肖忠意等：机构投资者的信息获取如何抑制股权质押?——来自中国的证据-文章推荐-西南政法大学.pdf

International Review of Financial Analysis 89 (2023) 102746 Contents lists available at ScienceDirect International Review of Financial Analysis journal homepage: www.elsevier.com/locate/irfa How does institutional investors’ information acquisition inhibit share pledging? Evidence from China Zhongyi Xiao a, b, Haitao Chen c, Kang Chen d, * a Business School, Southwest University of Political Science & Law, No. 301 Baosheng Avenue, Yubei District, Chongqing, China Research Center for Audit and Rule of Law, Southwest University of Political Science & Law, No. 301 Baosheng Avenue, Yubei District, Chongqing, China c School of Economics, Southwest University of Political Science & Law, No. 301 Baosheng Avenue, Yubei District, Chongqing, China d Institute of Chinese Financial Studies, Southwestern University of Finance and Economics, No. 555 Liutai Avenue, Wenjiang District, Chengdu City, Sichuan Province, China b A R T I C L E I N F O A B S T R A C T JEL classifications: G23 G32 This study investigates whether and how institutional investors’ information acquisition affects controlling shareholder’s share pledging. Taking a unique data of institutional investors’ corporate site visits in China, we find that institutional investors’ corporate site visits significantly inhibit controlling shareholder’s share pledging. This effect is robust to a series of robustness checks, including controlling for endogeneity concerns, propensity score matching method, alternative model specifications, and alternative measures of the indepen­ dent variable. We then provide evidence that the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced for listed firms with less efficient in­ formation environment and weaker corporate governance. Further analysis indicates that the negative relation is also more pronounced when controlling shareholders are under higher margin call pressure and when the visiting institutional investors consist of more fund companies. Overall, our study is the first to provide direct evidence of the governance mechanism of financial intermediaries on shareholders’ pledging decisions. Keywords: Corporate site visits Share pledging Information environment Corporate governance 1. Introduction Share pledging is the practice that shareholders use their share­ holdings as collaterals to obtain external financing. Share pledging has expanded over the last few years and has become one common financing channel in both developed markets and emerging markets. In the US, over 17% of S&P 1500 firms have pledged shares (Anderson & Puleo, 2020); in India, at least one third of controlling shareholders in listed firms have pledged shares (Singh, 2018). Under a developing financial system in China, where enterprises heavily rely on indirect financing from the banking system, share pledging, a unique financing vehicle at the intersection of the banking system and the stock market (He, Liu, & Zhu, 2022), has become prevailing among shareholders and has raised the attention from both regulatory authorities and the academia. Extant studies majorly focus on the share pledging by controlling shareholders and it is documented to be associated with agency problems and opportunistic behaviors, which might induce earning management (Deren and Ke, 2018), erosion of corporate value (Dou, Masulis, & Zein, 2019) and corporate fraud (Kryzanowski, Li, Xu, et al., 2021). Hence, a comprehensive analysis about the determinants underlying controlling shareholder’s share pledging decisions, especially from the perspective of corporate governance, might help alleviate pledging-related agency problems and provide practical implications to protect medium and small shareholder’s benefits. However, the determinants of controlling shareholder’s share pledging are still under-explored. Literature suggests that financial institutions, as one of the typical channels of corporate governance, are important contributors to infor­ mation production (Boone & White, 2015), investor relationships management (Su, Feng, & Tang, 2021) and shareholders’ wealth pro­ tection (Cheng, Du, Wang, et al., 2019). However, Pound (1988) argues that institutional investors might also exert negative impacts on corpo­ rate governance by conspiring with managers. It might require further investigations and empirical evidences to verify the governance mech­ anisms of financial institutions on firm decision making. On July 17 Hobson, Mayew, & Venkatachalam, 2012, Shenzhen Stock Exchange (SZSE) mandated all its listed firms to disclose standard and detailed information about corporate site visits, which provided a unique data base facilitating researches on the influence of institutional investors’ * Corresponding author. E-mail address: kangchen@swufe.edu.cn (K. Chen). https://doi.org/10.1016/j.irfa.2023.102746 Received 20 October 2022; Received in revised form 7 March 2023; Accepted 27 June 2023 Available online 28 June 2023 1057-5219/© 2023 Elsevier Inc. All rights reserved. Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 on-site visits to firm headquarters. In this study, we are the first to take the perspective of corporate site visits, a specific approach of informa­ tion acquisition and corporate governance of institutional investors, to investigate their roles in protecting medium and small shareholders’ benefits by affecting the share pledging decisions of controlling shareholders. Corporate site visits provide institutional investors access to addi­ tional information through face-to-face communications with corporate executives and staff, and such additional information acquired might alleviate the information asymmetry between corporate insiders and outside investors (Bowen, Dutta, Tang, et al., 2018). On the other hand, visiting institutions could also exert pressure on corporate decision making with voting rights and exit threat (Lu, Fung, & Su, 2018). As a result, a higher frequency of institutional investors’ corporate site visits might be associated with enhanced information environment and strengthened corporate governance. Hence, we predict a negative rela­ tion between institutional investors’ corporate site visits and controlling shareholder’s share pledging due to following reasons. First, form the perspective of alleviating information asymmetry, on the one hand, after pledging shares, controlling shareholders are incentivized to maintain stock price, so they often involve themselves in agency problems. Those potential pledging-related agency problems might be more easily detected and exposed under frequent investors’ visits (Su et al., 2021), thus it would add on pressure and risks for controlling shareholder to pursue or continue share pledging. On the other hand, as external financing for investments might be one of the major purposes for share pledging, the reduction in financing costs resulted from frequent visits (Saci & Jasimuddin, 2021) might alleviate financial constrains for visited firms, thus reduce controlling shareholder’s incentive to pledge shares. Second, form the perspective of enhancing corporate gover­ nance, visiting investors might exert pressure on controlling share­ holders’ pledging decisions with face-to-face inquiries and utilize exit threat to protect the benefits of small and medium shareholders, as controlling shareholder’s share pledging is mostly self-interested (He et al., 2022) and widely considered as an irresponsible use of equity by institutions (Pang & Wang, 2020). Taking a unique data of institutional investors’ corporate site visits and a sample of SZSE listed firms from 2013 to 20201, we empirically investigated whether and how institutional investors’ corporate site visits influence controlling shareholders’ share pledging measured by the shares pledged to the total shares held by the controlling shareholder within a fiscal year. The results indicate a significantly negative relation between institutional investors’ corporate site visits and controlling shareholders’ share pledging, which is robust to several robustness checks, including controlling for endogeneity issues with instrumental variables, propensity score matching approach, alternative model specifications, and alternative measures of institutional investors’ corporate site visits. The advantages of institutional investors’ on-site visits are to acquire additional information and to exsert governance on visited firms (Jiang & Yuan, 2018), which might expose and inhibit potential misconducts associated with controlling shareholder’s share pledging. Hence insti­ tutional investors’ corporate site visits are hypothesized to affect controlling shareholder’s share pledging through channels of informa­ tion acquisition and corporate governance. To identify these channel hypotheses, we divide our sample into subsamples based on visited firm’s information and governance characteristics. Consistent with our hypothesis, we find that the negative relation between institutional in­ vestors’ corporate site visits and controlling shareholder’s share pledging is more pronounced for subsamples with a less efficient in­ formation environments and weaker corporate governance. When the price of pledged share falls below a maintenance requirement, a margin call would be triggered asking for additional stocks pledged or cash deposited; otherwise, the controlling shareholder would face the risk of control transfer and the crash of stock price as pledgees are entitled to enforce forced sales. Hence, we predict the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging to be more pronounced when controlling shareholders pledge shares under higher potential margin call pressure. To verify this, we also divide our sample into subsamples based on visited corporates’ stock volatility and based on the macro-economic condition when the visits were conducted. we find that the negative relation is more pronounced when visited firms experience high-volatility stock returns within the last fiscal year and when insti­ tutional visitors conduct site visits during an economic recession, which is consistent with our ex-ante hypothesis. Given that brokerage companies like securities firms are themselves potential pledgees but fund companies are not, we also predict that their visits to have differentiated impacts on controlling shareholder’s share pledging. By examining the moderating effect of the proportions of se­ curity firms and fund companies in visiting institutional investors, we find that a higher proportion of fund companies’ site visits further en­ hances the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging, while the pro­ portion of security firms does not. These findings are consistent with the fact that security companies might have business ties with visited firms in terms of share pledging and the prior arguments that fund companies are usually active external monitors (Borochin & Yang, 2017). The contributions of this paper are threefold. First, it adds to the literature on the economic consequences of institutional investors’ corporate site visits. This strand of literature has investigated the in­ formation acquisition and governance effects of institutional investors’ corporate site visits to influence stock price crash risk (Gao, Cao, & Liu, 2017), corporate innovation (Jiang & Bai, 2022; Jiang & Yuan, 2018)), corporate fraud (Broadstock & Chen, 2021), earning management (Qi, Zhou, & Chen, 2021), managerial myopia (Luo, Wu, Ying, et al., 2022) and etc. This paper enriches the literature by revealing the inhibitory effect of institutional investors’ corporate site visits on controlling shareholder’s share pledging. Second, this paper extends the literature on the determinants of controlling shareholders’ share pledging decisions. Extant literature has mainly focused on the economic consequences of controlling share­ holders’ share pledging on agency problems (Chen, Kao, & Chen, 2007), corporate value (Dou et al., 2019; Kao, Chiou, & Chen, 2004), corporate risks (Anderson & Puleo, 2020; Chauhan, Mishra, & Spahr, 2021) and corporate decision makings (Chan, Chen, & Hu, 2018; Pang & Wang, 2020; Zhu, Xia, & Zheng, 2021), but little has been revealed about the determinants of controlling shareholders’ share pledging. Puleo and Kozlowski (2021) documented that insider shareholders might utilize information asymmetry to time share pledging to maximize personal benefits. This paper adds to this strand of literature by providing the first reliable empirical evidence indicating that enhanced information envi­ ronment and corporate governance resulted from frequent institutional investors’ corporate site visits might facilitate to inhibit controlling shareholder’s share pledging. Third, this study provides a comprehensive analysis revealing the underlying mechanisms through which institutional investors’ corpo­ rate site visits affect controlling shareholder’s share pledging, including information acquisition channel, corporate governance channel, 1 Our sample starts from 2013 due to the following reasons: first, on May 24, 2013, the Shanghai Stock Exchange, Shenzhen Stock Exchange, and China Se­ curities Depository and Clearing Corp promulgated the Business Measures on Trading, Registration, and Settlement of Collateralized Repo of Shares (Trial), allowing brokerage companies to enter the share pledging market; second, on July 17, 2012, the Shenzhen Stock Exchange issued a new regulation requiring that when a SZSE listed firm conducts a site visit, it must disclose a standard summary report about the visit on the firm’s official website within two trading days after the visit. Thus, the sample period from 2013 to 2020 enables access to complete research data and excludes regulation changes that might affect our results. 