‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises

Xue Li; Jin-hui Luo; Zeyue Huang

doi:10.1080/21697213.2019.1643068

‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises

Li, Xue; Luo, Jin-hui; Huang, Zeyue 2019-01-02 00:00:00 CHINA JOURNAL OF ACCOUNTING STUDIES 2019, VOL. 7, NO. 1, 119–143 https://doi.org/10.1080/21697213.2019.1643068 ARTICLE ‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises Xue Li, Jin-hui Luo and Zeyue Huang School of Management, Xiamen University, Xiamen, China ABSTRACT KEYWORDS ‘Original sin’ suspicion; The suspicion of ‘original sin’ is an unavoidable problem in the devel- private enterprises; opment of China’s private enterprises. Using a sample of A-share corporate philanthropy; private listed ﬁrms from 2004–2016, we investigate whether and institutional environment how the suspicion of ‘original sin’ of private enterprises aﬀects their corporate philanthropy. We ﬁnd that private enterprises with the suspicion of ‘original sin’ have stronger incentives to take corporate philanthropy as a strategic tool to reduce the threat arising from the ‘original sin’. High development levels of institutional environment can attenuate the positive eﬀect of ‘original sin’ suspicion on private enterprises’ corporate philanthropy. Furthermore, we ﬁnd that the positive association between the ‘original sin’ suspicion of private enterprises and their corporate philanthropy is more pronounced when they do not have political connections and/or after their ulti- mate controllers are showed up on the Hurun Rich List. 1. Introduction Since the Reform and Opening policy, the private sector has developed vigorously from scratch and contributes greatly to China’s economic growth. But until nowadays, private enterprises are still faced with the suspicion of ‘original sin’. ‘Origin sin’ means that some private enterprises are suspected of engaging in some ‘unruly’ behaviours when they ﬁrst obtained their initial capital. Those private enterprises may not be formally recog- nized by the public (Tang, Wen, & Sun, 2017; Ye, Li, & Li, 2012). In some extreme cases, ‘original sin’ suspicion even leads to the private entrepreneurs being jailed (for example, the founder of WuMart, Zhang Wenzhong and the eﬀective controller of Greencool, Gu Chujun). Faced with the threat of not being recognized and protected formally in legitimacy, how to improve their legitimacy has become a strategic problem for these private enterprises for a long time. In other words, it has become a very important question that how those private entrepreneurs can deal with the lack of legitimacy due to ‘original sin’ suspicion. Existing studies show that the suspicion of ‘original sin’ of private enterprises will bring higher political costs to enterprises. In order to decrease those political costs, these enterprises tend to lower their transparency (Tang et al., 2017; Ye et al., 2012). CONTACT Jin-hui Luo jinhuiluo@xmu.edu.cn School of Management, Xiamen University, No. 422, Siming South Road, Xiamen, Fujian 361005, P. R. China Paper accepted by Kangtao Ye. © 2019 Accounting Society of China 120 X. LI ET AL. Speciﬁcally, the ‘original sin’ suspicion leads to that the legitimacy of the private companies cannot be recognized by the government and the public. A large number of studies in recent years have shown that, as one of demonstrations of social respon- sibility, corporate philanthropy is often strategically used by companies to achieve their political goals, such as to promote the legitimacy and gain goodwill from the public (Dai, Pan, & Feng, 2014; Du, 2015; Hornstein & Zhao, 2018; Li, Chen, & Song, 2016; Li, Wang, & Xu, 2015; Sánchez, 2000; Wang & Qian, 2011). Therefore, this paper is intended to investigate that whether corporate philanthropy is employed by private enterprises with ‘original sin’ suspicion as a strategic tool to alleviate the impact of ‘original sin’ and gain legitimacy. The following research will be conducted from two aspects: (1) how will the suspicion of ‘original sin’ aﬀect the philanthropic behaviour of private enter- prises? (2) can formal institutional environment have an impact on the relationship between ‘original sin’ suspicion and corporate philanthropy? Using a sample of A-share private listed ﬁrms between 2004 and 2016, we measure private enterprises’‘original sin’ according to whether their original property rights are transferred from