Indifferent to CSR performance? Market liberalization brings about changes

Abstract

The relationship between stock market performance and corporate social responsibility (CSR) activities indicates the extent of acceptance by investors of such activities. Based on a sample of Chinese-listed companies and a quasi-natural experiment, we explore the impact of an important form of market liberalization—the Shanghai-Hong Kong and Shenzhen-Hong Kong Stock Connect—on the extent of investors' acceptance of CSR using a difference-in-differences approach. The results show that Stock Connect reinforces the favorable correlation between CSR performance and the market response to CSR reports, as well as the positive association between CSR performance and Tobin's Q. This effect is particularly pronounced for nonstate-owned enterprises,firms with lower agency costs, and firms with a lower possibility of using CSR reports for impression management. Thus, market liberalization results in the improved short- and long-term stock market performance of companies with strong CSR records. Further research shows that the benefits of the program are more prominent for companies without Qualified Foreign Institutional Investors (QFII) shareholdings, and the impact of significant foreign ownership through the Stock Connect program on the stock market performance related to CSR engagement is incremental over QFII ownership. The findings support the continuous liberalization of China's capital market.

1. Introduction

China's reform and opening are key to its economic success. In November 2014 and December 2016, the Shanghai-Hong Kong Stock Connect and Shenzhen-Hong Kong Stock Connect (hereafter, SHSH Stock Connect) programs were launched, respectively, allowing for increased liberalization of China's stock market. Past literature documented that the SHSH Stock Connect not only led to higher liquidity, information efficiency, and stability of the capital market [[1], [2], [3], [4]] but also improved corporate governance, promoted corporate green innovation and reduced litigation risk for firms [[5], [6], [7]]. However, little attention has been paid to the social effects of market liberalization.

In becoming increasingly aware of their importance, firms now highlight the decision to engage in corporate social responsibility (CSR). Management must reconcile the demands of shareholders with those of a wider group of stakeholders when making decisions on expenditures on CSR. To some extent, firms' willingness to engage in CSR is affected by the stock market acceptance of CSR [8,9]. With low awareness and acceptance of CSR, market investors do not value CSR, which reduces the incentives for firms to pursue these activities. CSR is highly valued by investors from developed economies [9,10]. However, in contrast [11], found no significant relationship between CSR scores and market reactions to CSR reports before the implementation of Stock Connect in China. In addition [12,13], documented that CSR engagement does not appear to collectively improve Chinese-listed companies' long-term market performance (Tobin's Q). These studies suggested that better CSR performance does not necessarily yield higher market acceptance and superior stock market performance for Chinese-listed companies. A potential explanation for these findings is that Chinese investors may be indifferent to firms' CSR initiatives [14] or may lack the ability to accurately and comprehensively evaluate and interpret the details of a firm's CSR practices, resulting in the failure to incorporate this information into their valuation processes.

Nevertheless, China's market liberalization initiatives could lead to positive changes that increase market acceptance of CSR practices. The SHSH Stock Connect has facilitated mutual market access to trading-designated stocks, allowing Hong Kong and foreign investors to trade eligible A-shares listed on the Shanghai and Shenzhen Stock Exchanges, while mainland investors can now trade eligible stocks listed on the Hong Kong Stock Exchange. There is strong community acceptance of the need for CSR, supported by a regulatory framework that encourages CSR engagement in developed economies [9]. The importance imposed on CSR is greater in developed economies than in emerging markets [15]. The implementation of Stock Connect changes the investor base and opens the market for global investors with greater awareness of CSR [16]. In addition, these investors also have stronger abilities to interpret, process and analyze information about a company's CSR performance and use it for stock valuation [2]. Based on the above analysis, we expect that SHSH Stock Connect will improve the stock market performance related to CSR as it introduces foreign investors.

Consistent with our predictions, we find that the SHSH Stock Connect improves the short- and long-term stock market performance related to CSR engagement by employing the implementation of the Stock Connect program as a quasi-natural experiment. Stock Connect strengthens the positive relationship between the CSR score and CSR report announcement return and between the CSR score and Tobin's Q. This implies a higher market valuation of CSR and improved stock market performance of companies with strong CSR records after market liberalization that will, in turn, encourage firms to fulfill CSR.

This paper's possible contributions are as follows. First, it adds to the literature on the economic consequences of CSR activities. Prior studies documented that the fulfillment of CSR has been valued by investors from developed economies [9,10], while empirical evidence from China suggests that active CSR engagement may not necessarily lead to superior stock market performance [[11], [12], [13], [14]]. We provide empirical evidence that market liberalization helps improve the short- and long-term stock market performance of Chinese firms that actively engage in CSR. Our findings help alleviate the concerns of managers who are torn between satisfying the interests of shareholders and those of other stakeholders.

Second, our study contributes to the literature on the economic impacts of market liberalization. While previous studies focused primarily on the influence of market liberalization on overall stock market efficiency and a firm's behavior and decisions, our research extends this to demonstrate that market liberalization also has social implications. Specifically, we provide empirical evidence that market liberalization leads to improved stock market performance for companies with strong CSR records. This finding highlights the positive social effects of market liberalization.

Third, we explore the effect of the two important forms of liberalization in the Chinese stock market, while most prior literature discussed only one form of market liberalization. We consider the impact of not only the SHSH Stock Connect program but also the Qualified Foreign Institutional Investors (QFII) scheme on the stock market performance related to CSR engagement. We show that improvements in the stock market performance related to CSR engagements brought by Stock Connect are more pronounced for firms without QFII shareholdings. In addition, the impact of significant foreign ownership through the Stock Connect program on the stock market performance related to CSR engagement is incremental over that of QFII ownership.

2. Institutional background

The business environment in China is frequently described as involving ethically questionable and frequently socially irresponsible business decisions [17]. According to Refs. [18,19]; Chinese businesses that struggle to maximize profits frequently pursue irresponsible development strategies. The development of CSR initiatives in China differs from that of many other countries in that the Chinese government took the initiative of promoting CSR practices. Although there is no mandatory legislation on corporate social responsibility (CSR) in China, the government has issued a series of policy documents to encourage and support enterprises in fulfilling their social responsibilities. For instance, in January 2002, the China Securities Regulatory Commission released the Code of Corporate Governance for Listed Firms to promote CSR awareness and encourage firms to fulfill their social responsibilities towards the environment and stakeholders. In 2007, the State-owned Assets Supervision and Administration Commission of the State Council initiated a research project on the social responsibility of central government-owned enterprises and planned to issue guiding opinions to promote these enterprises in fulfilling their social responsibilities, with CSR becoming a crucial assessment criterion for enterprises. On January 4, 2008, the State-owned Assets Supervision and Administration Commission of the State Council issued the “Guiding Opinions on the Fulfillment of Social Responsibilities by Central Government-owned Enterprises,” requiring these enterprises to serve as models in fulfilling social responsibilities and fully integrating social responsibility concepts and requirements into their development strategies.

CSR disclosure has emerged within the last two decades in China [20]. Several guidelines have been established by the Chinese regulatory parties to meet the increasing demand for CSR disclosures. In 2008, the Shanghai and Shenzhen stock exchanges mandated the disclosure of an annual CSR report for a subset of firms. Specifically, the Shanghai Stock Exchange (SSE) requires firms included in the SSE Corporate Governance Sector, firms with shares listed overseas, and financial companies to disclose the annual CSR reports. The Shenzhen Stock Exchange (SZSE) requires firms included in the Shenzhen 100 index to release CSR reports [21]. Currently, there are no mandatory policies or regulations in China that stipulate the content and format of CSR reports. Enterprises can voluntarily choose the content and format of disclosure based on their actual situation and needs. However, some industry associations and research institutes have prepared relevant CSR standards and guidelines, such as the China CSR Evaluation Guidelines prepared by the China Enterprise Evaluation Association (CEEA) and the Guidelines for Preparation of CSR Reports in China prepared by the Chinese Academy of Social Sciences (CASS), which can provide reference and guidance for enterprises in preparing CSR reports. In addition, regulatory authorities have also issued a number of guiding documents to encourage listed companies to publish CSR reports and provide guidelines on the content and manner of disclosure of CSR information, including the Guidelines on Social Responsibility for Listed Companies on the Shenzhen Stock Exchange.