2 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 potential margin call pressure and the roles of different types of visiting institutions. Since China’s capital market is severely plagued by weak information transparency and property right protection (Wu, Johan, & Rui, 2016), the findings of our mechanism analysis are of significant reference value in protecting the benefits of medium and small share­ holders in China, as well as offering practical implications for other emerging markets suffering from the similar problems. The remainder of this paper is organized as follows. Section 2 dis­ cusses the institutional background of share pledging and institutional investors’ corporate site visits. Section 3 reviews the related literature on both strands, and develops our hypothesis. Section 4 introduces the sample, data, and research methodology. Section 5 reports our empirical results and robustness checks. Section 6 reports the additional analysis. Section 7 concludes the paper. revised Measures for Stock-Pledge Type Repurchase Transaction and Registration and Settlement Business to tighten regulations on share pledging. 2.2. Institutional background of site visits in China Corporate site visits refer to the information acquisition activity that institutional investors, namely analysts and managers from security firms, fund companies, and investment firms, travel to corporates’ headquarters and operating facilities. During a site visit, institutional investors are able to have access to collect first-hand information on visited firm’s operating and governance quality by observing and communicating with its managers and staff. This would help them obtain especially informal knowledge and form impressions on the visited firm’s future prospects and potential risks, which are challenging to be analyzed from financial reports. Besides, corporate site visits are also considered an essential approach to exert external governance as investors can have face-to-face inquiries with senior executives and express their concerns. In terms of information disclosure, in 2009, SZSE mandated that all of its listed firms to disclose institutional investors’ corporate site visits in annual reports; in Hobson et al., 2012, SZSE further required its listed firms to disclose detailed and standard sum­ maries of institutional investors’ corporate site visits on their official websites within two trading days after a visit was conducted. This regulation allows the academia to access data on institutional investors’ corporate site visits in the Chinese capital market and facilitates studies in the related fields. 2. Institutional background 2.1. Institutional background of share pledging in China Ever since the promulgation of the Chinese Security Law in 1995, Share pledging has gradually gained prevalence among shareholders as an external financing channel. Before 2013, share pledging was solely organized in an over-the-counter (OTC) market with a relatively small scale and with main pledgees are commercial banks and trust firms. After 2013, brokerage companies such as security firms were allowed to enter the share pledging market, and both Shenzhen and Shanghai stock exchanges offered a new trading system to facilitate the share pledging process. As a result, the market cap of pledged shares of listed firms in both Shanghai and Shenzhen stock exchanges skyrocketed. By the end of December 2020, over 75% of Chinese listed corporates had pledged shares, and the market value of pledged shares reached 1057 billion RMB, accounting for approximately 9% of the total market cap of China’s floating A shares (He et al., 2022). Compared with traditional financing channels, share pledging is considered faster, easier, and less costly for both pledgees and pledgers (Pang & Wang, 2020). Unlike other types of collateralized materials, collateralized shares are openly traded in the capital market with a convenience of pricing, this reduces pledgees’ risk taking from lending. Share pledging guarantees pledgers a loan of 30%–60% of the market value of the collateralized shares and remains their voting rights, enormously benefiting pledgers’ investments or personal consumption. However, according to the signed contract, if the price of pledged shares falls below a particular maintenance requirement, usually ranging from 130%–150% of the loan amount, a margin call would be triggered, requiring additional stocks pledged or cash deposited. If the pledger is unable to satisfy the maintenance requirement, the pledgee would be entitled to sell off the shares to cover up the loss. In parallel with the soaring share pledging activities, their risks and damages were soon revealed. First, share pledging, usually by control­ ling shareholders, is often associated with corporate scandals in China. For instance, Jia Yueting, the controlling shareholder of Le Shi Internet Information& Technology Corp. (Le Shi TV), had pledged 77% of his shareholdings to 19 pledgees to obtain a 14.2 billion RMB loan during 2011–2015. In order to boost the stock price and obtain more loans by share pledging, Jia exaggerated the corporates’ future prospects and participated in misconducts such as severe overinvestments and ac­ counting manipulations. Till 2018 when his misconducts were exposed, the slump of stock price led bankruptcy of Le Shi TV and left Jia’s 17.6 billion RMB loans unrepaid. Second, share pledging risks are associated with stock price volatility. For example, many pledgers received margin calls during the 2015 Chinese A-share market crash as the market fluctuated. Those who unable to provide additional collaterals or de­ posits were forced to sell off their shareholdings, which caused a downside spiral in stock price. This not only eroded pledged firm’s value but also caused panic in the whole market (He et al., 2022). Conse­ quently, in late 2017, China Securities Regulatory Commission issued 3. Literature review and hypothesis development Controlling shareholder’s share pledging, though as an individual behavior, has essential impacts on the benefits of minority investors and the stability of the whole capital market. Since utilizing shareholdings as collaterals for external financing is easy and flexible, it has become prevalent in emerging markets like China. However, it is not rare that share-pledging-related issues are the cores of corporate scandals, which reveals the potential risks behind the convenience of share pledging (Wang & Chou, 2018). It is argued that external financing granted by share pledging might cover controlling shareholder’s investments and alleviate financial constrain (Deren & Ke, 2018), but it impairs cash flow rights (which are temporarily frozen) of shareholders without affecting the voting rights. This would cause a deviation between these two rights and would raise agency problems between insider shareholders and outside investors (Berkman, Cole, & Fu, 2010), especially in a Chinese institutional background where controlling shareholders are more entrenched and voting rights are strongly emphasized (Pang & Wang, 2020). One pri­ mary concern for a controlling shareholder with shares pledged is that a fall of the stock price might trigger margin calls, which may cause forced sales of the pledged shares and control transfer. Under margin call pressure, the controlling shareholder have stronger incentives to take actions to influence corporate strategies and induce agency problems in order to maintain stock price, even at the expense of other minority shareholders (Zhu et al., 2021). Thus, significant interest and effort have been put into the investigations of the economic consequences of share pledging. One strand of literature has argued that controlling share­ holder’s share pledging is associated with negative impacts on corporate value by reshaping corporate strategy and aggravating agency problems: Kao et al. (2004) find that collateralized shares impair corporate per­ formance; insiders’ share pledging is documented to be positively related to agency problems which would eventually erode corporate value and outside shareholders’ wealth (Chan et al., 2018; Chen et al., 2007; Dou et al., 2019). Pang and Wang (2020) suggest that controlling shareholder’s share pledging is associated with weaker innovation outputs and quality, which may undermine corporates’ long-term competitiveness. Kryzanowski et al. (2021) provide evidence that 3 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 controlling shareholder’s share pledging has a positive relation with corporate fraud. Another strand of the literature suggests that despite the risk-aversion incentives of controlling shareholders after they pledged their shares, their share pledging is in fact positively associated with corporate risks. The findings of Anderson and Puleo (2020) indicate that share pledging is associated with greater equity risk; Chauhan et al. (2021) suggest that larger proportions of share pledging exacerbate stock volatility; Zhou, Li, Yan, et al. (2021) point out that share pledging increases stock price crash risk via bad news hoarding by insiders. With the rapid development of Chinese financial institutions, insti­ tutional investors’ corporate site visits are becoming an important channel to manage investor relationships (Su et al., 2021) and protect shareholders’ wealth (Cheng et al., 2019). Extant literature has docu­ mented positive relations between institutional investors’ corporate site visits and enhanced environment and strengthened corporate gover­ nance. First, institutional investors’ corporate site visits are a tremen­ dously efficient way to acquire firm-specific information (Brown, Call, Clement, et al., 2015). Prior literature on this strand suggests that site visits could alleviate information asymmetry (Bowen et al., 2018), which might improve analysts’ forecast accuracy and portfolio invest­ ment (Bushee, 1998; Wahal & McConnell, 2000). Second, institutional investors are considered to be more experienced and sophisticated compared to other shareholders (Chen, Harford, & Lin, 2015); thus they might play the role of active monitors on corporate decision making (An & Zhang, 2013). Together with the information channel and governance channel, institutional investors’ corporate site visits are suggested to be positively related to better corporate performances, including future stock price (Cheng et al., 2019), corporate innovation (Jiang & Yuan, 2018), social responsibility (Chen, Wan, & Sial, 2021), cash holding (Wang, Lai, Cao, et al., 2022) and reductions of cost of equity (Saci & Jasimuddin, 2021). Literature also provides evidence that site visits reduce agency problems including corporate fraud (Broadstock & Chen, 2021), managerial myopia (Luo et al., 2022) and earning management (Qi et al., 2021). However, Gao et al. (2017) and Lu et al. (2018) note that the bad news hoarding incentives would limit the information ac­ quired by institutional visitors during a site visit, so institutional in­ vestors’ corporate site visits might increase stock price crash risk, while Wang, Liu, and Xiong (2020) argue that it is an inverted U-shape relation between the frequency of institutional visits and stock price crash risk. Yang, Lu, and Xiang (2020) further note that institutional investors’ corporate site visits are negatively associated with future stock price crash risk. Institutional investors’ site visits might impact controlling share­ holder’s share pledging due to the following reasons. First, institutional investors’ corporate site visits might alleviate information asymmetry between insiders and outside investors. Visiting institutional investors can obtain first-hand information, especially informal information, through face-to-face contact with corporate executives and through visiting the operating facilities (Hobson et al., 2012). On the one hand, site visits by institutional investors are possibly to obtain cues about corporate misconduct and opportunistic behaviors when they realize the inconsistency between financial reports and the firm’s actual operations (Firth, Lin, Wong, et al., 2019), which might later be disseminated