (TOBINQ), and tangible assets (TANG) averaging 2.03 and 0.21, respectively.
Table 2.
Descriptive statistics.
| Variable | Obs. | Mean | SD | Min | Max |
|---|---|---|---|---|---|
| CSR | 9939 | 13.09 | 9.90 | 0 | 36 |
| DD | 27340 | 7.92 | 3.77 | 1.55 | 27.97 |
| SIZE | 22889 | 22.21 | 1.87 | 0 | 26.51 |
| LEV | 22889 | 0.44 | 0.21 | 0 | 0.93 |
| LIQ | 27340 | 2.62 | 2.86 | 0.26 | 29.76 |
| TOBINQ | 27080 | 2.03 | 1.38 | 0.81 | 15.61 |
| TANG | 27340 | 0.21 | 0.16 | 0 | 0.72 |
Fig. 1, Fig. 2 provide a more intuitive presentation of CSR scores and default risk. As shown in Fig. 1, CSR scores are mainly concentrated in the range of 10–20. This is consistent with the average CSR score of 13.09, as obtained in Table 2. From Fig. 2, we can find that the default risk of enterprises is mainly concentrated in the range of 0–10. The average value of default risk in Table 2 is 7.92. At the same time, the maximum and minimum values of CSR and default risk are also consistent with the results in Table 2.
Fig. 1.
Csr score.
Fig. 2.
Default distance.
4.2. Main results
First, we examined the hypothesis that CSR is associated with an inverse U-shape of default risk. Table 3 corroborates our hypothesis, indicating that the coefficient of the primary term in CSR displays a negative significance. Meanwhile, the coefficient of the secondary term demonstrates a positively significant relationship with the default distance. The findings reveal that the influence of CSR on the default distance exhibits a U-shaped pattern, whereas its impact on default risk demonstrates an inverted U-shaped trend. Once a specific threshold is crossed, a positive connection exists between CSR and default risk. However, once a particular threshold is reached, CSR has a counterproductive effect on default risk, effectively supporting Hypothesis 1. This finding complements the existing literature, which usually assumes that the fulfilment of CSR reduces default risk. Firms can only benefit from mitigating default risk if they ensure their social responsibility score exceeds a certain threshold level.
Table 3.
The effect of CSR on default risk.
| (1) DD | (2) DD | |
|---|---|---|
| CSR | 0.003 (0.30) | −0.015* (−1.79) |
| CSR^2 | 0.001 (1.42) | 0.001*** (4.55) |
| SIZE | 0.203*** (13.70) | |
| LEV | −7.476*** (−34.23) | |
| LIQ | 0.150*** (5.76) | |
| TOBINQ | 0.512*** (19.18) | |
| TANG | 1.082*** (4.80) | |
| Constant | 7.988*** (144.93) | 5.489*** (16.83) |
| Industry effect | No | YES |
| Year effect | No | YES |
| Observations | 9936 | 9436 |
| R2 | 0.391 | 0.575 |
4.3. Robustness test
4.3.1. Other measurements of CSR
In the main framework, we integrate the strengths and challenges of each of the six aspects to identify CSR's beneficial and detrimental aspects. Subsequently, we deduct CSR's constructive and destructive elements to calculate the overall social responsibility index. This section draws upon previous research [9] to employ alternative indicators for assessing a company's CSR. More specifically, we utilize a company's ESG ranking as a proxy for evaluating its CSR. The ESG framework is a measurement tool for assessing a corporation's environmental, social, and governance performance. These assessments include employee welfare, the firm's operational environment, and compliance strategies. From the first column of Table 4, we conclude that the primary outcome remains consistently stable and reliable. ESG exhibits an inverted U-shaped relationship with default risk, which is significant at the 5 % level.
Table 4.