state ownership to private ownership, and examine the relationship between the ‘original sin’ suspicion of private enterprises and their corporate philan- thropy. We ﬁnd that: (1) the suspicion of ‘original sin’ of ﬁrms has a signiﬁcantly positive eﬀect on corporate philanthropy. Due to public and government’s doubts about the legitimacy of enterprises and even the threat of their private property rights being takeover caused by the suspicion of ‘original sin’, private entrepreneurs with ‘original sin’ suspicion have stronger motivation to reduce or eliminate the adverse eﬀect through corporate philanthropy; (2) the positive eﬀect of ‘original sin’ suspicion on charitable donations of private enterprises is found to be weaker for ﬁrms located in regions with more developed institutions. In poorer institutional environment, private enterprises with suspicion of ‘original sin’ may face more serious threats of being liquidated and the lack of protection of their legitimacy, which may further strengthen their motivation to improve corporate legitimacy through corporate philanthropy. Our results are robust after adopting alternative variable measurements and tacking the potential endogeneity concerns. Furthermore, we also ﬁnd that the positive eﬀect of ‘original sin’ suspicion of private enterprises on corporate is more pronounced in those ﬁrms without political connections and when the entrepreneur was listed on the Hurun rich list. Therefore, our ﬁndings indicate that private enterprises with suspicion of ‘original sin’ have strong motivation to enhance the legitimacy of enterprises and protect their property rights through corporate philanthropy. Our study contributes to the literature in several ways. First, this paper contributes to the literature about ‘original sin’ suspicion of private enterprises in developing econo- mies like China. The suspicion of ‘original sin’ is an unavoidable problem in the devel- opment of private enterprises, and has triggered widespread discussion throughout the country several times. We empirically examine the impact of the suspicion of ‘original sin’ of private enterprises on corporate philanthropy and ﬁnd that the private enterprises with the suspicion of ‘original sin’ tend to strategically donate more to alleviate and oﬀset the costs caused by the lack of legitimacy, which can deepen our understanding of the outcome of suspicion of ‘original sin’ problem faced with private enterprises. Second, this paper also contributes to the literature regarding the determi- nants of corporate philanthropy. To the best of our knowledge, we are among the ﬁrst CHINA JOURNAL OF ACCOUNTING STUDIES 121 to examine the non-altruistic motivations of corporate philanthropic behaviours under the transitional institutional background from the perspective of ‘original sin’ suspicion of private enterprises. Our ﬁnding conﬁrms the strategic motivation of corporate philan- thropy, thus providing incremental empirical evidence for the research about corporate philanthropy. Third, we further analyze the moderating eﬀect of formal institutional development on the relationship between the suspicion of ‘original sin’ of private enterprises and corporate philanthropy, showing that the formal institutions can have a profound eﬀect on behaviours of the enterprises, especially private enterprises. 2. Literature review and hypotheses development 2.1. Literature review 2.1.1. ‘Original sin’ suspicion of private enterprises Under the background of China’s emerging and transitional economy, the transforma- tion of institutions and the establishment of new rules are a normal situation faced by enterprises. Under such background, due to the imperfect institution system and weak implementation, some private enterprises inevitably have some unruly or even illegal operational behaviours in the development of their enterprises, which may lead to ‘original sin’ suspicion from the public and the government. Such suspicion would have a profound impact on enterprise behaviours. ‘Original sin’ suspicion mainly refers to the suspicion from social public to the compliance and legality of private entrepre- neurs’ initial wealth accumulation. Such suspicion could lead to the problem of private enterprise legitimacy, and thus they could not get the acceptance of the public and the government department and legal protection. Therefore, the private entrepreneurs suﬀer the threats of higher