to the market through research reports, earnings forecasts (Cheng, Du, Wang, et al., 2016) or trading behaviors. Prior literature has documented that in order to maintain stock price to and obtain personal benefits, con­ trolling shareholders’ share pledging is often associated with agency problems like accounting manipulations (Dou et al., 2019) and frauds (Kryzanowski et al., 2021). Therefore, under frequent institutional site visits, the potential share-pledging-related opportunistic behaviors are more likely to be detected and exposed, making it costly and risky for controlling shareholders to obtain self-interested and undue gains by pledging shares. On the other hand, as an important channel to obtain external financing, one essential purpose of share pledging is to alleviate financial constraints and to facilitate corporate investments (Deren & Ke, 2018). With the existence of information asymmetry, it is hard for external investors to obtain the corporates’ actual operation and future prospects, and a risk of investment appraisal would result in higher costs of external financing compared to internal financing. Saci and Jasi­ muddin (2021) document that institutional investors’ corporate site visits enable outsiders to access up-to-date corporate information, which leads to reductions in equity costs. Hence, institutional investors’ corporate site visits might reduce controlling shareholders’ incentives to pledge shares for external financing as they facilitate to alleviate financial constrains for visited firms. Second, from the perspective of agency theory, without adequate monitoring, corporate insiders, including controlling shareholders, would have incentives to tunnel corporate resources for personal use (Yang & Ma, 2022). The tunneling incentive might be strengthened with a more profound deviation between voting rights and cash flow rights after controlling shareholders pledge their shares (Berkman et al., 2010) and would possibly results in a series of detrimental firm-specific out­ comes, including impediments on both short-term and long-term corporate performance (Dou et al., 2019; Pang & Wang, 2020) and increased risks (Anderson & Puleo, 2020; Chauhan et al., 2021). Liter­ ature has argued that institutional investors are an active governance mechanism on listed firms (Chen et al., 2021). Therefore, their site visits might have significant governance effects on visited firms (Broadstock & Chen, 2021; Lu et al., 2018). Furtherly, most of the site visits are con­ ducted by sell-side institutions (Su et al., 2021), their exit threats are powerful tools to influence corporate decisions (Helwege, Intintoli, & Zhang, 2012). Given that share pledging is argued to be mostly selfinterested (He et al., 2022) and widely considered as an irresponsible use equity by institutions (Pang & Wang, 2020), visiting institutional investors might have governance and monitoring effect on controlling shareholder’s share pledging by exerting stronger pressure with face-toface inquiries and expressing concerns about the purpose and potential risks of the share pledging decisions based on the information obtained, and threat to exit until an agreement is reached, which might probably be a reduction of share pledging, especially when controlling share­ holder pledges shares due to self-interest incentives. Given the above considerations, it is reasonable to predict a negative relation between institutional investors’ corporate site visits and controlling share­ holder’s share pledging. Therefore, we formulate our first testable hy­ pothesis as follows: H1. Institutional investors’ corporate site visits will inhibit controlling shareholder’s share pledging. As discussed above, site visits allow institutional investors to obtain first-hand information about the visited firm’s operating conditions by contacting with executives and staff. The additional information ob­ tained and disseminated would help mitigate information asymmetry between corporates’ insiders and outside investors (Bowen et al., 2018). Intuitively, both the amount and benefits of additional information that institutional investors can acquire through site visits would be greater when the visited firms suffer from heavier information asymmetry and are more prone to sugar up or withhold the information essential to assess their performance due to the self-interest incentives (Su et al., 2021). In contrast, if visited firms are inclined to disseminate up-to-date firm-specific information to outside investors, visiting institutions wouldacquire limited additional information as most of the information is already available to the market. Hence, if institutional investors’ corporate site visits inhibit controlling share holders’ share pledging through alleviating information asymmetry, it is predicted that this ef­ fect would be more pronounced when visited firms suffer from heavier information asymmetry. Given this consideration, we formulated the second testable hypothesis as follows: H2a. The negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging will be more pronounced for firms with less efficient information environment, ceteris paribus. 4 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 Yang and Ma (2022) suggest that corporate insiders, including con­ trolling shareholders, will have incentives to tunnel corporate resources for personal use under weak corporate governance, while institutional investors’ corporate site visits might expose potential tunneling behav­ iors and exert pressure on controlling shareholders with exit threat to protect the interests of minority shareholders (Broadstock & Chen, 2021). Similarly, institutional investors’ corporate site visits might better complement corporate governance when the visited firms suffer from weaker corporate governance, where controlling shareholders have more substantial incentives to tunnel corporate resources by pledging their shares. In contrast, if visited firms have strong corporate governance, the visiting institutions would contribute little to their governance as controlling shareholders’ incentives of tunneling are already well inhibited. Hence, if institutional investors’ corporate site visits inhibit controlling share holders’ share pledging through exerting corporate governance, it is also reasonable to predict that this effect would be more pronounced when visited firms suffer from weaker corporate governance. Given this consideration, we formulated the third testable hypothesis as follows: Chinese institutional background. IDR, measured as the number of in­ dependent directors to the number of all board directors, which controls for the potential impact from corporate governance. Top1, the owner­ ship of the largest shareholder, which controls for the potential impacts from shareholding structure. Dual, an indicator variable which equals one if the CEO also chairs the board of directors and zero otherwise, which controls for the potential impact from CEO’s power in corporate decision makings. 4.3. Empirical model To investigate the impact of institutional investors’ corporate site visits on controlling shareholder’s share pledging, we estimate the following standard OLS regression model: PledgedSharesi,t =α + β1 InvestorVisitsi,t + β2 ControlVariablesi,t + Year + Industry + εi,t (1) where i indexes firm and t indexes time, the dependent variable Pledg­ edShares is the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed firm within fiscal year t. InvestorVisits is the independent variable of interests, which equals the natural logarithm of one plus the number of institutional investors’ corporate site visits a firm receives within a fiscal year. ControlVariables is a set of firm-specific control variables that potentially affect controlling shareholder’s share pledging, including ROA, Growth, Age, Lev, SOE, IDR, Top1, Dual. To address the effects of outliers, we winsorize all continuous variables at 1% and 99% level. Standard errors in all regressions are clustered at the firm level to account for correlations between firms. The definition of all variables is in Appendix A. We include both Year and Industry to control for possible year-to-year and industry-to-industry variations in control­ ling shareholder’s share pledging. H2b. The negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging will be more pronounced for firms with weaker corporate governance, ceteris paribus. 4. Sample, data, and methodology 4.1. Sample and data We obtain our data mainly from two sources. The data about insti­ tutional investors’ corporate site visits is from the Chinese Research Data Services Platform (CNRDS) database, and other financial information is from China Stock Market & Accounting Research (CSMAR) database. The sample includes all Chinese A-share listed firms in the SZSE from 2013 to 2020. We then exclude (1) financial listed firms and (2) ob­ servations without enough data for the control variables. The sample consists of 13,350 firm-year observations from 2019 unique firms. 4.4. Descriptive statistics and correlation analysis Panel A of Table 1 reports the descriptive statistics for the variables in our analysis. The mean value of PledgedShares is 0.191, indicating that controlling shareholders of SZSE-listed firms averagely pledge 19.1% of their shares within a fiscal year. The mean value of the number of institutional investors’ corporate site visits is 4.115, indicating that a sample firm averagely holds around four institutional investors’ corpo­ rate site visits within a fiscal year. Thus, the mean value (standard de­ viation) of the independent variable InvestorVisits, which equals the natural logarithm of one plus the number of institutional investors’ corporate site visits a sample firm receives, is 1.105(1.099). Panel B of Table 1 reports the by-year descriptive statistics of interested variables of PledgedShares and InvestorVisits. Panel C of Table 1 reports the Pearson correlation matrix and Spearman correlation matrix, with the Pearson correlations below the main diagonal and the Spearman correlations above it. The bold values indicate statistical significance at the 5% level or lower. The results indicates that the Pearson correlation of PledgedShares and InvestorVisits is not statistically significant, but the Spearman correlation between PledgedShares and InvestorVisits is 0.038, significant at the 5% level. Fig. 1 presents the time trend of the mean values for controlling shareholder’s share pledging and institutional investors’ corporate site visits. As illustrated in Fig. 1, controlling shareholder’s share pledging has shown a general downward trend since 2013, with a more pro­ nounced decline after 2017. This is consistent with the series of regu­ latory policies on equity pledge issued by the China Securities Regulatory Commission and the stock exchanges around 2017. In January 2018, approved by the China Securities Regulatory Commis­ sion, the Shanghai Stock Exchange, the Shenzhen Stock Exchange, and China Securities Depository and Clearing Corporation released the ’Measures for Stock-Pledge Type Repurchase Transaction and Regis­ tration and Settlement Business (Revised in 2018)’ on January 12. The 4.2. Variables definition To capture the variation in the number of pledged shares by con­ trolling shareholders, following Liu and Tian (2021), we employ the dependent variable PledgedShares, as the ratio of controlling share­ holder’s pledged shares to their shareholdings in the listed firm within a fiscal year. To investigate whether and how the intensity of institutional in­ vestors’ corporate site visits impacts controlling shareholder’s share pledging, following Jiang and Yuan (2018), we employ the independent variable InvestorVisits, measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits that a firm holds within a fiscal year. Following Wu, Chen, and Gu (2022), we control for a series of control variables that might impact controlling shareholder’s share pledging in our regressions as follows: ROA, measured as the net margin to assets, which controls for the potential impact from visited firm’s profitability. Size, measured as the natural logarithm of firm’s assets, which controls for the potential impact from visited firm’s scale. Growth, measured as the firm’ operating revenue growth rate, which controls for the potential impact from visited firm’s future prospects. Age, measured as the natural logarithm of one plus the number of years a firm has been listed, which is concerned because firms with longer time being listed might earn more trust from pledgees. Lev, measured as total debt over total assets, which controls for the potential impact from visited firm’s financing structure and risk taking. SOE, an indicator variable equals one if a firm is a stateowned enterprise and zero otherwise, which controls for the potential impact from the vast differences between SOEs and Non-SOEs in a 5 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 