To sum up, there is no mandatory legal requirement for the fulfillment of social responsibility in China. Regarding the disclosure of social responsibility reports of listed companies, stock exchanges have mandated the disclosure of an annual CSR report for a subset of firms, while the rest of the companies can choose to voluntarily disclose their reports. Although government agencies and industry associations have provided some guidelines on the content format and aspects of social responsibility reports, there is no strict supervision by laws and regulations.

3. Literature review and hypothesis development

Fulfilling social responsibility can help companies gain legitimacy [22], enhance their reputations [23], attract socially responsible consumers [24], facilitate financing [25], assist in recovering quickly from financial distress [26], and reduce the impact of negative public news on investor sentiment [27]. For these reasons, investors may support the fulfillment of CSR and increase the valuation of companies with high social responsibility performance. Consistent with the above studies [9,10], found that the fulfillment of CSR has been valued by investors, indicating a positive correlation between a firm's CSR rating and its stock market performance.

However, empirical evidence from China suggests that there may not be a significant relationship between CSR and its short- or long-term stock market performance. Using data from the Chinese capital market between 2009 and 2016 [11], revealed that, overall, there was no significant correlation between the CSR scores and the cumulative abnormal return around the release of CSR reports. They concluded that the release of CSR reports with higher CSR scores did not lead to significant stock market reactions. Similarly, based on merger and acquisition samples from the Chinese market [14], documented that the CSR score does not have a substantial effect on the cumulative abnormal returns over the initial announcement of acquirers, indicating that Chinese investors are not concerned about the CSR engagement of acquirers. In addition [12,13], found that CSR engagement has no significant impact on long-term stock market performance as measured by Tobin's Q.

In summary, superior social responsibility performance has not yet brought higher capital market valuation and acceptance by Chinese investors, meaning that the release of CSR reports with higher CSR ratings may not necessarily result in significantly positive stock market reactions, and there is no significant positive relationship between a company's CSR score and its long-term market performance as measured by Tobin's Q. We believe there may be two main reasons for this. First, as noted by Ref. [11]; market participants in China have too low awareness and sensitivity to CSR engagement and CSR information. The effectiveness of CSR information disclosure and the impact of CSR on firms' stock market performance are determined by the absorptive and cognitive capacities of stakeholders [28]. Research on multinational corporations has shown that CSR has a greater impact on firm value and other financial performance indicators in more stakeholder-oriented economies [29,30]. Investors from emerging markets are more concerned about short-term profits than long-term-oriented CSR performance [31]. When stakeholders have lower awareness and sensitivity to CSR performance, pursuing CSR may not have a significant positive impact on the company's stock market performance. Second, investors have poor information acquisition and analysis capabilities. For a long time, individual investors have prevailed in China's capital markets, with relatively insufficient feedback on and mining of corporate-specific information [32]. These investors have failed to accurately and fully analyze and interpret information about a company's CSR activities and use it for valuation and investment decision making. Consequently, such information has not been fully incorporated into the market price.

If firms’ strong performance in social responsibility fails to translate into better stock market performance, they may feel less motivated to continue their social responsibility efforts. However, the implementation of Stock Connect offers hope for improvement in this regard for the following two reasons.

First, foreign investors may have higher sensitivity and awareness of CSR. The SHSH Stock Connect enables investors from the Hong Kong stock market to participate in mainland China's A-share market while also allowing investors from other regions, including those from developed economies, to access this market via this channel. According to institutional theory, developed economies exercise higher standards for CSR practices than emerging markets. The importance imposed on CSR is greater in developed economies than in emerging markets [15]. The idea of CSR originated in developed regions and countries across Europe and America. In the 1980s, these places initiated a movement promoting corporate social responsibility, with an emphasis on environmental protection and employee rights. This movement led to a shift in the focus of the market from solely business performance to encompassing various aspects of social responsibility, such as product quality and safety, environmental protection, employee rights, and community maintenance [33]. Therefore, investors from developed areas introduced through market liberalization may have a greater understanding of social responsibility and take a keen interest in its fulfillment and disclosure [15,34,35]. They often possess long-standing and ingrained attitudes toward CSR. With the growing sensitivity and awareness of investors toward CSR, the influence of CSR engagement on investment decision making becomes more significant. Consequently, firms' CSR activities will be more valued by the market as Stock Connect introduces more foreign investors.

Second, foreign investors who participate in the A-share capital market have a greater ability to recognize value and are often more sophisticated in their investment approaches than local investors [36]. In more developed capital markets, such as Hong Kong's stock market, institutional investors are prevalent. These investors often possess wealthier investment experience, stronger investment teams, and more professional investment capabilities than those in mainland China's A-share market. As a result, foreign investors may have a competitive advantage in terms of information acquisition, processing, and analysis because of greater investment funds, technology, and human resources [37,38]. Therefore, foreign investors introduced through the SHSH Stock Connect program are better equipped to accurately interpret the information surrounding CSR performance, evaluate its impact on corporate value in a logical and objective manner, and utilize this insight to inform their investment decisions.

Based on the above arguments, we propose the following hypothesis:

Hypothesis 1

The implementation of the SHSH Stock Connect program has strengthened the positive correlation between firms' corporate social responsibility (CSR) scores and their stock market performance.

Previous research showed that management may fulfill social responsibility to pursue their own interests, even if at the expense of investors [39,40]. This behavior can lead to negative outcomes, such as lower company value, hampered growth, and increased risk of stock price downturns [[41], [42], [43]]. Companies with higher agency costs are more likely to engage in such practices, where managers undertake CSR efforts for their own benefit. Investors introduced to the Stock Connect program from overseas have demonstrated exceptional analytical and information processing skills [37,38]. This skillset allows them to make better use of CSR information and rationally evaluate the impact of CSR on corporate value. They do this by considering a company's CSR engagement and other relevant factors, such as internal corporate governance. Overseas investors can analyze information effectively and correctly comprehend the motivation and economic consequences of pursuing CSR in companies with high agency costs. This enables them to recognize that the fulfillment of CSR by these companies is driven by management self-interest alone and may not add value to the organization. Thus, overseas investors may not be sensitive to the CSR engagement of companies with high agency costs but may choose instead to support those of companies with lower agency costs. Taken together, investment decisions made by overseas investors are more inclined toward companies with lower agency costs that possess a strong CSR rating. Based on the above arguments, we propose the following hypothesis:

Hypothesis 2

The impact of Stock Connect on the stock market performance related to CSR is more pronounced for companies with lower agency costs than those with higher agency costs.