Robustness analysis.
| (1) DD | (2) DD | (3) DD | (4) DD_Bhsh | |
|---|---|---|---|---|
| ESG | −0.014* (−1.76) | |||
| ESG^2 | 0.000** (2.28) | |||
| SIZE | 0.223*** (16.30) | 0.225*** (15.00) | 0.196*** (15.51) | 0.144*** (7.49) |
| LEV | −6.651*** (−50.01) | −7.843*** (−35.98) | −7.599*** (−33.56) | −1.304*** (−4.65) |
| LIQ | 0.126*** (9.57) | 0.133*** (4.86) | 0.151*** (5.32) | 0.167*** (4.26) |
| TOBINQ | 0.547*** (31.95) | 0.532*** (20.41) | 0.499*** (19.15) | 0.096*** (2.90) |
| TANG | 1.385*** (9.62) | 0.802*** (3.35) | 0.723*** (2.96) | 1.749*** (5.61) |
| L.CSR | −0.015* (−1.70) | |||
| L.CSR^2 | 0.002*** (4.24) | |||
| L2.CSR | −0.019* (−2.00) | |||
| L2.CSR^2 | 0.002*** (4.73) | |||
| CSR | −0.028** (−2.57) | |||
| CSR^2 | 0.003*** (5.73) | |||
| Constant | 4.319*** (13.51) |
5.015*** (15.25) |
5.281*** (19.67) |
7.596*** (18.56) |
| Industry effect | YES | YES | YES | YES |
| Year effect | YES | YES | YES | YES |
| Observations | 22301 | 8253 | 7171 | 7740 |
| R2 | 0.534 | 0.591 | 0.588 | 0.574 |
4.3.2. The role of lagged time effect
The potential delayed impact of CSR on default risk could be observed. Therefore, Table 4 presents the findings of the effects of CSR on default risk, with a one-period lag and a two-period lag. As seen in columns (2) and (3) of Table 4, our findings demonstrate significant robustness, covering one-year and two-year time lags. An inverse U-shaped relationship exists between CSR and default risk, with substantial characteristics at the 1 % statistical significance level.
4.3.3. Other measurements of DD
In this section, we modify the evaluation method by changing the dependent variable's default distance to assess the stability of our results. Instead of using DD_merton as a measure of default risk, we use DD_Bhsh. DD_Bhsh uses a more simplified method to calculate the default distance. Observing column (4) of Tables 4 and it can be discerned that employing DD_Bhsh as an alternative to DD_Merton for assessing default risk yields a robust outcome. This examination explores the existence of an inverted U-shaped relationship between CSR and default risk.
4.4. Mediating effect
Columns (1) through (3) of Table 5 demonstrate the mediating effect of financial constraints within the correlation between CSR and default risk. Column (1) of Table 5 reveals that the coefficient associated with the quadratic term of CSR concerning the default distance is 0.001, demonstrating statistical relevance at a 10 % significance level. This finding corroborates the significance of β2 in Model (1). Column (2) shows that the quadratic term coefficient of CSR on the default distance is −0.005 and significant at a 1 % level, which supports the significance of model (2) α2. The fulfilment of social responsibility on the left side of the U-shape intensifies financing constraints, while on the right side, an increase in social responsibility scores mitigates these constraints. Column (3) indicates that the impact of financial constraints on corporate default distance is portrayed by a coefficient of −0.040. The statistical significance of this coefficient has been established at a 10 % level, which supports the significance of model (3) σ3. Additionally, an analysis of the significance of σ2 reveals that the coefficient associated with the quadratic term of CSR about the default distance is 0.001. Statistically speaking, the findings hold significant value at the 10 % level. This suggests that financing constraints serve as a mediating pathway through which CSR influences default risk. The result supports the transmission path CSR → financing constraints → default risk, and the establishment of H2.
Table 5.
SA's mediating influence on the correlation between CSR and default risk.