Table 1 Descriptive statistics. Panel A Basic statistics. Variables N Mean Std. Dev. p25 Median p75 PledgedShares InvestorVisits ROA Size Q Age Lev SOE IDR Top1 Dual 13,350 13,350 13,350 13,350 13,350 13,350 13,350 13,350 13,350 13,350 13,350 0.191 1.105 0.037 21.978 2.183 1.987 0.389 0.235 0.065 0.322 0.328 0.307 0.980 0.068 1.147 1.329 0.819 0.196 0.424 0.150 0.139 0.470 0.000 0.000 0.015 21.155 1.342 1.386 0.230 0.000 0.003 0.215 0.000 0.000 1.099 0.039 21.838 1.769 2.079 0.376 0.000 0.004 0.300 0.000 0.315 1.792 0.069 22.634 2.539 2.565 0.531 0.000 0.004 0.410 1.000 Panel B By year descriptive statistics of share pledging and institutional site visits. Year N Mean of InvestorVisits Mean of Pledge Shares 2013 2014 2015 2016 2017 2018 2019 2020 1308 1338 1443 1580 1801 1910 1982 1988 1.399 1.453 1.266 1.264 1.088 0.935 0.839 0.875 0.144 0.175 0.202 0.236 0.232 0.212 0.174 0.149 Panel C Pearson and Spearman correlation matrix. Variables 1 2 3 4 5 6 7 8 9 10 11 1.InvestorVisits 2. PledgedShares 3. ROA 4. Size 5. Q 6. Age 7. Lev 8. SOE 9. IDR 10. Top1 11. Dual 1 − 0.004 0.223 0.187 0.067 ¡0.078 ¡0.030 ¡0.065 ¡0.091 0.017 ¡0.020 0.038 1 ¡0.057 0.018 − 0.005 − 0.010 0.046 ¡0.249 ¡0.055 ¡0.055 0.081 0.252 ¡0.025 1 − 0.013 0.157 ¡0.195 ¡0.325 ¡0.060 ¡0.060 0.138 0.013 0.165 0.043 ¡0.093 1 ¡0.383 0.511 0.531 0.316 0.078 0.094 ¡0.186 0.12 0.034 0.294 ¡0.514 1 ¡0.091 ¡0.312 ¡0.121 − 0.003 ¡0.045 0.056 ¡0.121 ¡0.071 ¡0.264 0.537 ¡0.252 1 0.342 0.423 0.130 ¡0.060 ¡0.218 ¡0.03 0.04 ¡0.4 0.515 ¡0.403 0.341 1 0.242 0.051 0.037 ¡0.103 ¡0.073 ¡0.294 ¡0.135 0.304 ¡0.204 0.451 0.233 1 − 0.003 0.149 ¡0.216 ¡0.087 0.004 ¡0.041 0.004 0.009 0.052 0.014 ¡0.064 1 ¡0.066 − 0.017 0.021 − 0.016 0.128 0.046 ¡0.046 ¡0.077 0.021 0.137 ¡0.024 1 0.014 − 0.013 0.103 0.043 ¡0.19 0.089 ¡0.219 ¡0.102 ¡0.216 0.069 0.023 1 Table 1 reports descriptive statistics for the main variables in Panel A, the by-year descriptive statistics of the dependent variable InvestorVisits and independent variable PledgedShares in Panel B, and the correlation matrix of the main variables in panel C. The sample period for all variables is from 2013 to 2020. p25 and p75 stand for 25th percentile and 75th percentile respectively. The bold value in Panel C indicate statistical significance at least at the 5% level. All variables are defined in Appendix A. new regulatory measures impose stricter risk management and super­ vision on share pledge, explicitly stating that the maximum stock pledge ratio should not exceed 60%. shareholders’ share pledging. For controls variables, the coefficients of ROA and SOE are negative and significant, suggesting that controlling shareholders from firms with better profitability and firms that are state-owned will have fewer in­ centives to pledge their shares. The coefficients of Size and Age are positive and significant, indicating that it might be easier for controlling shareholders from firms with larger scales and longer time being listed to pledge their shares. The coefficient of Lev is positive and significant, suggesting that share pledging might be an important source of financing for firms with heavier debt burden. The coefficient of Dual is positive and significant, indicating that controlling shareholders might pledge more shares when CEO also chairs the board of directors. In order to investigate whether a change in the frequency of insti­ tutional visits would lead to changes in controlling shareholder’s share pledging, we run a difference regression model based on Eq. (1), in which the dependent variable, independent variable and control vari­ ables are the remainders of the deduction of the variable between year t and t-1. Column (3) in Panel A of Table 2 reports the regression result of the difference regression model, the coefficient of InvestorVisits is − 0.007, significant at the 10% level. This result indicates that an in­ crease in the frequency of institutional visits from year t-1 to year t 5. Empirical results and discussion 5.1. Main regression results 5.1.1. Baseline regression results This section tests Hypothesis H1, namely whether institutional in­ vestors’ corporate site visits inhibit controlling shareholder’s share pledging. Column (1) in Panel A of Table 2 shows the regression results of Eq. (1) without controlling for year and industry fixed effects. The coefficient of InvestorVisits is − 0.009, significant at the 1% level. Column (2) in Panel A of Table 2 includes all control variables and year and industry fixed effects. The coefficient of InvestorVisits is − 0.010, signif­ icant at 1% level, suggesting a negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging, as a one-standard-deviation increase in the number of insti­ tutional investors’ corporate site visits is associated with a 5.79% (= − 0.010 × 1.105 / 0.191) decrease of a standard deviation in controlling 6 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 Fig. 1. By year descriptive statistics of controlling shareholder’s share pledging and institutional investors’ corporate site visits. This figure illustrates the by year mean value of InvestorVisits and Pledged Share. Worth noting that there is a decline in the mean value of PledgedShares since 2017, which is consistent with the issuance of the revised Measures for Stock-Pledge Type Repurchase Transaction and Registration and Settlement Business aiming to regulate share pledging market. It would make more sense if we also investigate whether institutional investors’ corporate site visits are associated with future changes in controlling shareholder’s share pledging. Hence, we utilize one-year, two-year and three-year lagged values of InvestorVisits as the indepen­ dent variables and rerun Eq.(1). Column (1) to Column (3) of Table 3 report the regression results from lagged independent variable. The coefficients of InvestorVisitst-1, InvestorVisitst-2 and InvestorVisitst-3 are negative and significant at least at the 5% level. The regression results indicate that institutional investors’ corporate site visits significantly associate with future decreases in controlling shareholder’s share pledging, indicating that institutional investors’ corporate site visits might exert not only current, but long-term information and governance effects on visited firms. Table 2 Baseline regression results. Dep. Var. = PledgedShares (1) (2) (3) InvestorVisits − 0.009*** (− 3.28) − 0.187*** (− 3.94) 0.020*** (6.37) 0.004* (1.71) 0.028*** (6.91) 0.087*** (5.00) − 0.223*** (− 38.35) − 0.161*** (− 10.21) − 0.028 (− 1.55) 0.031*** (5.35) − 0.270*** (− 4.23) No 13,350 0.090 − 0.010*** (− 3.40) − 0.157*** (− 3.31) 0.017*** (5.29) − 0.002 (− 0.83) 0.033*** (8.34) 0.112*** (6.29) − 0.223*** (− 36.82) − 0.042 (− 0.65) − 0.008 (− 0.43) 0.032*** (5.47) − 0.236*** (− 3.47) Yes 13,350 0.102 − 0.007* (− 1.65) − 0.063 (− 1.28) 0.041*** (3.13) 0.002 (0.70) 0.110*** (5.06) 0.029 (0.71) − 0.003 (− 0.31) 0.051 (0.62) 0.372*** (4.66) 0.010 (1.02) 0.011 (0.37) Yes 10,978 0.011 ROA Size Q Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 5.2. Robustness tests 5.2.1. Endogeneity The baseline regression results of Eq. (1) might be subject to several endogeneity concerns, for instance, institutional investors’ site visit decisions might be determined by other endogenous factors or unob­ servable reasons, which would cause omitted variable bias to the results. On the other hand, firms with controlling shareholders who pledged a large portion of their shares might raise institutional investors’ atten­ tion, thus attracting more site visits, in the case that Eq. (1) might suffer from a reversed causality problem. To further address the potential endogeneity problems, we perform a two-stage least square (2SLS) regression with instrumental variables (IVs). Following Wang et al. (2022), we employ a dummy variable HighSpeed, which equals one if a sample firm’s located city is connected to a high-speed railway network in a fiscal year t and zero otherwise. Following Su et al. (2021) we next employ a continuous variable Flight, which equals the natural logarithm of one plus the number of direct flights towards the sample firm’s located city. Both of the two IVs are expected to be positively correlated to the number of institutional investors’ corporate site visits that a sample firm receives but not the controlling shareholders’ share pledging, as institutional investors’ corporate site visits are limited to the transport links of a firm’s located city, however, the transport links of a city where a firm locates might not directly affect the share pledging decisions of its controlling shareholder. Column (1) of Table 4 reports the first stage of 2SLS regression with both IVs of HighSpeed and Flight and other control variables in Eq. (1). The results of the first stage 2SLS analysis show that both the two IVs are positively correlated with InvestorVisits at the 1% significance level, Table 2 reports the impact of institutional investors’ corporate site visits on controlling shareholder’ s share pledging of the visited firm with a standard OLS regression model. In Column (1) and Column (2), the dependent variable PledgedShares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed firm within fiscal year t, the inde­ pendent variable InvestorVisits is measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits a firm receives within fiscal year t. In Column (1), we control for all ControlVariables, in Column (2) we add on year and industry fixed effects to control for unobservable heterogeneity across time and across industries. Column (3) reports the results of a difference regression model where all variables are the remainders of the deduction of the variable between year t and t-1. All other variables are defined in Appendix A. The t statistics reported in parentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. directly leads to a reduction in the shares that controlling shareholder pledges in year t, which strengthens our baseline regression results in Column (1) and Column (2). 7 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 Table 3 Regression results from lagged independent variable. Dep. Var. = PledgedShares (1) InvestorVisitst-1 − 0.008*** (− 2.62) InvestorVisitst-2 InvestorVisitst-3 ROA Size Q Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 − 0.183*** (− 3.62) 0.019*** (5.38) − 0.002 (− 1.02) 0.020*** (3.67) 0.109*** (5.53) − 0.220*** (− 32.29) − 0.045 (− 0.70) − 0.022 (− 1.10) 0.037*** (5.71) − 0.224*** (− 2.98) Yes 10,985 0.104 (2) − 0.010*** (− 2.98) − 0.160*** (− 3.00) 0.022*** (5.69) − 0.005** (− 2.06) 0.016** (2.42) 0.082*** (3.84) − 0.218*** (− 28.89) − 0.031 (− 0.46) − 0.010 (− 0.46) 0.040*** (5.52) − 0.236*** (− 2.88) Yes 9057 0.105 Table 4 2SLS analysis. (3) Dep. Var. = InvestorVisits Dep. Var. = PledgedShares (1) (2) InvestorVisits HighSpeed − 0.009** (− 2.47) − 0.149** (− 2.53) 0.020*** (4.76) − 0.011*** (− 3.66) 0.004 (0.45) 0.079*** (3.23) − 0.217*** (− 25.09) − 0.027 (− 0.39) − 0.004 (− 0.16) 0.035*** (4.33) − 0.177* (− 1.96) Yes 7239 0.110 Flight ROA Size Q Age Lev SOE IDR Top1 Dual Constant Kleibergen-Paap Wald F Hansen J p-value Year and Industry FE Observations Adjusted R2 Table 3 reports the impact of institutional investors’ corporate site visits on controlling shareholder’s future share pledging of the visited firm by utilizing lagged InvestorVisits as independent variables. The dependent variable Pledged­ Shares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed corporates within fiscal year t. The independent variable InvestorVisitst-1 in column (1) is InvestorVisits lagged by one year, the independent variable InvestorVisitst-2 in column (2) is InvestorVisits lagged by two years, and the independent variable InvestorVisitst-3 in column (3) is InvestorVisits lagged by three years. We control for all ControlVariables and control for year and industry fixed effects to control for unobservable heterogeneity across time and across industries in all regressions. All other variables are defined in Ap­ pendix A. The t statistics reported in parentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. 