Different types of ownership lead to varying motivations and economic consequences of CSR initiatives [44]. State-owned enterprises (SOEs) in China have been naturally entrusted with the duty of fulfilling specific social functions since their establishment, including employment rates and improving societal welfare [45]. The inherent association between the government and SOEs makes it easier for governments to achieve political goals by intervening in the initiatives of SOEs' social responsibilities, especially in a financially constrained regional economy [46]. Consequently, compared to nonstate-owned enterprises, CSR engagement for SOEs is a manifestation of their political mission rather than their efforts to attain economic objectives. The acknowledgment of SOEs’ CSR performance by foreign investors may be relatively low, first, because state-owned enterprises are under greater pressure to fulfill social responsibilities, including those that are more compulsory [47], which may not align well with the values of overseas investors who have stronger social awareness. Second, investors introduced via the Stock Connect program possess more robust information analysis and processing capabilities, allowing them to evaluate rationally the impact of CSR on corporate value by making full and effective use of CSR information and other relevant information. Considering that CSR engagement by SOEs for political purposes does not necessarily enhance their economic value, foreign investors may be less supportive of the CSR activities of SOEs but may only acknowledge the CSR performance of nonstate-owned enterprises. When nonstate-owned enterprises have better CSR ratings, investors tend to believe that the valuation of these companies should be higher. Therefore, it can be concluded that the impact of the Stock Connect program on the stock market performance of nonstate-owned enterprises related to CSR is more pronounced. This is attributed to the fact that foreign investors tend to recognize and value the robust CSR performance of these companies, as it aligns more closely with their values and investment principles. Based on the above arguments, we propose the following hypothesis:

Hypothesis 3

The impact of Stock Connect on the stock market performance related to CSR is less pronounced for SOEs than non-SOEs.

As discussed in Section 2, although government agencies and industry associations have provided some guidelines on the content and format of social responsibility reports, there is no strict supervision by laws and regulations, and the published CSR reports have not undergone independent auditing and verification. Consequently, it is plausible that certain companies may engage in impression management through their CSR reports [48,49]. Overseas investors may have a higher ability to detect impression management in social responsibility reports compared to local retail investors, due to their greater experience and sophistication in analyzing corporate information. These investors often have access to a wider range of information sources and may be more skilled at discerning patterns and inconsistencies that suggest impression management [37,38]. If a company is more likely to engage in impression management in its CSR report, overseas investors may be more cautious in assessing its true level of CSR performance. Even if the report appears to demonstrate a high level of fulfillment, these investors may discount the information or take a more skeptical view of the company's claims. This is because they may recognize that impression management can distort the true state of a company's social responsibility performance, making it difficult to accurately assess its value. As a result, overseas investors may be less likely to place a higher valuation on companies that are more likely to engage in impression management through their CSR reports. Based on the above arguments, we propose the following hypothesis:

Hypothesis 4

The influence of Stock Connect on stock market performance related to CSR is less pronounced for companies with a higher possibility of using CSR reports for impression management than those with a lower possibility of doing so.

4. Research design

4.1. Sample and data source

Taking China's A-share companies from 2011 to 2021 as the initial sample, we obtain 25,226 firm-year samples used to estimate the baseline regression after applying the following sample selection criteria: (1) delete firms in the finance industry; (2) delete firms removed from the SHSH Stock Connect list during the sample period to ensure the validity of the difference-in-differences setting; (3) delete samples with missing data; and (4) delete delisted companies. The sample period covers the years before and after the implementation of Stock Connect.

The data on CSR scores are from Hexun.com. Hexun. com is one of the leading financial websites in China and evaluates the CSR performance of China's listed companies on a yearly basis, mainly based on the information from CSR reports and also on information from annual reports and other sources. The overall CSR score is a weighted average of the subscores relating to ① stockholders, ② employees, ③ suppliers, clients, and customers, ④ the environment, and ⑤ society and the community. The data on the announcement date of CSR reports, corporate finance indicators, corporate governance indicators, QFII shareholding, SHSH shareholding, and stock market trading data are from the China Stock Market Accounting Research (CSMAR) database.

4.2. Variables and models

Following [50]; the difference-in-differences methodology is adopted to construct regression Equations (1), (2). The market reaction to CSR reports, CAR, represents the cumulative abnormal returns calculated using the market-adjusted model, i.e., the sum of the differences between the firm's daily stock return and the market return over the three-day interval (−1, 1) around the release day of CSR reports. This variable measures the short-term market performance of CSR engagement. The firm's long-term stock market performance measure is Tobin's Q (Q), calculated as the sum of the market value of equity and book value of liabilities divided by the book value of total assets The overall CSR score from Hexun, CSR, is used to measure the CSR performance of listed companies. The dummy variable, SC, equals one if the firm is eligible for trading under either the Shanghai-Hong Kong or Shenzhen-Hong Kong Stock Connect program in that year and zero otherwise. Equations (1), (2) include the firm (FIRM) and year (YEAR) fixed effects, which control for the fixed differences between treated and nontreated firms and the aggregate fluctuations.

Following [11]; in Equation (1), we include the following control variables that may affect the market reaction to CSR reports: SIZE (firm size), LEV (debt ratio), ROA (return on assets), EPS (earnings per share), CASH (free cash flow), VOL (stock return volatility), BM (book-to-market ratio), and TURNOVER (stock turnover ratio).

Following [51]; we control for a series of corporate finance and governance variables in Equation (2): SIZE, LEV, ROA, TURNOVER, GROWTH (sales growth rate), TANG (tangible assets), TOP1 (controller shareholding), and BOARD (size of the board). Appendix A summarizes the definitions of these variables.

CAR_i,t = a₀+b₁SC_i,t-1*CSR_i,t-1+b₂CSR_i,t-1+b₃SC_i,t-1+b₄CONTROL_i,t-1+FIRM + YEAR+ε (1)

Q_i,t = a₀+b₁SC_i,t-1*CSR_i,t-1+b₂CSR_i,t-1+b₃SC_i,t-1+b₄CONTROL_i,t-1+FIRM + YEAR+ε (2)

5. Empirical results

5.1. Descriptive statistics and baseline results

Table 1 reports the descriptive statistics. All of the variables, except for the dummy variables are winsorized at the 1st and 99th percentiles. The standard deviation of CAR is much greater than its median and mean, implying that the announcement of CSR reports could result in very different market reactions. The minimum and maximum of CSR are −18.45 and 90.87, respectively, suggesting a wide variation in the CSR engagement of Chinese-listed companies. SC has a mean value of 0.260, meaning that 26.0 % of the listed companies are eligible for trading under the SHSH Stock Connect as the underlying stocks.

Table 1.

Summary statistics of the main variables.

Variables	Median	Mean	St. Dev.	Min.	Max.
CAR	−0.003	0.001	0.056	−0.308	0.475
Q	1.612	2.114	1.520	0.872	9.724
CSR	21.79	24.23	16.80	−18.45	90.87
SC	0	0.260	0.439	0	1
SIZE	21.926	22.139	1.411	18.963	26.873
LEV	0.429	0.439	0.221	0.048	0.978
ROA	0.051	0.053	0.065	−0.316	0.242
TURNOVER	5.829	5.592	1.215	−4.934	8.499
EPS	0.382	0.526	0.693	−5.573	8.112
CASH	0.019	0.008	0.099	−0.535	0.271
VOL	0.054	0.059	0.024	0.022	0.172
BM	0.696	0.682	0.250	0.102	1.135
GROWTH	0.141	0.480	1.4736	−0.829	11.652
TANG	0.178	0.213	0.167	0.002	0.733
TOP1	0.327	0.348	0.152	0.003	0.899
BOARD	2.197	2.139	0.208	1.099	3.044

Table 2 reports the results of the baseline regression Equations (1), (2). In Columns 1 and 3, we regress the short-term stock market performance (measured as the market reaction to the release of CSR reports) and the long-stock market performance (measured by Tobin's Q) on the CSR score, respectively. The results show that the coefficients on CSR are not significant in Columns 1 and 3. These findings align with [11]; who reported no significant association between CSR scores and market response to CSR reports, as well as [13,14]; who observed no significant correlation between firms' CSR scores and Tobin's Q. These studies concluded that CSR engagement did not result in superior short- or long-term stock market performance for Chinese companies.

Table 2.

Baseline results.