| (1) DD | (2) SA | (3) DD | |
|---|---|---|---|
| CSR | −0.015* (−1.79) | 0.212*** (35.26) | −0.011 (−1.01) |
| CSR^2 | 0.001* (4.55) | −0.005*** (−22.15) | 0.001* (2.47) |
| SA | −0.040* (−1.88) | ||
| SIZE | 0.203*** (13.70) | 0.021*** (3.67) | 0.160*** (10.91) |
| LEV | −7.476*** (−34.23) | −0.204*** (−1.86) | −6.754*** (−27.43) |
| LIQ | 0.150*** (5.76) | −0.011 (−0.85) | 0.198*** (6.68) |
| TOBINQ | 0.512*** (19.18) | 0.016 (1.38) | 0.495*** (16.70) |
| TANG | 1.082*** (4.80) |
−0.351*** (−2.58) | 1.312*** (5.16) |
| Constant | 5.489*** (16.83) | −4.068*** (−32.46) | 5.857*** (17.75) |
| Industry effect | YES | YES | YES |
| Year effect | YES | YES | YES |
| Observations | 9436 | 6709 | 6709 |
| R2 | 0.575 | 0.354 | 0.584 |
4.5. Heterogeneity analysis
The previous study analyzed the impact of CSR on default risk, utilizing a comprehensive dataset as the research framework, and confirmed the relationship between them through robustness tests. This section further embedded ownership nature and firm size into the total sample for the sub-sample test.
4.5.1. Ownership
Under different ownership natures, CSR may affect corporate default risk asymmetrically. Based on their ownership structure, Chinese companies listed on stock exchanges are categorized into state-owned enterprises (SOEs) and non-state-owned enterprises (NSOEs). SOEs maintain substantial connections with the administrative sector and possess a considerable advantage in allocating limited resources. NSOEs experience a deficiency in governmental assistance and face more severe credit discrimination, trade restrictions, and other unfair treatment [37]. Therefore, SOEs are more capable of carrying out CSR activities. In the primary effect test mentioned above, we find that only when CSR performance reaches a certain threshold can the default risk be reduced; otherwise, the practice of CSR activities will only increase the default risk. Therefore, CSR can better prevent state-owned enterprises from defaulting.
Table 6 presents the outcomes of group assessments based on ownership characteristics. The analysis of columns (1) and (2) suggests that the impacts of CSR's primary and secondary factors are not statistically significant in NSOEs. The positive trend of CSR among SOEs is significant and notable. The effectiveness of CSR in mitigating default risk is noticeably greater in SOEs. This is because SOEs are generally subject to weaker budgetary constraints, and close ties with the government provide sufficient funds for their operations. Hence, they can better invest in social responsibility activities, so the probability of default is relatively low.
Table 6.
Heterogeneity analysis.
|
Variables |
Dependent variable:DD |
|||
|---|---|---|---|---|
| (1) SOEs | (2) NSOEs | (3) Large |
(4) SMEs | |
| CSR | 0.003 (0.24) | −0.014 (−1.03) | 0.002 (0.20) | −0.060*** (−2.62) |
| CSR^2 | 0.001*** (3.44) | 0.000 (0.86) | 0.001* (1.77) | 0.002** (2.18) |
| SIZE | 0.247*** (10.16) | 0.136*** (7.74) | 0.628*** (14.01) | 0.105*** (6.26) |
| LEV | −9.179*** (−30.74) | −4.978*** (−13.87) | −9.472*** (−30.30) | −4.197*** (−10.44) |
| LIQ | 0.112*** (2.62) | 0.277*** (7.09) | 0.052 (0.95) | 0.283*** (6.98) |
| TOBINQ | 0.454*** (9.78) | 0.528*** (15.77) | 0.532*** (8.83) | 0.423*** (10.81) |
| TANG | 1.170*** (4.05) |
1.524*** (3.83) |
0.832*** (2.79) |
2.152*** (4.43) |
| Constant | 5.542*** (10.02) |
5.336*** (14.26) |
−3.493*** (-3.42) |
7.002*** (20.33) |
| Industry effect | YES | YES | YES | YES |
| Year effect | YES | YES | YES | YES |
| Observations | 5487 | 3525 | 5302 | 2434 |
| R2 | 0.612 | 0.555 | 0.586 | 0.564 |
In contrast, NSOEs are limited by limited resources and face more severe financial exclusion and credit discrimination. They usually need insufficient investment in social responsibility activities, and the influence of CSR on enhancing reputation, profitability, and alleviating financing constraints may not be explicitly evident, making it challenging to minimize default risk significantly. Consequently, CSR demonstrates a more substantial mitigating influence on the default risk of SOEs.