0.196*** (7.96) 0.023*** (7.58) 1.913*** (15.88) 0.382*** (36.98) 0.091*** (12.66) − 0.167*** (− 13.05) − 0.354*** (− 6.98) − 0.201*** (− 9.49) 0.285 (1.28) − 0.313*** (− 5.51) 0.001 (0.09) − 6.859*** (− 30.95) 93.416 0.823 Yes 13,350 0.218 − 0.123*** (− 4.58) 0.057 (0.81) 0.061*** (5.59) 0.010*** (2.64) 0.013** (2.03) 0.070*** (3.34) − 0.244*** (− 29.73) − 0.012 (− 0.17) − 0.047** (− 2.25) 0.033*** (5.31) − 1.003*** (− 5.13) Yes 13,350 0.279 Table 4 reports the impact of institutional investors corporate site visits on controlling shareholders’ share pledging of the visited firm. To address the po­ tential endogeneity problems, we adopt two instrumental variables (IVs) to perform an 2SLS analysis. The first instrumental variable is HighSpeed, an indi­ cator variable which equals one if a sample firm’s located city is connected to a high-speed railway network in fiscal year t and zero otherwise. The second instrumental variable is Flight, which equals the natural logarithm of one plus the number of direct flights towards the sample firm’s located city. InvestorVisits stands for the natural logarithm of one plus the number of institutional in­ vestors’ corporate site visits a firm receives within fiscal year t. Column (1) in Table 4 reports the first stage regression results of IVs on InvestorVisits. In Column (2) in Table 4, we utilize the predicted value of the frequency of investors’ site visits from first-stage regression to further investigate the impact of investors’ site visits on controlling shareholder’s share pledging after controlling for po­ tential endogeneity problems. We control for all Control Variables and control for year and industry fixed effects to control for unobservable heterogeneity across time and across industries in all regressions. All other variables are defined in Appendix A. The t statistics reported in parentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. indicating that better transport links facilitate institutional investors’ corporate site visits. The Kleibergen-Paap Wald F statistic is 93.416, indicating that both of the IVs are not weak instruments. The over­ identification test (Hansen Jp-Value) is >10% level, indicating that both of the IVs are valid and the structural model is correctly specified. After controlling for any possible endogeneity concerns, Column (2) of Table 4 reports the second stage of 2SLS analysis, the coefficient of InvestorVisits is still negative and significant at the 1% level with a greater magnitude. Thus, it can be concluded that institutional investors’ corporate site visits are negatively related to the controlling shareholder’s share pledging of the visited firm. between the control group and the treated group by comparing the mean value of these variables. Column (1) and Column (2) in Panel C of Table 5 reports the OLS regression results of Eq. (1) on the PSM sub­ sample, the coefficients of InvestorVisits are still negative and statistically significant at the 1% level, suggesting that the baseline regression results remain robust after controlling the potential sample selection bias. 5.2.2. Propensity score matching (PSM) method To concern the potential selection bias, we adopt a propensity score matching (PSM) method between firms that receive at least one insti­ tutional site visit within fiscal year t (Treated group) and those firms that do not (Control group). To calculate the propensity score, we employ a Logit model controlling other firm-specific ControlVariables in the baseline regression to predict the effect of InvestorVisits on PledgedShares. We then perform a 1:1 neighbor matching between the control group and the treated group. We finally obtain a subsample of 5261 matched firm-year observations. Panel A and Panel B of Table 5 report the uni­ variate analysis results on the full sample and PSM subsample respec­ tively, we find that most of the variables are not significantly different 5.2.3. Alternative model specification Our baseline regression results might suffer from biased model specification problem from two aspects. First, since the median value of PledgedShares in our sample equals zero, which means quite a few ob­ servations have no share pledging during a fiscal year, the error term of the dependent variable might not follow a normal distribution. Second, a standard OLS estimation result might be biased considering that the value of PledgedShares is limited between 0 and 1. To address these 8 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 InvestorVisits measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits a firm receives within fiscal year t. Panel A reports the univariate difference of PledgedShares and ControlVariables between the treated group and control group of the full sample. Panel B reports the univariate difference of PledgedShares and ControlVariables between the treated group and control group of the PSM subsample. Panel C reports the multivariable regression results from the subsample. In Column (1), we control for all ControlVariables, in Column (2) we add on year and industry fixed effects to control for unobservable heterogeneity across time and across industries. All other variables are defined in Appendix A. The t statistics reported in paren­ theses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. Table 5 PSM method. Panel A Univariate analysis of variables before PSM. PledgedShares ROA Size Q Age Lev SOE IDR Top1 Dual (1) (2) (3) (4) InvestorVisits = 0 InvestorVisits > 0 N = 4230 N = 9120 Mean Mean Difference T-value 0.176 0.018 21.770 2.072 2.112 0.404 0.292 0.083 0.320 0.324 0.197 0.046 22.070 2.230 1.930 0.383 0.205 0.059 0.322 0.333 − 0.023 − 0.028 − 0.309 − 0.164 0.184 0.021 0.084 0.026 − 0.003 − 0.005 − 3.960*** − 22.897*** − 14.618*** − 6.627*** 12.140*** 5.829*** 10.711*** 9.434*** − 0.984 − 0.614 concerns, we employ a Poisson model and a Tobit model to re-estimate Column (2) of Table 2 respectively. The result of Poisson model esti­ mation is reported in Column (1) of Table 6 and the result of Tobit model estimation is reported in Column (2) of Table 6. The coefficients of InvestorVisits from alternative models are both negative and significant at least at 5% level, suggesting that the baseline regression results remain robust after applying alternative model specifications. Panel B Univariate analysis of variables after PSM. (1) (2) InvestorVisits = 0 InvestorVisits > 0 N = 2849 N = 2772 PledgedShares ROA Size Q Age Lev SOE IDR Top1 Dual (3) (4) Mean Mean Difference T-value 0.184 0.029 21.862 2.100 2.015 0.394 0.068 0.323 0.184 0.029 0.196 0.029 21.879 2.086 2.043 0.395 0.072 0.322 0.196 0.029 − 0.012 0.001 − 0.016 0.014 − 0.028 − 0.001 0.050 − 0.004 0.001 0.006 − 1.472 0.426 − 0.558 0.406 − 1.251 − 0.138 4.280*** − 0.939 0.161 0.511 5.2.4. Alternative measurement for independent variable In order to further enhance our baseline regression results, following Luo et al. (2022), we construct alternative measurements for the inde­ pendent variable. The first alternative measurement InvestorNumber is Table 6 Alternative model specification. Dep. Var. = PledgedShares InvestorVisits ROA Size Q Panel C OLS regression on PSM sample. Dep. Var. = PledgedShares (1) (2) InvestorVisits − 0.010** (− 2.41) − 0.172** (− 2.51) 0.023*** (4.49) 0.004 (1.16) 0.026*** (4.21) 0.092*** (3.56) − 0.229*** (− 24.54) − 0.141*** (− 5.93) − 0.005 (− 0.18) 0.032*** (3.43) − 0.344*** (− 3.24) No 5621 0.092 − 0.014*** (− 3.14) − 0.138** (− 2.03) 0.022*** (4.18) − 0.002 (− 0.40) 0.035*** (5.45) 0.109*** (4.13) − 0.231*** (− 23.84) − 0.151 (− 1.62) 0.007 (0.26) 0.034*** (3.71) − 0.297*** (− 2.66) Yes 5621 0.100 ROA Size Q Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Psuedo R2 Poisson Model Tobit Model (1) − 0.054*** (− 3.66) − 0.676*** (− 3.33) 0.087*** (5.09) − 0.020 (− 1.62) 0.177*** (8.15) 0.602*** (6.76) − 1.704*** (− 27.50) − 0.105 (− 0.24) − 0.024 (− 0.25) 0.171*** (6.19) − 0.054*** (− 3.66) Yes 13,350 0.063 (2) − 0.018** (− 2.33) − 0.315*** (− 2.65) 0.072*** (8.01) 0.009 (1.42) 0.056*** (4.88) 0.314*** (6.56) − 0.784*** (− 32.47) 0.029 (0.14) 0.065 (1.29) 0.095*** (6.53) − 0.018** (− 2.33) Yes 13,350 0.094 Table 6 reports the impact of institutional investors corporate site visits on controlling shareholders’ share pledging of the visited firm with alternative model specifications. The dependent variable PledgedShares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed firm within fiscal year t, the independent variable InvestorVisits is measured as the natural logarithm of one plus the number of institutional in­ vestors’ corporate site visits a firm receives within fiscal year t. In Column (1), we utilize a Poisson model to address the concern that error term of the dependent variable might not follow a normal distribution. In Column (2) we utilize a Tobit model to address the potential bias considering that the value of PledgedShares is limited between 0 and 1. We control for all Control Variables and control for year and industry fixed effects to control for unobservable hetero­ geneity across time and across industries in all regressions. All other variables are defined in Appendix A. The t statistics reported in parentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. Table 5 reports impact of institutional investors’ corporate site visits on con­ trolling shareholders’ share pledging of the visited firm by using propensity score matching method (PSM). The control group is selected based on a nearestneighbor matching of propensity scores from a logit regression controlling for other firm-specific ControlVariables. The dependent variable PledgedShares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed firm within fiscal year t, the independent variable is 9 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 measured as the number of visiting institutions to a firm within fiscal year t scaled by 100. The second alternative measurement VisitDummy is a dummy variable that equals one if a firm’s InvestorVisits is higher than the year-industry median value and zero otherwise. Column (1) and Column (2) in Table 7 report the regression results employing alterna­ tive measurements of Investor Number and Visit Dummy respectively, the results show that the coefficients of both Investor Number and Visit Dummy are negative and significant at the 1% level, indicating a rein­ forced negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging. and re-estimate Eq. (1) respectively. The results in Column (1) of Table 8 show that the coefficient of InvestorVisits for the low-REM group is not statistically significant, but for the high-REM group in Column (2), the coefficient of InvestorVisits is negative and significant at the 1% level, indicating that institutional investors’ corporate site visits have more pronounced inhibitory effect on controlling shareholder’s share pledging for firms with lower earning disclosure quality. 5.3.2. Analysts’ forecast deviation Following Behn and Kang (2008), another metric we use to measure listed firm’s information environment is analysts’ ForecastDeviation, measured as the mean value of deviations between forecasted earnings per share by analysts and the actual earnings per share of a sample firm at year-end. A higher ForecastDeviation indicates a lower quality of an­ alysts’ forecast, suggesting analysts can acquire less information from the market. We partition samples into the low-forecast-deviation group 5.3. Effect of the information environment In order to test Hypothesis H2a, namely whether the negative rela­ tion between institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced for firms with a less efficient information environment, we employ the following two metrics. Table 8 Cross sectional analysis by the visited firm’s information environment. 5.3.1. Real earning management The first metric is real earning management (REM). Francis, LaFond, Olsson, et al. (2005) have argued that a higher REM level indicates a lower quality of firm-specific earning disclosure, resulting in a lower quality of information for investors. We estimate firm-specific REM following Roychowdhury (2006) and partition the whole sample into low-REM group and High-REM group along the median value of REM Dep. Var. = PledgedShares InvestorVisits ROA Table 7 Alternative measurement for the independent variable. Dep. Var. = PledgedShares (1) InvestorNumber − 0.015*** (− 3.10) VisitDummy ROA Size Q Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 − 0.160*** (− 3.36) 0.017*** (5.18) − 0.001 (− 0.59) 0.034*** (8.33) 0.113*** (6.40) − 0.223*** (− 36.74) − 0.041 (− 0.64) − 0.008 (− 0.45) 0.032*** (5.46) − 0.239*** (− 3.46) Yes 13,350 0.102 Size Q (2) Age Lev − 0.014*** (− 2.64) − 0.164*** (− 3.45) 0.016*** (4.96) − 0.002 (− 1.00) 0.034*** (8.51) 0.113*** (6.37) − 0.222*** (− 36.79) − 0.043 (− 0.66) − 0.007 (− 0.37) 0.032*** (5.49) − 0.212*** (− 3.16) Yes 13,350 0.102 SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 Real Earning Management Forecast Deviation Low High Low High (1) (2) (3) (4) − 0.006 (− 1.39) − 0.003 (− 0.04) 0.016*** (3.39) 0.001 (0.38) 0.013 (1.56) 0.110*** (4.19) − 0.211*** (− 22.01) − 0.048 (− 0.51) − 0.004 (− 0.15) 0.031*** (3.44) − 0.190* (− 1.90) Yes 5875 0.103 − 0.019*** (− 4.41) − 0.182*** (− 2.70) 0.024*** (4.70) − 0.008*** (− 2.69) 0.018** (2.32) 0.091*** (3.42) − 0.221*** (− 24.58) 0.008 (0.08) − 0.015 (− 0.53) 0.040*** (4.37) − 0.308*** (− 2.94) Yes 5886 0.108 − 0.006 (− 1.43) − 0.579*** (− 4.76) 0.003 (0.66) − 0.003 (− 0.87) 0.050*** (8.57) 0.098*** (3.18) − 0.245*** (− 26.72) 0.074 (0.72) 0.002 (0.07) 0.031*** (3.41) 0.166 (1.46) Yes 5164 0.123 − 0.014*** (− 2.87) − 0.120* (− 1.87) 0.025*** (4.38) 0.002 (0.46) 0.023*** (3.08) 0.135*** (4.51) − 0.219*** (− 20.22) − 0.079 (− 0.77) − 0.005 (− 0.16) 0.046*** (4.69) − 0.410*** (− 3.37) Yes 5140 0.087 Table 8 reports the cross sectional analysis results