VARIABLES	(1)	(2)	(3)	(4)
	CAR	CAR	Q	Q
CSR	0.000	0.001	0.001	0.001
	(0.970)	(1.012)	(1.209)	(1.038)
SC*CSR		0.001***		0.003**
		(2.594)		(2.216)
SC		−0.009		0.034
		(-0.851)		(0.528)
SIZE	−0.003	−0.002	−0.347***	−0.357***
	(-1.560)	(-1.433)	(-11.518)	(-12.759)
LEV	0.022	0.021	0.502***	0.509***
	(1.572)	(1.326)	(5.063)	(5.309)
ROA	−0.000	−0.001	0.480***	0.481***
	(-0.008)	(-0.052)	(2.598)	(2.745)
EPS	0.003	0.003
	(0.946)	(0.987)
CASH	−0.007	−0.006
	(-0.388)	(-0.369)
VOL	−0.126*	−0.125*
	(-1.765)	(-1.723)
BM	0.009	0.009
	(1.011)	(1.042)
TURNOVER	0.005***	0.005***	0.014**	0.015**
	(3.556)	(3.527)	(2.006)	(2.148)
GROWTH			−0.009	−0.008
			(-1.278)	(-1.183)
TANG			0.404***	0.392***
			(2.917)	(2.766)
TOP1			0.099	0.094
			(0.577)	(0.549)
BOARD			−0.343***	−0.341***
			(-4.372)	(-4.185)
Constant	−0.097	−0.084	9.181***	9.367***
	(-0.812)	(-0.754)	(13.998)	(14.135)
FIRM&YEAR	YES	YES	YES	YES
Observations	5,639	5,639	25,226	25,226
R2	0.036	0.037	0.231	0.232

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

Next, we include the interaction term SC*CSR in Columns 2 and 4 of Table 2. The coefficients on the interaction term are significantly positive. Thus, the positive correlation between CSR scores and the market's reaction to CSR reports becomes more significant after the implementation of Stock Connect. This suggests that as Stock Connect introduces investors with greater awareness of CSR and stronger abilities to process, analyze and utilize information, the market will react more favorably to CSR reports that demonstrate that the firm has actively engaged in CSR, resulting in the firm having better short-term stock market performance. In addition, the long-term stock market performance (measured by Tobin's Q) of firms with higher CSR scores also improves after the implementation of Stock Connect. Thus, firms with strong CSR records are rewarded with superior stock market performance and are likely to continue investing in CSR initiatives in the future as they see the positive impact on their stock prices.

5.2. Robustness

To test robustness, first, the propensity score matching (PSM) model is adopted to mitigate the endogeneity issue that firms included in the Shanghai-Hong Kong or Shenzhen-Hong Kong Stock Connect list were not randomly selected. Following [2]; a one-on-one nearest matching without placement was carried out based on the following matching variables: SIZE, ROA, GROWTH, DIV (dividend payout ratio), AGE (natural logarithm of the number of listed years), TURNOVER, BOARD, BIG4 (big four auditing dummy) and DUAL (=1 if the board chairman concurrently serves as the general manager; otherwise 0). Panel A of Table 3 demonstrates that the differences in the means for the covariates between the treatment group and the matched control sample are nonsignificant except for the firm size and size of the board, suggesting a successful matching process. According to Panel B of Table 3, the coefficients on SC*CSR remain significantly positive after re-estimating the baseline regression using the PSM sample.

Table 3.

Propensity score matching.

Panel A Covariate balance between the matched pairs
VARIABLES	Treated (1)	Control (2)	Difference (1)–(2)	t-statistics
SIZE	23.161	23.234	−0.073	−2.52
ROA	0.062	0.061	0.001	0.67
GROWTH	0.391	0.401	−0.010	−0.45
DIV	0.299	0.300	−0.001	−0.07
AGE	2.469	2.453	0.016	1.32
TURNOVER	4.952	5.012	−0.060	−0.78
BOARD	2.163	2.173	−0.010	−2.26
BIG4	0.126	0.135	−0.009	−1.61
DUAL	0.228	0.218	0.010	1.27

Panel B Results based on the PSM sample
VARIABLES	(1)	(2)
	CAR	Q
CSR	0.000	0.002**
	(1.025)	(2.418)
SC*CSR	0.002**	0.003**
	(1.975)	(2.076)
SC	−0.005	0.016
	(-0.988)	(0.410)
SIZE	0.001	−0.324***
	(0.717)	(-12.036)
LEV	0.007	0.441***
	(0.829)	(4.015)
ROA	0.026	0.244*
	(0.997)	(1.720)
TURNOVER	−0.001	−0.008
	(-0.844)	(-0.635)
EPS	0.004*
	(1.852)
CASH	−0.001
	(-0.079)
VOL	0.020
	(0.345)
BM	−0.001
	(-0.248)
GROWTH		−0.008
		(-1.487)
TANG		0.350***
		(2.998)
TOP1		0.091
		(0.766)
BOARD		−0.180**
		(-2.187)
Constant	−0.025	8.886***
	(-0.843)	(16.259)
FIRM&YEAR	YES	YES
Observations	4,072	12,896
R-squared	0.018	0.196

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

Second, we alter the method of measuring the key variables. Following [52]; we calculate the industry-adjusted CSR score, ADJCSR, using Equation (3). CSRMIN and CSRMAX are the minimum and maximum values of CSR within the industry in a given year, respectively.

$$\mathrm{ADJCSR}=(\mathrm{CSR}-\mathrm{CSRMIN})/(\mathrm{CSRMAX}\mathit{-}\mathrm{CSRMIN})$$ (3)

In addition, we use the market model instead of the market-adjusted model to calculate the cumulative abnormal return and denote it as CAR1. The parameters are estimated between trading days (−300, 50) before the announcement (t = 0) of CSR reports.

The results in Columns 1–3 of Table 4 show that the conclusion still holds after altering the method of measuring the CSR score and the cumulative abnormal return.

Table 4.

Altering the method of measuring key variables and delete the COVID-19 samples.

VARIABLES	(1)	(2)	(3)	(4)	(5)
	CAR	Q	CAR1	CAR	Q
ADJCSR	0.005	−0.046
	(1.156)	(-1.105)
SC*ADJCSR	0.018**	0.422***
	(2.403)	(6.447)
CSR			0.000	0.000*	−0.001*
			(1.267)	(1.802)	(-1.856)
SC*CSR			0.001***	0.000*	0.005***
			(2.762)	(1.877)	(4.490)
SC	−0.013**	−0.152***	−0.009	−0.008*	−0.011
	(-2.523)	(-3.688)	(-1.634)	(-1.753)	(-0.266)
SIZE	0.001	−0.347***	−0.000	0.001	−0.150***
	(1.363)	(-10.909)	(-0.200)	(1.195)	(-4.366)
LEV	0.007	0.560***	0.017**	0.004	0.281**
	(1.061)	(4.738)	(2.282)	(0.584)	(2.148)
ROA	0.031	0.346*	0.051	0.031	0.289
	(1.094)	(1.678)	(1.534)	(1.024)	(1.251)
TURNOVER	0.001	−0.016	0.004***	−0.001	0.010
	(0.611)	(-1.156)	(2.971)	(-0.674)	(1.344)
EPS	0.003*		0.004**	0.004*
	(1.876)		(1.989)	(1.905)
CASH	−0.001		−0.009	−0.009
	(-0.145)		(-0.711)	(-0.910)
VOL	0.011		−0.108*	−0.002
	(0.221)		(-1.952)	(-0.031)
BM	−0.002		0.004	−0.007
	(-0.422)		(0.625)	(-1.141)
GROWTH		−0.007			−0.011
		(-1.056)			(-1.375)
TANG		0.337**			0.707***
		(2.192)			(3.951)
TOP1		0.015			0.135
		(0.105)			(0.707)
BOARD		−0.314***			−0.378***
		(-3.704)			(-3.741)
Constant	−0.042*	9.782***	−0.016	−0.028	6.017***
	(-1.903)	(13.796)	(-0.623)	(-1.203)	(7.873)
FIRM&YEAR	YES	YES	YES	YES	YES
Observations	5,622	23,238	5,545	4823	20176
R-squared	0.017	0.221	0.016	0.015	0.235