4.5.2. Size
In the context of stakeholder theory and social capital, Russo and Perrini concluded that large enterprises and small and medium enterprises (SMEs) must be distinguished because they have different characteristics to consider. Large financial institutions are often more willing to lend money to large businesses based on comprehensive financial information. SMEs usually obtain expensive financing through informal financial channels, resulting in significant debt repayment burdens and heightened default risks. The fulfilment of CSR increases the information transparency of SMEs, mitigates the disparity of information between financial institutions and corporate entities, optimizes the risk monitoring of SMEs by banks, and thus reduces the threshold and cost of financing for SMEs through formal financial channels such as banks. Therefore, we infer that CSR can prevent SMEs from defaulting.
Table 6 shows the group test results based on firm size. Based on the findings in columns (3) and (4), the influence of CSR on default risk varies among enterprises of varying sizes. While the coefficients of the second term of CSR for both large enterprises and SMEs exhibit significantly positive values, the effect of CSR on the latter is comparatively more significant. This demonstrates that CSR's influence on minimizing SME's default risk is more pronounced. Large enterprises have high credit ratings, solid political connections, sufficient collateral, easy access to low-cost financing in the financial market, and low debt default risk. As a result, CSR has no noticeable inhibiting effect on the default risk of large enterprises.
Conversely, the operational scale, market share, information transparency, etc., all impact SMEs. The demand for capital scale is small and scattered, resulting in financial institutions needing to invest more human and material capital to lend to them, which increases their operating costs. Financial institutions seeking efficiency are more willing to lend to large enterprises. Therefore, CSR is more effective in preventing the default of SMEs.
5. Conclusion
In recent years, with the increase in corporate liquidity and debt repayment pressure, the “insurance” role of CSR has received more and more attention. However, the existing literature rarely links CSR and default risk, and the mechanism that CSR affects default risk needs to be revealed. We conducted a comprehensive empirical analysis of 4988 listed companies in the Chinese context during 2011–2020. Our research has three findings: First, the initial increase in CSR will lead to a rise in default risk. However, once the CSR level exceeds a specific threshold, the default risk decreases as the CSR rises. We tested the robustness of the results by replacing the explanatory and the explained variables and taking into account the lag time effect, which proved the reliability of our research conclusions. Second, the mediation analysis shows that financing constraints play an important mediating role in this inverted U-shaped relationship. On the left side of the U-shape, CSR performance intensifies financing constraints, while on the right side, increasing CSR reduces financing constraints. Finally, we confirm heterogeneity in the impact of CSR on the default risk of different enterprises' ownership and size. Unlike previous studies, we do not find that CSR has an advantage in reducing the default probability of NSOEs [38]. In contrast, SOEs use their inherent policy and financial benefits to minimize the default risk by fulfilling CSR more effectively. Exploring scale heterogeneity confirmed the research results [39], emphasizing CSR's important role in reducing SMEs' default risk.
Our research results have specific theoretical significance. Existing studies based on stakeholder theory believe that CSR can alleviate default risk [14]. Alternatively, based on the shareholder and agency theory, CSR is thought to increase the default risk [15]. Our research combines the two theories and finds that the impact of CSR on default risk is not a simple linear relationship but a more complex U-shaped relationship. Before reaching the threshold, CSR will lead to an increase in default risk, consistent with the shareholder and agency theory. After reaching the threshold, CSR will lead to a decline in default risk, which aligns with the stakeholder theory. Our research results extend the stakeholder and shareholder agency theories to consider the relationship between CSR and default risk in a more comprehensive theoretical framework.
Our findings have practical implications for policymakers, companies, and investors. From the perspective of policymakers, enterprises should not be blindly encouraged to carry out CSR activities. If companies cannot improve their CSR performance, they will fail to meet the threshold and increase the default risk. Therefore, due to risk considerations, such companies should be prevented from unthinkingly carrying out CSR activities. From the perspective of enterprises, while continuously improving the transparency of CSR performance, CSR concepts should also be integrated into corporate culture's construction and development process. However, a lower level of CSR will increase operating costs and the default risk. Therefore, enterprises should appropriately carry out CSR activities according to their capabilities. From the perspective of investors, when choosing an investment object, in addition to whether a company discloses its CSR performance, it should also pay attention to the level of disclosure. A higher default risk premium should be required for companies with no disclosure and low scores.