by visited firm’s information environment. We adopt two metrics to measure the information environment of visited firms. The first metric is real earning management (REM), estimated by the method following Roychowdhury (2006). The second metric is analysts’ Forecast Deviation measured as the mean value of deviations between forecasted earnings per share by analysts and the actual earnings per share of a sample firm at year-end. The higher the value of REM (ForecastDeviation), the more ineffi­ cient the firm’s information environment is. We partition the whole sample into two groups along the median value of REM and ForecastDeviation respectively. The dependent variable PledgedShares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed firm within fiscal year t, the independent variable InvestorVisits is measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits a firm receives within fiscal year t. Column (1) and Column (2) reports the stan­ dard OLS regression results from groups with lower and higher level of REM. Column (3) and Column (4) reports the standard OLS regression results from groups with lower and higher level of ForecastDeviation. We control for all Control Variables and control for year and industry fixed effects to control for unobservable heterogeneity across time and across industries in all regressions. All other variables are defined in Appendix A. The t statistics reported in pa­ rentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. Table 7 reports the impact of institutional investors corporate site visits on controlling shareholders’ share pledging of the visited firm with alternative measurements for the independent variable. The independent variable in Col­ umn (1) is InvestorNumber measured as the number of visiting institutions to a listed firm within fiscal year t scaled by 100. The independent variable in Col­ umn (2) is VisitDummy, a dummy variable that equals one if a sample firm’s InvestorVisits is higher than the year-industry median value, and zero otherwise. All regressions are based on a standard OLS model. We control for all Control Variables and control for year and industry fixed effects to control for unob­ servable heterogeneity across time and across industries in all regressions. All other variables are defined in Appendix A. The t statistics reported in paren­ theses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. 10 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 and high-forecast-deviation group along the median value of Fore­ castDeviation and re-estimate Eq. (1) respectively. The results in Column (3) of Table 8 show that the coefficient of InvestorVisits for the lowforecast-deviation group is not statistically significant, but for the high-forecast-deviation group in Column (4), the coefficients of Invest­ orVisits is negative at the 1% significant level, indicating that institu­ tional investors’ corporate site visits have more pronounced inhibitory effect on controlling shareholder’s share pledging for firms that have higher analysts’ forecast deviation. To sum up, the above results indicate that the negative relation be­ tween institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced for firms with a less efficient information environment, which supports Hypothesis H2a. Table 9 Cross sectional analysis by the visited firm’s corporate governance. Dep. Var. = PledgedShares InvestorVisits ROA Size Q Age 5.4. Effect of corporate governance Lev In order to test Hypothesis H2b, namely whether the negative rela­ tion between institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced for firms with weaker corporate governance, we employ the following two metrics. SOE IDR Top1 5.4.1. Total assets turnover Following Ang, Cole, and Lin (2000), we employ total assets turnover to proxy for a firm’s agency cost, which is expected to be lower among firms with lower firm-specific agency efficiency, indicating weaker corporate governance. We partition the whole sample into the low-totalassets-turnover group and the high-total-assets-turnover group along the median value of total assets turnover and re-estimate Eq. (1) respec­ tively. The results in Column (1) of Table 9 show that the coefficient of InvestorVisits for the low-total-assets-turnover group is significant at the 1% level, but for the high-total-assets-turnover group in Column (2), the coefficient of InvestorVisits is not statistically significant, indicating that institutional investors’ corporate site visits have more pronounced inhibitory effect on controlling shareholder’s share pledging for firms with lower total assets turnover. Dual Constant Year and Industry FE Observations Adjusted R2 Total Assets Turnover Independent Director Rate Low High Low High (1) (2) (3) (4) − 0.013*** (− 3.04) − 0.126* (− 1.92) 0.032*** (6.55) − 0.003 (− 0.86) 0.025*** (3.93) 0.118*** (4.76) − 0.227*** (− 25.48) − 0.065 (− 0.71) − 0.020 (− 0.76) 0.037*** (4.44) − 0.499*** (− 4.86) Yes 6730 0.112 − 0.004 (− 1.13) − 0.100 (− 1.31) 0.002 (0.46) − 0.001 (− 0.29) 0.039*** (7.47) 0.137*** (4.93) − 0.215*** (− 26.20) − 0.006 (− 0.06) 0.026 (1.04) 0.026*** (3.13) 0.014 (0.16) Yes 6618 0.094 − 0.012*** (− 3.28) − 0.181*** (− 2.69) 0.025*** (5.70) 0.002 (0.48) 0.030*** (5.77) 0.084*** (3.54) − 0.215*** (− 27.71) − 0.219** (− 1.97) − 0.062** (− 2.50) 0.039*** (4.69) − 0.411*** (− 4.41) Yes 7096 0.113 − 0.007 (− 1.62) − 0.121* (− 1.78) 0.008* (1.69) − 0.005* (− 1.66) 0.039*** (6.19) 0.147*** (5.50) − 0.233*** (− 23.93) 0.009 (0.10) 0.049* (1.81) 0.024*** (2.84) − 0.039 (− 0.39) Yes 6252 0.090 Table 9 reports cross sectional analysis results by the visited firm’s corporate governance. We adopt two metrics to measure the corporate governance of visited firms. The first metric is Total Assets Turnover, measured as sales revenue to the average value of total assets. The second metric is Independent Director Rate measured as the number of independent directors on board to the number of total directors on board. The higher the value of Total Assets Turnover (Inde­ pendent Director Rate), the better the firm’s corporate governance is. We partition the whole sample into two groups along the median value of Total Assets Turn­ over and Independent Director Rate respectively. The dependent variable Pledg­ edShares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed firm within fiscal year t, the independent vari­ able InvestorVisits is measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits a firm receives within fiscal year t. Column (1) and Column (2) reports the standard OLS regression results from groups with lower and higher level of Total Assets Turnover. Column (3) and Column (4) reports the standard OLS regression results from groups with lower and higher level of Independent Director Rate. We control for all Control Variables and control for year and industry fixed effects to control for unobservable het­ erogeneity across time and across industries in all regressions. All other variables are defined in Appendix A. The t statistics reported in parentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. 5.4.2. Independent director rate We then employ independent director rate as another metric measuring corporate governance. Literature has argued that indepen­ dent directors can improve corporate governance by alleviating the conflicts between managers and shareholders, and supporting decision makings with professional knowledge and techniques (Fama & Jensen, 1983). Thus a lower ratio of independent directors on the board might indicate the weakness of a corporate’s governance. We partition the whole sample into the low-independent-director-rate group and the high-independent-director-rate group along the median value of IDR and re-estimate Eq. (1) respectively. The results in Column (3) of Table 9 show that the coefficient of InvestorVisits for the low-independentdirector-rate group is significant at the 1% level, but for the highindependent-director-rate group in Column (4), the coefficient of InvestorVisits is not statistically significant, indicating that institutional investors’ corporate site visits have more pronounced inhibitory effect on controlling shareholder’s share pledging for firms with less inde­ pendent directors on board. To sum up, the above results indicate that the negative relation be­ tween institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced for firms with weaker corporate governance, which supports Hypothesis H2b. falls below the loan maintenance requirement, a margin call would be triggered to require additional collaterals or deposits, or the pledger would face forced sales of the stock and a control transfer. In view of this, the controlling shareholders who pledge their shares under higher margin call pressure would have stronger incentives to involve in agency problems, thus they are more prone to participate in accounting ma­ nipulations and frauds. As for visiting institutional investors, their in­ formation acquisition and governance effect might also be more pronounced when the controlling shareholder conduct riskier share pledging or have a higher probability to be involved in agency problems. Thus, we hypothesize institutional investors’ corporate site visits would have more pronounced impact on controlling shareholder’s share pledging when they are under higher margin call pressure. To investigate whether the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging would be more pronounced when the share pledging is under 6. Additional analysis 6.1.1. Visited corporates’ stock volatility One essential concern for share pledging is that once the stock price 11 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 higher margin call pressure, we employ two metrics to measure the margin call pressure. The first metric is visited firm’s stock volatility within the last fiscal year. It is expected that institutional investors would assess controlling shareholder’s risk taking in pledging by judging the stock’s past performance, and it might be considered as riskier and under higher margin call pressure when the visited firm experienced a higher level of stock volatility within the last fiscal year. Given this consideration, we partition the whole sample into the low-stockvolatility group and the high-stock-volatility group along the median value of visited firm’s stock volatility within fiscal year t-1 and reestimate Eq. (1) respectively. Column (1) and Column (2) in Table 10 report the results from regressions of institutional investors’ corporate site visits on controlling shareholder’s share pledging for groups with different level of stock volatility. The results indicate that for with lower stock volatility in column (1), the coefficient of InvestorVisits is− 0.008, significant at the 5% level, but for firms with higher stock volatility in column (2), the coefficient of InvestorVisits is− 0.018, significant at the 1% level. The results imply that the magnitude and significance of the effect of institutional investors’ corporate site visits on controlling shareholders’ share pledging is greater for firms that experienced a larger stock volatility within in the last fiscal year, which is consistent with our ex-ante hypothesis. 