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

Third, we delete the samples during the COVID-19 epidemic. Our research sample includes the years 2020 and 2021, which corresponds to the period of the COVID-19 pandemic. This could potentially impact our research findings. The COVID-19 pandemic has had a significant global impact on the economy and financial markets, which might have influenced investors' acceptance of corporate social responsibility (CSR) activities. For example, due to the challenges faced by businesses during the pandemic and changes in social needs, investors may have increased their focus on CSR performance. Given this, although we have used the difference-in-differences models that include the year effect to control for unobservable time-varying characteristics or aggregate fluctuations, as a further robustness test, we still delete the samples during the COVID-19 epidemic and test the main regression again as a further robustness test. As presented in Columns 4 and 5 of Table 4, the results have not changed substantially from the original manuscript.

Fourth, we conduct the parallel trend test to ensure the validity of the difference-in-differences analysis. Following [53]; we replace SC with the following variables: SC⁻², SC⁻¹, SC⁰, SC¹, and SC²⁺. These variables take the value of one if the firm is included in (1) the Shanghai-Hong Kong or Shenzhen-Hong Kong Stock Connect list two or more years later, (2) the Stock Connect list in the next year, (3) the Stock Connect list in the current year, (4) the Stock Connect list one year ago, or (5) the Stock Connect list two or more years ago. As shown in Table 5, the coefficients of CSR*SC⁻² and CSR*SC⁻¹ are nonsignificant, whereas those of CSR*SC⁰, CSR*SC¹, and CSR*SC²⁺ are mostly significant. These results demonstrate that the baseline results satisfy the parallel trend test.

Table 5.

Parallel trend test.

VARIABLES	(1)	(2)
	CAR	Q
CSR	0.000	0.000
	(1.456)	(0.123)
SC−2	0.004	0.000
	(0.318)	(0.119)
SC−1	0.006	0.095**
	(0.462)	(1.975)
SC0	0.010	0.235***
	(0.906)	(4.252)
SC1	0.011	−0.053
	(1.047)	(-1.058)
SC2+	0.006	−0.281***
	(0.598)	(-5.527)
CSR*SC−2	−0.000	−0.002
	(-0.432)	(-1.210)
CSR*SC−1	−0.000	0.001
	(-0.388)	(0.740)
CSR*SC0	0.000	0.001*
	(0.187)	(1.672)
CSR*SC1	0.001**	0.002*
	(2.109)	(1.922)
CSR*SC2+	0.001**	0.003**
	(2.225)	(2.003)
SIZE	0.002*	−0.293***
	(1.662)	(-15.106)
LEV	0.007	0.523***
	(0.098)	(7.033)
ROA	0.039	0.485***
	(1.521)	(3.572)
TURNOVER	0.003***	0.010**
	(3.864)	(1.995)
EPS	0.003*
	(1.786)
CASH	−0.001
	(-0.158)
VOL	−0.077
	(-1.654)
BM	−0.004
	(-0.651)
GROWTH		−0.007
		(-1.310)
TANG		0.380***
		(4.092)
TOP1		0.016
		(0.165)
BOARD		−0.336***
		(-5.070)
Constant	−0.056**	8.406***
	(-2.319)	(18.642)
FIRM&YEAR	YES	YES
Observations	5,639	25,226
R-squared	0.018	0.225

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

5.3. Heterogeneity

In this section, we explore cross-sectional variations in the effect of market liberalization on the stock market performance related to CSR engagement and test hypotheses 2, 3, and 4 by focusing on the following aspects of this heterogeneity: the role of agency costs, the status of SOE, and the possibility of impression management.

5.3.1. Agency costs

Overseas investors possess the ability to effectively analyze information and accurately understand the motivations and economic ramifications of pursuing corporate social responsibility (CSR) in companies with high agency costs. This understanding enables them to recognize that the fulfillment of CSR by such companies is driven solely by management's self-interest and may not contribute value to the organization. Consequently, investment decisions made by overseas investors tend to favor companies with lower agency costs that have a strong CSR rating. In light of this, we anticipate that the influence of Stock Connect on stock market performance related to CSR will likely be more pronounced for companies with lower agency costs than those with higher agency costs.

To test hypothesis 2, following [54]; we measure agency costs as the ratio of operating expenses to sales revenue. We use the annual median of this indicator to split the sample and define company-year samples to have high (low) agency costs if the ratio of operating expense to sales revenue is higher than (equal to or lower than) the annual median. As shown in Table 6, the coefficients of the interaction term, SC*CSR, are significantly positive in Columns 1 and 3 (the group with low agency costs) but not significant in Columns 2 and 4 (the group with high agency costs). These results suggest a more significant positive effect of Stock Connect on the stock market performance related to CSR for firms with lower agency costs.

Table 6.

Heterogeneity: the role of agency costs.

VARIABLES	(1)	(2)	(3)	(4)
	Low agency costs	High agency costs	Low agency costs	High agency costs
	CAR	CAR	Q	Q
CSR	0.000*	0.000	−0.001	0.002***
	(1.800)	(0.525)	(-0.912)	(3.002)
SC*CSR	0.000**	0.000	0.006***	−0.001
	(1.993)	(0.830)	(2.831)	(-1.292)
SC	−0.011*	−0.004	−0.067	0.076**
	(-1.656)	(-0.637)	(-1.011)	(2.064)
SIZE	0.002	0.001	−0.382***	−0.227***
	(1.092)	(1.046)	(-7.427)	(−5.325)
LEV	0.016*	−0.009	0.671***	0.405***
	(1.697)	(-1.031)	(3.668)	(3.503)
ROA	0.045	−0.002	0.144	1.013***
	(1.237)	(-0.050)	(0.495)	(3.402)
TURNOVER	0.004**	0.003**	−0.003	0.013*
	(2.465)	(2.073)	(−0.265)	(1.672)
EPS	0.004	0.004*
	(1.242)	(1.859)
CASH	0.002	−0.002
	(0.146)	(-0.129)
VOL	−0.153**	0.005
	(-2.287)	(0.100)
BM	−0.021**	0.006
	(-2.548)	(0.762)
GROWTH			−0.018	0.002
			(-1.412)	(0.382)
TANG			0.492*	0.181
			(1.934)	(1.097)
TOP1			0.120	−0.048
			(0.391)	(−0.315)
BOARD			−0.448***	−0.191**
			(-3.006)	(-2.357)
Constant	−0.050	−0.034	10.752***	6.603***
	(-1.360)	(-1.217)	(9.582)	(7.072)
FIRM&YEAR	YES	YES	YES	YES
Observations	2,829	2,761	12,721	12,438
R-squared	0.031	0.010	0.260	0.198

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

5.3.2. Status of SOE

The acknowledgment of SOEs' CSR performance by foreign investors introduced via the Stock Connect program may be relatively lower than that of nonstate-owned enterprises (non-SOEs). This could be due to a mismatch between the values of overseas investors, who often have a stronger social consciousness, and the CSR initiatives undertaken by SOEs. Additionally, foreign investors possess sophisticated information analysis and processing capabilities, enabling them to discern that CSR engagement by SOEs for political purposes may not necessarily enhance their economic value.