Finally, our study has some limitations. First, our research methods have limitations. The data we used were all secondary data from the database, and although we conducted various robustness tests, the study's results may still be biased. In the future, we can choose more diversified research methods to support our conclusions, such as data capture and semi-structured interviews. Second, our mediation mechanism has limitations. We mainly analyze the mediating role of financing constraints in the relationship between CSR and default risk. However, other factors may also influence the relationship. Therefore, it is possible to explore whether different mediation mechanisms will lead to varying effects of CSR on default risk in the future.
Funding
The present study was funded by the [National Natural Science Foundation of China] under Grant [72171162] and the [Youth Fund for Humanities and Social Sciences of the Ministry of Education] under Grant [20YJC630148].
Data availability statement
Data will be made available on request.
CRediT authorship contribution statement
Wenli Wang: Writing – review & editing, Funding acquisition, Conceptualization. Jiasi Yang: Writing – original draft, Investigation, Data curation.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
We appreciate the insightful comments the anonymous reviewers and editors provided on our manuscript.
References
- 1.Boubaker S., et al. Does corporate social responsibility reduce financial distress risk? Econ. Modell. 2020;91:835–851. [Google Scholar]
- 2.Yang S.L., et al. Can country trade flows benefit from improved corporate social responsibility ratings? Econ. Modell. 2019;80:192–201. [Google Scholar]
- 3.Desender K.A., et al. Corporate social responsibility and cost of financing-The importance of the international corporate governance system. Corporate Governance-an International Review. 2020;28(3):207–234. [Google Scholar]
- 4.Lee D.D., Faff R.W. Corporate sustainability performance and idiosyncratic risk: a global perspective. Financ. Rev. 2009;44(2):213–237. [Google Scholar]
- 5.Goss A., Roberts G.S. The impact of corporate social responsibility on the cost of bank loans. J. Bank. Finance. 2011;35(7):1794–1810. [Google Scholar]
- 6.Mishra S., Modi S.B. Positive and negative corporate social responsibility, financial leverage, and idiosyncratic risk. J. Bus. Ethics. 2013;117(2):431–448. [Google Scholar]
- 7.Kim Y., Li H.D., Li S.Q. Corporate social responsibility and stock price crash risk. J. Bank. Finance. 2014;43:1–13. [Google Scholar]
- 8.Bouslah K., Kryzanowski L., M'Zali B. Social performance and firm risk: impact of the financial crisis. J. Bus. Ethics. 2018;149(3):643–669. doi: 10.1007/s10551-016-3017-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Farah T., et al. The non-linear effect of CSR on firms? systematic risk: international evidence q. J. Int. Financ. Mark. Inst. Money. 2021;71 [Google Scholar]
- 10.Nadarajah S., et al. Stock liquidity and default risk around the world. J. Financ. Mark. 2021;55 [Google Scholar]
- 11.Giesecke K., et al. Corporate bond default risk: a 150-year perspective. J. Financ. Econ. 2011;102(2):233–250. [Google Scholar]
- 12.Do T.K. Corporate social responsibility and default risk: international evidence. Finance Res. Lett. 2022;44 [Google Scholar]
- 13.Badayi S.A., et al. Does corporate social responsibility influence firm probability of default? Int. J. Finance Econ. 2021;26(3):3377–3395. [Google Scholar]
- 14.Freeman R.E. Cambridge university press; 2010. Strategic Management: A Stakeholder Approach. [Google Scholar]
- 15.Jensen M.C., Meckling W.H. Theory of firm - managerial behavior, agency costs and ownership structure. J. Financ. Econ. 1976;3(4):305–360. [Google Scholar]
- 16.Godfrey P.C., Merrill C.B., Hansen J.M. The relationship between corporate social responsibility and shareholder value: an empirical test of the risk management hypothesis. Strat. Manag. J. 2009;30(4):425–445. [Google Scholar]