6.1.2. Macroeconomic condition The second metric is the macro-economic condition when the corporate site visits are conducted. During a recession of the economy, firms might be more likely to suffer from decreases in revenue and stock price, which may raise the probability of margin call pressure. Hence we hypothesize that during a recession, institutional investors should have stronger incentives to inhibit controlling shareholder’s share pledging. Following Cogley and Nason (1995), we partition the full sample into recession group and boom group depending on the trend term of China’s annual real GDP growth after utilizing Hodrick Prescott filter. If the trend term is below 0, then the year is partitioned into recession group, and boom group otherwise. Column (3) and Column (4) in Table 10 report the results from regressions of institutional investors’ corporate site visits on controlling shareholder’s share pledging for different macroeconomic conditions. The results in column (3) indicate that during a recession, the coefficient of InvestorVisits is negative and sig­ nificant at the 1% level, but during the boom period in column (4), the coefficient of InvestorVisits is not statistically significant, implying that the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced when institutional investors conduct site visits during an economic recession. To sum up, the above results indicate that the negative relation be­ tween institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced when the share pledging is associated with higher margin call pressure. Table 10 Cross sectional analysis by the visited firm’s margin call pressure. Dep. Var. = PledgedShares InvestorVisits ROA Size Q Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 Stock Volatility Macroeconomic Condition Low High Recession Boom (1) (2) (3) (4) -0.008** (− 1.96) − 0.246*** (− 3.69) 0.012** (2.47) 0.001 (0.33) 0.026*** (3.07) 0.134*** (4.97) − 0.215*** (− 21.46) − 0.069 (− 0.71) 0.007 (0.25) 0.042*** (4.68) − 0.138 (− 1.32) Yes -0.018*** (− 3.53) − 0.151** (− 2.35) 0.039*** (6.37) − 0.005 (− 1.34) 0.019** (2.43) 0.086*** (3.06) − 0.231*** (− 18.20) − 0.011 (− 0.10) − 0.048 (− 1.49) 0.032*** (3.55) − 0.521*** (− 3.94) Yes − 0.016*** (− 4.28) − 0.125* (− 1.95) 0.017*** (4.01) 0.003 (0.99) 0.028*** (4.89) 0.144*** (6.24) − 0.222*** (− 27.90) − 0.040 (− 0.61) − 0.013 (− 0.58) 0.028*** (3.49) − 0.239*** (− 2.68) Yes 0.001 (0.19) − 0.193*** (− 2.76) 0.015*** (2.99) − 0.013*** (− 3.08) 0.039*** (6.86) 0.068** (2.44) − 0.226*** (− 23.71) − 0.584*** (− 4.30) 0.001 (0.03) 0.037*** (4.21) − 0.150 (− 1.39) Yes 5461 0.120 5424 0.100 7657 0.101 5693 0.101 6.1.3. Types of visiting institutions In addition, different types of institutional investors’ visits might play different roles in influencing controlling shareholder’s share pledging based on their different nature. Specifically, “transient” insti­ tutional investors are investors who are willing to have business ties with listed firms and they lack the motivation and ability to participate in corporate governance; on the contrary, “dedicated” institutional in­ vestors are investors who have no business ties with listed firms and they are actively participate in corporate governance in seeking of long-term value (Brickley, Lease, & Smith Jr., 1988; Panicker, Mitra, & Upad­ hyayula, 2019). Literature suggests that ownership by “dedicated” type of institutional investors has positive implications on firm’s future value and governance characteristics, while ownership by “transient” type of institutional investors does not (Borochin & Yang, 2017; Bushee, 1998). Besides ownership by different types of institutional investors, Jiang and Bai (2022) find that site visits by “transient” institutional investors, namely security firms, can promote only strategic green innovation of visited firms due to catering incentives, while site visits by “dedicated” type of institutional investors, namely fund companies, are able to promote visited firm’s substantial green innovation. In terms of share pledging, the roles and impacts of visits done by security firms and fund companies may further differ, because since 2013, brokerage companies have been allowed to enter the share pledging market, this makes se­ curity firms potential pledgees and may offset their impact on control­ ling shareholders’ share pledging. In view of this, we hypothesize that corporate site visits done by security firms would have limited impacts on controlling shareholder’s pledging decisions as they are usually reluctant to participate in corporate governance and might have business ties with visited firms; in contrast, corporate site visits done by fund companies might better inhibit controlling shareholder’s pledging decisions as they are usually Table 10 reports cross sectional analysis results by the visited firm’s margin call pressure. We adopt two metrics to measure the margin call pressure for con­ trolling shareholders. The first metric is visited corporate’s StockVolatility within fiscal year t-1. The second metric is MacroeconomicCondition measured as the trend term of China’s annual real GDP growth after Hodrick Prescott filter. The higher the value of Stock Volatility, the higher the controlling shareholder’s margin call pressure is, the above-zero value of the trend term represents good macroeconomic condition, where controlling shareholders face less margin call pressure when pledging shares. We partition the whole sample into two groups along the median value of Stock Volatility and by Macroeconomic Condition. The dependent variable PledgedShares is measured as the ratio of controlling share­ holder’s pledged shares to their shareholdings in the listed firm within fiscal year t, the independent variable InvestorVisits is measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits a firm re­ ceives within fiscal year t. Column (1) and Column (2) reports the standard OLS regression results from groups with lower and higher level of Stock Volatility. Column (3) and Column (4) reports the standard OLS regression results from groups with bad and good macroeconomic condition. We control for all Control Variables and control for year and industry fixed effects to control for unob­ servable heterogeneity across time and across industries in all regressions. All other variables are defined in Appendix A. The t statistics reported in paren­ theses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. 12 Z. Xiao et al. International Review of Financial Analysis 89 (2023) 102746 active external monitors to corporate decisions and do not involve in pledging business. Thus, the proportion of security firms’ visits might not impact controlling shareholder’s share pledging, while a higher proportion of fund companies’ visits might strengthen the information channel and governance channel to better inhibit controlling share­ holder’s share pledging. Following Luo et al. (2022), we employ the following Eq. (2) and Eq. (3) to examine the impact of different types of institutional investors’ corporate site visits on controlling shareholder’s share pledging. Table 12 Corporate site visits, the proportions of different visiting institutions, and con­ trolling shareholder’s share pledging. (1) InvestorVisits×SecRatio − 0.016 (− 0.96) InvestorVisits×FundRatio SecRatio FundRatio PledgedSharesi,t = α + λ1 InvestorVisitsi,t × SecRatioi,t + λ2 SecRatioi,t (2) + λ3 InvestorVisitsi,t + λ4 ControlVariablesi,t + Year + Industry + εi,t InvestorVisits ROA PledgedSharesi,t = α + η1 InvestorVisitsi,t × FundRatioi,t + η2 FundRatioi,t Size + η3 InvestorVisitsi,t + η4 ControlVariablesi,t + Year + Industry + εi,t (3) Q where SecRatio (FundRatio) is defined as the number of visiting security firms (fund companies) to the number of the total visiting institutions. Table 11 shows the summary statistics of SecRatio and FundRatio, which indicates that averagely 37.9% of institutional visitors are from security firms, and 20.2% of institutional visitors are from fund companies. Column (1) of Table 12 reports the impact of the proportion of se­ curity firm’s visits on controlling shareholder’s share pledging. The coefficient of InvestorVisits×SecRatio is not statistically significant, while the coefficient of InvestorVisits is − 0.012, significant at the 5% level. Column (2) of Table 12 reports the impact of the proportion of fund company’s visits on controlling shareholder’s share pledging. The co­ efficient of InvestorVisits×FundRatio is − 0.079, significant at the 1% level, while the coefficient of InvestorVisits is not statistically significant. The above results indicate that, as hypothesized, the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging is mainly contributed by fund companies’ visits. Age Lev SOE IDR Top1 Dual Constant Year and Industry FE Observations Adjusted R2 As an ideal way of granting external financing, share pledging has become prevalent among controlling shareholders of listed firms around the globe, however, it is also proved to be associated with various agency problems and negative outcomes. Institutional investors’ corporate site visits emerge and as an essential approach to acquire in­ formation and enhance corporate governance, but their impacts on share pledging is unrevealed. Taking a unique database on SZSE-listed firms in China, our study finds a negative relation between institu­ tional investors’ corporate site visits and controlling shareholder’s share pledging, and this relation is suggested to be casual after controlling for potential endogeneity problems by employing instrumental variables and a PSM method. Further analysis indicates that institutional investors’ corporate site visits might inhibit controlling shareholder’s share pledging through alleviating information asymmetries and strengthening corporate governance, as the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging is more pronounced and significant among firms with less efficient information Std. Dev. p25 Median p75 0.379 0.202 0.241 0.167 0.228 0.063 0.333 0.200 0.500 0.300 − 0.012** (− 2.40) − 0.159*** (− 3.33) 0.018*** (5.40) − 0.001 (− 0.57) 0.034*** (8.41) 0.113*** (6.38) − 0.222*** (− 36.68) − 0.045 (− 0.71) − 0.009 (− 0.49) 0.032*** (5.44) − 0.259*** (− 3.75) Yes 13,350 0.103 − 0.079*** (− 3.66) 0.107*** (3.62) − 0.000 (− 0.05) − 0.162*** (− 3.39) 0.017*** (5.13) − 0.002 (− 0.79) 0.034*** (8.42) 0.112*** (6.33) − 0.222*** (− 36.65) − 0.040 (− 0.62) − 0.007 (− 0.39) 0.032*** (5.46) − 0.230*** (− 3.38) Yes 13,350 0.103 environment and weaker corporate governance. The negative relation is also more pronounced when controlling shareholder’s share pledging is under higher margin call pressure as when firms experienced high volatility in stock price within the last fiscal year and when the visits are conducted during an economic recession. Lastly, we also investigate whether the type of visiting institutions matters to inhibit controlling shareholder’s share pledging, the empirical results indicate the higher proportion of fund companies in visiting institutions that firm receives, the more controlling shareholder’s share pledging would be reduced, but the proportion of security companies in visiting institutions does not moderate the negative relation. This study contributes to the literature mainly from three aspects. First, it adds to the literature on the economic consequences of Table 11 Basic statistics on the proportions of different visiting institutions. Mean 0.063*** (3.51) (2) Table 11 reports the impact of institutional investors corporate site visits on controlling shareholders’ share pledging of the visited firm with different por­ tions of security firms (fund companies) in visiting institutions. The dependent variable PledgedShares is measured as the ratio of controlling shareholder’s pledged shares to their shareholdings in the listed corporates within fiscal year t. InvestorVisits is measured as the natural logarithm of one plus the number of institutional investors’ corporate site visits a corporate receives within fiscal year t. SecRatio (FundRatio) is defined as the number of visiting security firms (fund companies) to the number of the total visiting institutions. The variable of interests in Column (1) is the interaction term of InvestorVisits and SecRatio. The variable of interests in Column (2) is the interaction term of InvestorVisits and FundRatio. A negatively significant coefficient of the interaction term indicates that the proportion of visits done by this type of institutional investors further enhances the negative relation between institutional investors’ corporate site visits and controlling shareholder’s share pledging. All regressions are based on a standard OLS model. We control for all Control Variables and control for year and industry fixed effects to control for unobservable heterogeneity across time and across industries in all regressions. All other variables are defined in Ap­ pendix A. The t statistics reported in parentheses are based on standard errors clustered by firm. ***, **, * indicate that statistical significance at the 1%, 5% and 10% significance level respectively. 