To test hypothesis 3, we partition the full sample into SOEs and non-SOEs. The results in Table 7 report significantly positive coefficients of the interaction term, SC*CSR, for nonstate-owned enterprises in Columns 2 and 4 but not significant coefficients of SC*CSR for state-owned enterprises in Columns 1 and 3. These findings demonstrate that the impact of Stock Connect on the stock market performance related to CSR is stronger for nonstate-owned enterprises than for state-owned enterprises.

Table 7.

Heterogeneity: SOE versus. non-SOE.

VARIABLES	(1)	(2)	(3)	(4)
	SOE	Non-SOE	SOE	Non-SOE
	CAR	CAR	Q	Q
CSR	0.000**	0.000	0.001	−0.001
	(1.989)	(0.832)	(0.653)	(-0.851)
SC*CSR	−0.000	0.001***	0.001	0.004**
	(-0.857)	(3.689)	(1.033)	(1.989)
SC	0.004	−0.024***	0.038	0.004
	(0.776)	(-3.131)	(0.921)	(0.083)
SIZE	0.002*	0.001	−0.303***	−0.309***
	(1.876)	(0.505)	(-6.079)	(-6.982)
LEV	−0.005	0.015	0.327**	0.585***
	(-0.626)	(1.455)	(2.249)	(3.798)
ROA	0.007	0.058	0.309	0.394
	(0.208)	(1.264)	(0.957)	(1.464)
TURNOVER	0.005***	0.001	0.035***	−0.001
	(3.754)	(0.872)	(2.588)	(-0.115)
EPS	0.006***	−0.000
	(2.766)	(-0.093)
CASH	−0.019	0.012
	(-1.550)	(0.854)
VOL	−0.001	−0.107
	(-0.028)	(-1.480)
BM	0.007	−0.012
	(0.993)	(-1.374)
GROWTH			−0.012*	−0.004
			(-1.693)	(-0.397)
TANG			0.150	0.508**
			(0.752)	(2.553)
TOP1			0.077	0.051
			(0.369)	(0.221)
BOARD			−0.090	−0.454***
			(-0.719)	(-3.850)
Constant	−0.068***	−0.024	8.233***	9.067***
	(-2.727)	(-0.548)	(7.470)	(9.565)
FIRM&YEAR	YES	YES	YES	YES
Observations	922	4,601	2,813	21,899
R-squared	0.020	0.024	0.184	0.247

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

5.3.3. The possibility of impression management

Overseas investors introduced via the Stock Connect program may have a higher ability to detect impression management compared to local individual investors, due to their stronger information analysis and processing capabilities. If a company is more likely to engage in impression management through its CSR report, even if the report appears to demonstrate a high level of CSR fulfillment, overseas investors may not place a higher valuation on the company. This is because they may be able to discern the authenticity of the information presented in the report and factor in any potential negative implications of impression management on the company's true social responsibility performance.

To test hypothesis 4, we run tests considering the possibility of CSR reports being used as a tool of impression management. We can indirectly infer the likelihood of companies using social responsibility reports for impression management from the quality of information disclosure. Every year, the Shanghai Stock Exchange and the Shenzhen Stock Exchange release quality ratings of the information disclosure of Chinese-listed companies. The rating reflects the authenticity, completeness, reliability, and timeliness of the financial and nonfinancial information disclosed by a listed company. A higher rating signifies better information disclosure quality and a reduced likelihood of firm managers resorting to CSR reports for impression management [55].

We assign values of 4, 3, 2, and 1 to ratings A, B, C, and D (i.e., excellent, good, pass, and fail) released by the stock exchange, respectively, with higher values indicating a higher quality of corporate information disclosure. We use the annual median of this rating score to split the sample and define firm-year observations as having a high (low) quality of information disclosure if their value is higher than or equal to (lower than) the annual median. As shown in Table 8, the coefficients of the interaction term, SC*CSR, are significantly positive for the group that has a higher quality of information disclosure (Columns 2 and 4) and therefore lower possibility of using CSR report for impression management, but not significant coefficients for the group with lower quality of information disclosure (Columns 1 and 3) and higher possibility of using CSR report for impression management. Thus, our results suggest that the influence of Stock Connect on the stock market performance related to CSR is more pronounced for firms that are less likely to use CSR reports to manage impressions.

Table 8.

Heterogeneity: the possibility of impression management.

VARIABLES	(1)	(2)	(3)	(4)
	Low quality	High quality	Low quality	High quality
	CAR	CAR	Q	Q
CSR	0.000	0.000*	0.002**	−0.000
	(0.893)	(1.934)	(2.153)	(-0.285)
SC*CSR	0.000	0.001***	−0.002	0.010***
	(1.115)	(3.247)	(-1.304)	(3.623)
SC	−0.009	−0.018**	0.012	−0.152**
	(-1.458)	(-2.281)	(0.204)	(-2.196)
SIZE	0.003*	0.001	−0.425***	−0.236***
	(1.787)	(0.438)	(-8.069)	(-6.323)
LEV	−0.001	0.014	0.693***	0.314**
	(-0.060)	(1.606)	(3.587)	(2.328)
ROA	0.081*	−0.018	0.402	0.417
	(1.783)	(-0.489)	(1.208)	(1.353)
TURNOVER	0.004**	0.003*	0.005	0.007
	(2.478)	(1.921)	(0.379)	(0.983)
EPS	0.001	0.005**
	(0.213)	(2.029)
CASH	−0.012	0.007
	(-0.863)	(0.498)
VOL	0.104*	−0.186***
	(1.789)	(-2.835)
BM	0.016*	−0.020***
	(1.940)	(-2.733)
GROWTH			−0.005	−0.004
			(-0.507)	(-0.501)
TANG			0.115	0.896***
			(0.489)	(4.488)
TOP1			−0.046	0.161
			(-0.172)	(0.752)
BOARD			−0.404***	−0.244**
			(-3.059)	(-2.314)
Constant	−0.087***	−0.009	11.599***	6.971***
	(-2.793)	(-0.288)	(9.927)	(8.837)
FIRM&YEAR	YES	YES	YES	YES
Observations	532	2088	5269	11138
R-squared	0.018	0.024	0.187	0.278

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

5.4. Additional tests

5.4.1. QFII and the impact of Stock Connect on the stock market performance related to CSR

In 2002, Chinese regulators initiated the QFII (Qualified Foreign Institutional Investors) scheme to accelerate the opening of stock markets. Companies might attract foreign investors through the QFII scheme [56]. documented evidence that foreign investors through QFII are more likely to demand CSR activities. In our study, we posit that firms with more foreign investors have superior stock market performance in terms of their CSR engagement. If this is true, we anticipate that foreign investors participating in the SHSH Stock Connect program will have a more substantial impact on firms with limited international engagement through QFII ownership. In other words, we expect that the enhancement of CSR-related stock market performance will be more significant for companies without QFII ownership.

To test the above prediction, we re-estimate separately our baseline regression Equations (1), (2) for firm-year samples with and without QFII shareholding. As presented in Table 9, the coefficients on SC*CSR are only significant for samples without QFII shareholding (Columns 1 and 3). These findings indicate that improvements in the short- and long-term stock market performance related to CSR engagements brought by the SHSH Stock Connect are more pronounced for firms without QFII shareholdings.

Table 9.

QFII, SHSH Stock Connect and the stock market performance related to CSR.