- 17.Luo X.M., Bhattacharya C.B. Corporate social responsibility, customer satisfaction, and market value. J. Market. 2006;70(4):1–18. [Google Scholar]
- 18.Spicer B.H. INVESTORS, corporate social performance and information disclosure - empirical-study. Account. Rev. 1978;53(1):94–111. [Google Scholar]
- 19.Herremans I.M., Akathaporn P., McInnes M. An investigation of corporate social responsibility reputation and economic performance. Account. Org. Soc. 1993;18(7–8):587–604. [Google Scholar]
- 20.Albuquerque R., Koskinen Y., Zhang C.D. Corporate social responsibility and firm risk: theory and empirical evidence. Manag. Sci. 2019;65(10):4451–4469. [Google Scholar]
- 21.Barnea A., Rubin A. Corporate social responsibility as a conflict between shareholders. J. Bus. Ethics. 2010;97(1):71–86. [Google Scholar]
- 22.Godfrey P.C. The relationship between corporate philanthropy and shareholder wealth: a risk management perspective. Acad. Manag. Rev. 2005;30(4):777–798. [Google Scholar]
- 23.Cespa G., Cestone G. Corporate social responsibility and managerial entrenchment. J. Econ. Manag. Strat. 2007;16(3):741–771. [Google Scholar]
- 24.Wang H., Hou C., Liu H. Strategic deviance, performance aspiration disparity and default risk. Nankai Business Review. 2019;22(4):4–19. [Google Scholar]
- 25.Merton R.C. On the pricing of corporate debt: the risk structure of interest rates. J. Finance. 1974;29(2):449–470. [Google Scholar]
- 26.Duan J.C., Sun J., Wang T. Multiperiod corporate default prediction-A forward intensity approach. J. Econom. 2012;170(1):191–209. [Google Scholar]
- 27.Bharath S.T., Shumway T. Forecasting default with the Merton distance to default model. Rev. Financ. Stud. 2008;21(3):1339–1369. [Google Scholar]
- 28.Das S.R., Hanouna P., Sarin A. Accounting-based versus market-based cross-sectional models of CDS spreads. J. Bank. Finance. 2009;33(4):719–730. [Google Scholar]
- 29.Fu R.C.Y., Tang Y., Chen G.L. Chief sustainability officers and corporate social (Ir)responsibility. Strat. Manag. J. 2020;41(4):656–680. [Google Scholar]
- 30.Hadlock C.J., Pierce J.R. New evidence on measuring financial constraints: moving beyond the KZ index. Rev. Financ. Stud. 2010;23(5):1909–1940. [Google Scholar]
- 31.Ju X., Lu D., Yu Y. Financing constraints, working capital management and the persistence of firm innovation. Econ. Res. J. 2013;1:4–16. [Google Scholar]
- 32.Ali S., Liu B., Su J.J. Does corporate governance quality affect default risk? The role of growth opportunities and stock liquidity. Int. Rev. Econ. Finance. 2018;58:422–448. [Google Scholar]
- 33.Atif M., Ali S. Environmental, social and governance disclosure and default risk. Bus. Strat. Environ. 2021;30(8):3937–3959. [Google Scholar]
- 34.Chiang S.M., Chung H.M., Huang C.M. A note on board characteristics, ownership structure and default risk in Taiwan. Account. Finance. 2015;55(1):57–74. [Google Scholar]
- 35.Schultz E.L., Tan D.T., Walsh K.D. Corporate governance and the probability of default. Account. Finance. 2017;57:235–253. [Google Scholar]
- 36.Nakano M., Nguyen P. Foreign ownership and firm performance: evidence from Japan's electronics industry. Appl. Financ. Econ. 2013;23(1):41–50. [Google Scholar]
- 37.Zhang M., et al. Do suppliers applaud corporate social performance? J. Bus. Ethics. 2014;121(4):543–557. [Google Scholar]
- 38.Shih Y.C., et al. Corporate environmental responsibility and default risk: evidence from China. Pac. Basin Finance J. 2021;68 [Google Scholar]
- 39.Wu D.X., et al. Does government initiated corporate social responsibility lower the default risk? Evidence from the targeted poverty alleviation campaign in China. Pac. Basin Finance J. 2022;76 [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data will be made available on request.