7. Conclusions SecRatio FundRatio Dep. Var. = PledgedShares Table 11 reports the basic statistics of the proportions of different visiting in­ stitutions, where SecRatio (FundRatio) is defined as the number of visiting se­ curity firms (fund companies) to the number of the total visiting institutions. p25 and p75 stand for 25th percentile and 75th percentile respectively 13 International Review of Financial Analysis 89 (2023) 102746 Z. Xiao et al. shareholders’ share pledging decisions. Third this study provides a comprehensive analysis revealing the underlying mechanisms through which institutional investors’ corporate site visits affect controlling shareholder’s share pledging, including information acquisition chan­ nel, corporate governance channel, potential margin call pressure and the roles of different types of visiting institutions. institutional investors’ corporate site visits by directly linking institu­ tional investors’ corporate site visits to controlling shareholder’s share pledging. Second, while most of the prior literature focuses on the economic consequences of controlling shareholders’ share pledging, we provide the first reliable evidence indicating institutional investors’ corporate site visits inhibit controlling shareholders’ share pledging, which extends the literature on the determinants of controlling Appendix A. Variable definitions Variable Definition The measure of dependent Variable Following Liu and Tian (2021), we construct a continuous variable, PledgedShares, as the ratio of controlling shareholder’s pledged shares to their PledgedShares shareholdings in the listed corporates within a fiscal year. The measure of independent Variable Following Jiang and Yuan (2018), we construct a continuous variable InvestorVisits measured by the natural logarithm of one plus the number of site visits InvestorVisits within a fiscal year. Alternative measures of independent Variable Investor Number Defined as the number of visiting institutions to a listed firm within a fiscal year scaled by 100. Visit Dummy Defined as a dummy variable that equals one if a sample corporate’s InvestorVisits is higher than the year-industry median value, and zero otherwise. Measures of instrumental Variable Following Wang et al. (2022), we employ a dummy variable HighSpeed, which equals one if a sample firm’s located city is connected to high-speed railway HighSpeed network in fiscal year t and zero otherwise. Following (Su et al., 2021) we next employ a continuous variable Flight, which equals one the natural logarithm of one plus the number of direct flights Flight towards the sample corporate’s located city. Measures of the information environment Following Roychowdhury (2006), We run the following regressions on different industry–year observations: CFOt = α0 + α1 (1/At− 1 ) + β1 (St /At− 1 ) + β2 (ΔSt /At− 1 ) + εt PRODt = α0 + α1 (1/At− 1 ) + β1 (St /At− 1 ) + β2 (ΔSt /At− 1 ) + β3 (ΔSt− 1 /At− 1 ) + et DISEXPt = α0 + α1 (1/At− 1 ) + α1 (1/At− 1 ) + β1 (St− 1 /At− 1 ) + ut REM Where CFO is the cash flow from operating, PROD is the cost of production, DISEXP is the discretional expenditure. Then the abnormal CFO (AbCFO) is defined as the residual εt, the abnormal production cost (AbProd) is defined as the residual et, the abnormal expenditure (AbEXP) is defined as the residual ut. Then REM is defined as |REM| = |-AbCFOt + AbPRODt − AbDISEXPt |. Following Behn and Kang (2008), analysts’ Forecast Deviation is defined as follow: ⃒ 1 ∑n ⃒⃒ Forecast Deviation ForecasteDeviationi,t = ForecastedEPSi,j,t − ActualEPSi,t ⃒. j n Where i represents sample corporate, t represents forecasting year, j represents analyst, EPS is the earning per share of a corporate at year end. Measures of corporate governance Following Ang et al. (2000), we employ total assets turnover as a measure of a corporate’s agency cost, Total assets turnover is defined as sales revenue to the Total Assets Turnover average value of total assets. Independent Director Following Fama and Jensen (1983), Independent director rate is defined as the number of independent directors on board to the number of total directors on Rate board. Measures of margin call pressure √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ )2 ∑n ( d=1 returni,d − returni,t Stock Volatilityi,t = , n is the number of trading days of stock i within fiscal year t. Return is the market capitalization weighted Stock Volatility n daily return. Macroeconomic Defined as China’s annual real GDP growth after Hodrick Prescott filter following Cogley and Nason (1995). Condition Measures of the proportion of visiting institutions’ type Sec Ratio SecRatio is defined as the number of visiting security firms to the number of the total visiting institutions. Fund Ratio FundRatio is defined as the number of visiting security firms fund companies to the number of the total visiting institutions. Measures of control variables ROA Defined as net income over total assets. Size Defined as the natural logarithm of corporates’ assets. Growth Defined as the corporates’ operating revenue growth rate. Age Defined as the natural logarithm of one plus the number of years a firm has been listed. Lev Defined as total debt over total assets. SOE Defined as an indicator variable which equals one if a corporate is a state-owned enterprise and zero otherwise. Top1 Defined as the ownership of the largest shareholder. Defined as an indicator variable which equals one if the CEO also chairs the board of directors and zero otherwise. Dual References Berkman, H., Cole, R. A., & Fu, L. J. (2010). Political connections and minorityshareholder protection: Evidence from securities-market regulation in China. Journal of Financial and Quantitative Analysis, 45(6), 1391–1417. Boone, A., & White, J. (2015). The effect of institutional ownership on firm transparency and information production. Journal of Financial Economics, 117(3), 508–533. Borochin, P., & Yang, J. (2017). The effects of institutional investor objectives on firm valuation and governance. Journal of Financial Economics, 126(1), 171–199. Bowen, R. M., Dutta, S., Tang, S., et al. (2018). Inside the “black box” of private in-house meetings. Review of Accounting Studies, 23(2), 487–527. An, H., & Zhang, T. (2013). Stock price synchronicity, crash risk, and institutional investors. Journal of Corporate Finance, 21, 1–15. Anderson, R., & Puleo, M. (2020). Insider share-pledging and equity risk. Journal of Financial Services Research, 58(1), 1–25. Ang, J. S., Cole, R. A., & Lin, J. W. (2000). Agency costs and ownership structure. The. Journal of Finance, 55(1), 81–106. Behn, B. K., & Kang, C. T. (2008). Audit quality and properties of analyst earnings forecasts. Accounting Review, 83(2), 327–349. 14 International Review of Financial Analysis 89 (2023) 102746 Z. Xiao et al. Kao, L., Chiou, J. R., & Chen, A. (2004). The agency problems, firm performance and monitoring mechanisms: The evidence from collateralized shares in Taiwan. Corporate Governance: An International Review, 12(3), 389–402. Kryzanowski, L., Li, M., Xu, S., et al. (2021). Share pledging and corporate securities fraud. Working Paper. Liu, W., & Tian, G. G. (2021). Controlling shareholder share pledging and the cost of equity capital: Evidence from China. The British Accounting Review, 54(6), 101839. Lu, X., Fung, H. G., & Su, Z. (2018). Information leakage, site visits, and crash risk: Evidence from China. International Review of Economics and Finance, 58, 487–507. Luo, Y., Wu, H., Ying, S. X., et al. (2022). Do company visits by institutional investors mitigate managerial myopia in R&D investment? Evidence from China. Global Finance Journal, 51, Article 100694. Pang, C., & Wang, Y. (2020). Stock pledge, risk of losing control and corporate innovation. Journal of Corporate Finance, 60, Article 101534. Panicker, V. S., Mitra, S., & Upadhyayula, R. S. (2019). Institutional investors and international investments in emerging economy corporates: A behavioral risk perspective. Journal of World Business, 54(4), 322–334. Pound, J. (1988). Proxy contests and the efficiency of shareholder oversight. Journal of Financial Economics, 20, 237–265. Puleo, M. R., & Kozlowski, S. E. (2021). Asymmetric information and opportunism in insider share-pledging. Managerial Finance, 47(10), 1385–1407. Qi, Z., Zhou, Y., & Chen, J. (2021). Corporate site visits and earnings management. Journal of Accounting and Public Policy, 40(4), Article 106823. Roychowdhury, S. (2006). Earnings management through real activities manipulation. Journal of Accounting and Economics, 42(3), 335–370. Saci, F., & Jasimuddin, S. M. (2021). Does the research done by the institutional investors affect the cost of equity capital? Finance Research Letters, 41, Article 101834. Singh, P. (2018). Does pledging of shares by controlling shareholders always destroy firm value? (Working paper). Su, F., Feng, X., & Tang, S. (2021). Do site visits mitigate corporate fraudulence? Evidence from China. International Review of Financial Analysis, 78, Article 101940. Wahal, S., & McConnell, J. J. (2000). Do institutional investors exacerbate managerial myopia? Journal of Corporate Finance, 6(3), 307–329. Wang, J., Liu, G., & Xiong, Q. (2020). Institutional investors’ information seeking and stock price crash risk: Nonlinear relationship based on management’s opportunistic behavior. Accounting and Finance, 60(5), 4621–4649. Wang, Q., Lai, S., Cao, X., et al. (2022). The effect of institutional investors’ site visits: Evidence on corporate cash holdings. Applied Economics, 1–15. Wang, Y. C., & Chou, R. K. (2018). The impact of share pledging regulations on stock trading and firm valuation. Journal of Banking & Finance, 89, 1–13. Wu, M., Chen, H., & Gu, N. (2022). Research on the impact of stock liquidity on equity pledge——Theoretical explanation and empirical evidence. Finance Forum, 27(3), 43–52 (in Chinese). Wu, W., Johan, A., & Rui, M. (2016). Institutional investors, political connections, and the incidence of regulatory enforcement against corporate fraud. Journal of Business Ethics, 134, 709–726. Yang, J., Lu, J., & Xiang, C. (2020). Company visits and stock price crash risk: Evidence from China. Emerging Markets Review, 44, Article 100723. Yang, X., & Ma, Z. (2022). Institutional investors’ corporate site visits and dividend payouts. International Review of Economics and Finance, 80, 697–716. Zhou, J., Li, W., Yan, Z., et al. (2021). Controlling shareholder share pledging and stock price crash risk: Evidence from China. International Review of Financial Analysis, 77, Article 101839. Zhu, B., Xia, X., & Zheng, X. (2021). One way out of the share pledging quagmire: Evidence from mergers and acquisitions. Journal of Corporate Finance, 71, Article 102120. Brickley, J. A., Lease, R. C., & Smith, C. W., Jr. (1988). Ownership structure and voting on antitakeover amendments. Journal of Financial Economics, 20, 267–291. Broadstock, D., & Chen, X. (2021). Corporate site visits, private monitoring and fraud: Evidence from China. Finance Research Letters, 40, Article 101780. Brown, L. D., Call, A. C., Clement, M. B., et al. (2015). Inside the “black box” of sell-side financial analysts. Journal of Accounting Research, 53(1), 1–47. Bushee, B. J. (1998). The influence of institutional investors on myopic R&D investment behavior. Accounting Review, 305–333. Chan, K., Chen, H. K., Hu, S., et al. (2018). Share pledges and margin call pressure. Journal of Corporate Finance, 52, 96–117. Chauhan, Y., Mishra, A. K., & Spahr, R. W. (2021). Stock pledging and firm risk: Evidence from India. Financial Management, 50(1), 261–280. Chen, A., Kao, L., & Chen, Y. K. (2007). Agency costs of controlling shareholders’ share collateral with Taiwan evidence. Review of Pacific Basin Financial Markets and Policies, 10(2), 173–191. Chen, T., Harford, J., & Lin, C. (2015). Do analysts matter for governance? Evidence from natural experiments. Journal of Financial Economics, 115(2), 383–410. Chen, X., Wan, P., & Sial, M. S. (2021). Institutional investors’ site visits and corporate social responsibility: Implications for the extractive industries. The Extractive Industries and Society, 8(1), 374–382. Cheng, Q., Du, F., Wang, B. Y., et al. (2019). Do corporate site visits impact stock prices? Contemporary Accounting Research, 36(1), 359–388. Cheng, Q., Du, F., Wang, X., et al. (2016). Seeing is believing: Analysts’ corporate site visits. Review of Accounting Studies, 21(4), 1245–1286. Cogley, T., & Nason, J. M. (1995). Effects of the Hodrick-Prescott filter on trend and difference stationary time series: Implications for business cycle research. Journal of Economic Dynamics and Control, 19(1–2), 253–278. Deren, X., & Ke, L. (2018). Share pledging by controlling shareholders and real earnings management of listed corporates. China Journal of Accounting Studies, 6(2), 109–119. Dou, Y., Masulis, R. W., & Zein, J. (2019). Shareholder wealth consequences of insider pledging of company stock as collateral for personal loans. The Review of Financial Studies, 32(12), 4810–4854. Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. The Journal of Law and Economics, 26(2), 301–325. Firth, M., Lin, C., Wong, S. M., et al. (2019). Hello, is anybody there? Corporate accessibility for outside shareholders as a signal of agency problems. Review of Accounting Studies, 24(4), 1317–1358. Francis, J., LaFond, R., Olsson, P., et al. (2005). The market pricing of accruals quality. Journal of Accounting and Economics, 39(2), 295–327. Gao, S., Cao, F., & Liu, X. (2017). Seeing is not necessarily the truth: Do institutional investors’ corporate site visits reduce hosting firms’ stock price crash risk? International Review of Economics and Finance, 52, 165–187. He, Z., Liu, B., & Zhu, F. (2022). Share pledging in China: Funding listed corporates or funding entrepreneurship?. Working Paper,. NBER Working Paper. Helwege, J., Intintoli, V. J., & Zhang, A. (2012). Voting with their feet or activism? Institutional investors’ impact on CEO turnover. Journal of Corporate Finance, 18(1), 22–37. Hobson, J. L., Mayew, W. J., & Venkatachalam, M. (2012). Analyzing speech to detect financial misreporting. Journal of Accounting Research, 50, 349–392. Jiang, L., & Bai, Y. (2022). Strategic or substantive innovation? - the impact of institutional investors’ site visits on green innovation evidence from China. Technology in Society, 68, Article 101904. Jiang, X., & Yuan, Q. (2018). Institutional investors’ corporate site visits and corporate innovation. Journal of Corporate Finance, 48, 148–168. 15