VARIABLES	(1)	(2)	(3)	(4)
	Without QFII ownership	With QFII ownership	Without QFII ownership	With QFII ownership
	CAR	CAR	Q	Q
CSR	0.000*	0.000	−0.000	0.006***
	(1.809)	(0.334)	(-0.118)	(3.069)
SC*CSR	0.000*	0.000	0.003*	0.000
	(1.947)	(1.068)	(1.773)	(0.791)
SC	−0.006	−0.014	0.090**	0.051
	(-1.098)	(-1.373)	(2.354)	(0.470)
SIZE	0.002	0.001	−0.460***	−0.237***
	(1.558)	(0.643)	(-20.676)	(-3.224)
LEV	0.010	−0.005	0.413***	0.737**
	(1.337)	(-0.378)	(3.803)	(2.126)
ROA	0.042	0.044	0.408*	1.395**
	(1.182)	(0.897)	(1.869)	(2.122)
TURNOVER	0.003***	0.004	−0.014**	0.060***
	(2.606)	(1.463)	(-2.034)	(3.133)
EPS	0.000	0.008**
	(0.040)	(2.525)
CASH	−0.003	0.005
	(-0.305)	(0.202)
VOL	−0.029	−0.170**
	(-0.597)	(-2.011)
BM	−0.002	−0.002
	(-0.351)	(-0.200)
GROWTH			0.000	−0.034
			(0.065)	(-1.453)
TANG			−0.108	0.593
			(-0.978)	(1.602)
TOP1			−0.118	0.087
			(-1.072)	(0.179)
BOARD			−0.282***	−0.312
			(-3.901)	(-1.216)
Constant	−0.056**	−0.025	12.218***	6.868***
	(-2.254)	(-0.601)	(25.875)	(4.704)
FIRM&YEAR	YES	YES	YES	YES
Observations	4,342	1,297	20,250	4,976
R-squared	0.011	0.031	0.247	0.137

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

5.4.2. The impact of SHSH ownership

In this section, we examine the effect of foreign ownership through the SHSH Stock Connect on the stock market performance related to CSR. To gauge SHSH ownership, we define two variables as follows. The first is the shareholding ratio of SHSH shareholders, SC_RATIO. The second is a dummy variable, SC_ABOVE, that takes the value of 1 if the SHSH shareholding is above or equal to the tenth largest shareholder of the firm in a given year and 0 otherwise, following [16]. These two variables then interact with CSR, and the interaction terms SC_RATIO*CSR and SC_ABOVE*CSR are used to replace the key explanatory variable, SC*CSR, in the baseline regression Equation (1). We also include the industry (IND) and year (YEAR) effects in the regression. Columns 1–4 of Table 10 report significantly positive coefficients on SC_RATIO*CSR and SC_ABOVE*CSR. As such, the short- and long-term stock market performance related to CSR engagements is superior for firms with significant foreign ownership through the SHSH Stock Connect.

Table 10.

SHSH ownership and stock market performance related to CSR.

VARIABLES	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)
	CAR	CAR	Q	Q	CAR	CAR	Q	Q
CSR	0.000***	0.000***	−0.001	−0.005***	0.000***	0.000**	−0.002	−0.006***
	(2.978)	(2.960)	(-0.732)	(-2.719)	(2.739)	(2.235)	(-0.889)	(-3.340)
SC_RATIO*CSR	0.040**		0.561**		0.040**		0.511*
	(2.298)		(2.140)		(2.308)		(1.940)
SC_RATIO	−0.989*		8.153		−0.993*		9.397
	(-1.946)		(1.156)		(-1.951)		(1.330)
SC_ABOVE*CSR		0.002***		0.037***		0.002***		0.033***
		(4.531)		(8.511)		(4.574)		(7.576)
SC_ABOVE		−0.037***		−0.106		−0.037***		−0.063
		(-3.640)		(-1.026)		(-3.632)		(-0.609)
QFII_RATIO*CSR					0.000		−0.016
					(1.163)		(-1.175)
QFII_RATIO					−0.001		0.004**
					(-0.406)		(2.116)
QFII_ABOVE*CSR						0.000**		0.028***
						(2.158)		(3.294)
QFII_ABOVE						−0.008		−0.286
						(-1.545)		(-1.471)
SIZE	0.001	0.000	−0.555***	−0.566***	0.000	0.000	−0.555***	−0.568***
	(0.618)	(0.408)	(-25.553)	(-25.568)	(0.508)	(0.237)	(-25.555)	(-25.700)
LEV	0.007	0.008	0.808***	0.787***	0.007	0.008	0.811***	0.788***
	(1.117)	(1.234)	(7.536)	(7.319)	(1.120)	(1.343)	(7.559)	(7.327)
ROA	0.030	0.024	1.541***	1.695***	0.033	0.024	1.537***	1.623***
	(1.043)	(0.869)	(5.180)	(5.711)	(1.143)	(0.849)	(5.165)	(5.490)
TURNOVER	−0.000	−0.000	−0.160***	−0.177***	−0.000	−0.000	−0.160***	−0.171***
	(-0.406)	(-0.050)	(-9.139)	(-9.908)	(-0.440)	(-0.038)	(-9.145)	(-9.628)
EPS	0.003	0.003			0.003	0.003
	(1.577)	(1.515)			(1.372)	(1.499)
CASH	−0.003	−0.003			−0.003	−0.003
	(-0.279)	(-0.266)			(-0.323)	(-0.314)
VOL	0.021	0.016			0.024	0.019
	(0.447)	(0.333)			(0.500)	(0.405)
BM	0.000	−0.000			0.000	0.000
	(0.003)	(-0.026)			(0.086)	(0.011)
GROWTH			0.006	0.005			0.006	0.007
			(0.488)	(0.439)			(0.486)	(0.523)
TANG			−0.323***	−0.354***			−0.323***	−0.361***
			(-3.014)	(-3.289)			(-3.017)	(-3.368)
TOP1			0.142*	−0.083			0.147*	−0.085
			(1.682)	(-0.965)			(1.737)	(-0.986)
BOARD			−0.126*	−0.151**			−0.126*	−0.143**
			(-1.837)	(-2.175)			(-1.836)	(-2.063)
Constant	−0.023	−0.021	15.219***	15.740***	−0.021	−0.017	15.235***	15.752***
	(-1.091)	(-0.979)	(27.064)	(27.618)	(-0.986)	(-0.800)	(27.076)	(27.639)
IND&YEAR	YES	YES	YES	YES	YES	YES	YES	YES
Observations	5,639	5,639	25,226	25,226	5,639	5,639	25,226	25,226
R-squared	0.014	0.018	0.321	0.311	0.015	0.020	0.322	0.316

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

As previously mentioned, to appeal to foreign investors, the Chinese government implemented the QFII mechanism. Nevertheless, institutional investors are required to undergo a cumbersome, protracted, and time-intensive application process to obtain a QFII license [35]. Additionally, the QFII mechanism has strict quotas, and its approvals are subject to government discretion. Furthermore, it should be noted that QFII is a unilateral mechanism, while the SHSH Stock Connect is a bilateral liberalization mechanism. With these factors in mind, it is important to note that QFII provides limited entry to China's stock market compared to the SHSH Stock Connect, which is a more liberal and bilateral mechanism. In light of this, we are curious to explore the effect of SHSH ownership on the stock market performance related to CSR in comparison to QFII shareholding. Similarly, we define two variables to measure QFII ownership as follows. The first is the shareholding ratio of QFII shareholders, QFII_RATIO. The second is a dummy variable, QFII_ABOVE, taking the value of 1 if the QFII shareholding is above or equal to the tenth largest shareholder of the firm in a given year and 0 otherwise, following [16]. We then include the interaction terms QFII_RATIO*CSR and QFII_ABOVE*CSR in the regression. Columns 5–8 of Table 10 reveal that the coefficients on the interaction terms SC_RATIO*CSR and SC_ABOVE*CSR remain significantly positive after controlling for the interaction terms between CSR and QFII shareholding. These results suggest that the impact of significant foreign ownership through the SHSH Stock Connect on the stock market performance related to CSR engagement is incremental over that of QFII ownership.

5.4.3. Voluntary disclosure versus mandatory disclosure

As previously mentioned [11], found no significant relationship between CSR scores and market reactions to CSR reports. This was true for both mandatory and voluntary CSR disclosures, providing further support for their argument that Chinese investors are insensitive to CSR engagements. However, in earlier sections of our study, we demonstrated that the SHSH Stock Connect could have significant implications for the stock market performance related to CSR engagements. In this section, we aim to investigate whether these improvements are different for voluntary and mandatory CSR disclosures.

We re-examine separately our baseline regression Equations (1), (2) for voluntary and mandatory disclosures. Our findings show that the coefficient on SC*CSR is significantly positive for the voluntary disclosure group (Columns 2 and 4 of Table 11) but not significant for the mandatory disclosure group (Columns 1 and 3 of Table 11). This suggests that the improvements in stock market performance brought by Stock Connect only apply to voluntary CSR disclosures. Foreign investors perceive mandatory CSR disclosure as only fulfilling regulatory obligations; thus, they value only CSR commitments and achievements communicated by voluntary CSR reports.

Table 11.

Voluntary disclosure versus mandatory disclosure.

VARIABLES	(1)	(2)	(3)	(4)
	Mandatory ownership	Voluntary disclosure	Mandatory disclosure	Voluntary disclosure
	CAR	CAR	Q	Q
CSR	0.000	0.000	0.001	−0.002***
	(0.793)	(1.533)	(0.633)	(-2.697)
SC*CSR	−0.000	0.001***	−0.001	0.007***
	(-0.002)	(2.859)	(-0.753)	(3.712)
SC	0.001	−0.021***	0.142**	−0.084*
	(0.220)	(-2.712)	(2.262)	(-1.705)
SIZE	0.001	0.004*	−0.320***	−0.350***
	(0.398)	(1.664)	(-4.313)	(-9.913)
LEV	0.006	0.005	0.191	0.526***
	(0.737)	(0.494)	(0.804)	(4.200)
ROA	0.093**	−0.011	0.084	0.338
	(2.405)	(-0.248)	(0.151)	(1.523)
TURNOVER	0.003**	0.003**	0.069***	0.003
	(2.081)	(2.431)	(3.927)	(0.468)
EPS	0.002	0.005
	(1.099)	(1.495)
CASH	−0.019	0.011
	(-1.354)	(0.899)
VOL	−0.019	−0.088
	(-0.343)	(-1.343)
BM	0.011	−0.016*
	(1.572)	(-1.681)
GROWTH			0.007	−0.009
			(0.479)	(-1.246)
TANG			0.045	0.417**
			(0.239)	(2.435)
TOP1			−0.189	0.151
			(-0.669)	(0.855)
BOARD			−0.015	−0.393***
			(-0.103)	(-3.956)
Constant	−0.033	−0.084*	8.922***	9.797***
	(-1.147)	(-1.851)	(5.129)	(12.863)
FIRM&YEAR	YES	YES	YES	YES
Observations	3,098	2,539	4,101	20,917
R-squared	0.013	0.020	0.167	0.242

Note: *,** and*** indicate statistical significance at 10 %, 5 % and 1 %, respectively. The robust t-statistics are reported in parentheses.

6. Conclusion

We explore the impact of capital market openings on the extent of CSR acceptance by investors. We find robust evidence that due to the increase in investors' awareness of CSR and their ability to process and analyze information, the positive relationship between CSR scores and the market's reaction to CSR reports becomes more significant after market liberalization in China. In addition, firms with higher CSR scores enjoy better long-term stock market performance after market liberalization. The positive influence of the SHSH Stock Connect on the stock market performance related to CSR activities is more significant for non-SOE firms, firms with lower agency costs, and firms with a lower possibility of using CSR reports for impression management. Additional research indicates that the benefits of the program are more prominent for companies that do not have QFII shareholdings. Moreover, the impact of significant foreign ownership through Stock Connect on the stock market performance related to CSR engagement is incremental over QFII ownership.

Given the growing concerns over CSR engagement, our study has practical implications for corporate managers and regulators. First, our empirical results suggest that China's market liberalization improves the stock market performance of companies with strong CSR records, as it results in higher Tobin's Q and more favorable stock price reactions to CSR reports. The improvement in market performance will in turn encourage firms to actively engage in CSR. Thus, the implementation of Stock Connect starts a virtuous circle and helps managers balance the objectives of shareholders with those of a wider set of stakeholders.

Second, our results provide theoretical support for the continuous opening of China's capital market. A follow-on opening is the Shanghai-London Stock Connect, which will further attract foreign investors. Regulators could also consider Stock Connects with other countries/regions. Our findings offer pivotal insights for comprehending the debate around market liberalization. While the advantages of an increase in foreign investor access are widely acknowledged, the drawbacks of financial liberalization must not be ignored [57]. Amidst the ongoing argument over the significance of market liberalization, our research presents fresh evidence that supports the adoption of higher market openness. Our study highlights the constructive influence of market liberalization on the stock market performance related to CSR engagement, providing valuable guidance for policy-makers in emerging economies.

The present study contributes to the growing body of literature examining the relationship between corporate social responsibility (CSR) and stock market performance and the consequence of market liberalization, particularly in the context of emerging markets. However, several avenues for future research can be identified. Firstly, while this study focuses on the impact of Stock Connect on investor acceptance of CSR in Chinese-listed companies, it would be interesting to investigate the effect of similar market liberalization measures in other emerging markets. For example, future research can be conducted based on cross-national data to explore whether the positive impact of Stock Connect on investor acceptance of CSR is evident in most emerging countries or only in a specific type of country. Secondly, this study examines the impact of Stock Connect on the short- and long-term stock market performance of companies with strong overall CSR records. Future research can explore the impact of Stock Connect on investor acceptance of specific CSR practices, as well as the potential moderating role of industry-specific factors.

Funding

This work is supported by the Humanities and Social Sciences Youth Foundation, Ministry of Education of the People's Republic of China (No.22YJC630221) and the Special Innovative Projects of Universities in Guangdong Province (No.2022WTSCX179).

Data availability statement

The authors do not have permission to share data.

CRediT authorship contribution statement

Hui Zhou: Conceptualization, Formal analysis, Methodology, Writing – original draft, Writing – review & editing. Lin Xu: Conceptualization, Writing – original draft.

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.

Appendix.

A.

Appendix A. Definitions of the main variables

Variables	Definitions
CAR	Cumulative abnormal return calculated using the market-adjusted model over the three-day interval around the announcement of CSR reports.
Q	Tobin's Q, measured as the market value of assets to the book value of assets.
CSR	Overall CSR score from Hexun. A higher score means better CSR performance.
SC	Dummy = 1 if the firm is eligible for trading under either the Shanghai-Hong Kong or Shenzhen-Hong Kong Stock Connect scheme in that year, otherwise = 0.
SIZE	Firm's total assets, measured as the natural logarithm of total assets of the firm.
LEV	Debt ratio, measured as the ratio of total debt to total assets.
ROA	Return on total assets, measured as the ratio of net income to total assets.
TURN	Stock turnover, measured as the average daily turnover ratio during the year.
EPS	Earnings per share, measured as the ratio of net income to the total number of shares.
CASH	Free cash flow, measured as the ratio of free cash flow to total assets.
VOL	Volatility of stock return, measured as the standard deviation of weekly stock returns during the year.
BM	Book-to-market ratio, measured as the ratio of the book value of equity to the market value of equity.
GROWTH	Sales growth, measured as the growth rate of sales revenue.
TANG	Proportion of tangible assets, measured as the ratio of tangible assets to total assets.
TOP1	Controller shareholding, measured as the fraction of shares held by the controlling shareholder.
BOARD	Size of the board, measured as the natural logarithm of the number of directors on the board.