liquidation, even the deprivation of property rights and legal sanction (Tang et al., 2017; Ye et al., 2012). In recent years, some scholars have paid attention to the possible economic con- sequences of the suspicion of ‘original sin’ of private enterprises. Using the event of the entrepreneurs of the listed ﬁrms ranking on the Hurun Rich List, Ye et al. (2012) suggested that the suspicion of ‘original sin’ would bring more attention and question from public to the rich, and they may suﬀer higher threat of being liquidated resulting from ‘original sin’ suspicion. It makes the rich face a sudden increase in political costs after being listed. So the billionaires with ‘original sin’ have stronger motivation to reduce the political cost by reducing the quality of accounting information. Similarly, from perspective of political costs, Tang et al. (2017) argued that the private enterprises with ‘original sin’ suspicion have stronger motivation to reduce the political costs they bear by reducing the enterprise information transparency. 2.1.2. Motivations of corporate philanthropy As an important and explicit form of corporate social responsibility, corporate philanthropy has always attracted academic attention. According to FASB, corporate philanthropy is an uncon- ditional, voluntary and nonreciprocal transfers of cash and other assets to other entities such as non-proﬁt organizations and government agencies. This deﬁnition is in line with the altruistic motivation of corporate philanthropy identiﬁed by previous literature (Campbell, Gulas, & 122 X. LI ET AL. Gruca, 1999;Xiu &Zhou, 2016). In addition to altruistic motives, there are also self-interested motivations for corporate philanthropic giving, which can be classiﬁed into three types: (1) economic motivation, (2) managerial self-interest motivation, and (3) political and institutional motivation (Feng & Cheng, 2010;Lietal., 2016;Petrenko, Aime,Ridge, & Hill, 2016;Wang, Li, & Li, 2015; Zhang, Rezaee, & Zhu, 2010; Zhang, Zhu, Yue, & Zhu, 2010). Many studies have supported the economic motivation of corporate philanthropy, which suggests that one purpose of corporate philanthropy is to maximize economic beneﬁts. Philanthropic giving can create an advertising eﬀect that is helpful for the ﬁrm to establish positive image and reputation, leading to better achievement of both economic and social goals (Brammer & Millington, 2005; Lev, Petrovits, & Radhakrishnan, 2010; Pan, Wen, & Liu, 2017; Peng & Fan, 2016; Porter & Kramer, 2002; Shan, Gan, & Zheng, 2008; Wang, Jin, & Jiao, 2014; Zhang et al., 2010; Zheng & Xu, 2011). In addition, corporate philanthropy can be employed as a means to cope with negative events by shifting public attention away from bad news (Shiu & Yang, 2017), window- dressing poor performance from another aspects of corporate social responsibility (Chen, Patten, & Roberts, 2008; Du, 2015; Gao, Chen, & Zhang, 2012; Luo, Kaul, & Seo, 2018), attenuating risk perception of shareholder about performance decline, mitigating negative market reaction for earning restatements (Koehn & Ueng, 2010; Li, Chen, & Song, 2016), reducing litigation risks (Dai, Peng, & Pan, 2016; Fu & Ji, 2017), and even helping the ﬁrm to obtain a more favourable audit opinion (Wang & Pan, 2018). Second, corporate philanthropy can also be driven by managerial self-interest motivation (Haley, 1991). Jensen and Meckling (1976) suggested that corporate philanthropy can provide private beneﬁts to corporate managers, even though those beneﬁts may be at the expense of shareholders’ interest. Those private beneﬁts can be both pecuniary, such as higher compensation, and non-pecuniary, such as social recognition and higher reputa- tional status (Galaskiewicz, 1997; Jia & Zhang, 2010; Petrenko et al., 2016). Opportunistic behaviours of managers can also lead to higher philanthropic contributions (Petrenko et al., 2016;Tan & Xu, 2015). Finally, corporate philanthropy may also be politically driven, especially in developing economies with higher policy uncertainty (Du, Guo, & Lei, 2010; Gautier & Pache, 2015; Li, Lu, &Song, 2012; Tang, Zuo, & Li, 2014; Wang & Qian, 2011). By coordinating with the government’s scheme and sharing the costs, companies can quickly gain goodwill and trust from the government, which may help ﬁrms to gain more political support and legitimacy (Hornstein & Zhao, 2018;Sánchez, 2000; Su & He, 2010; Tang et al., 2014;Wang &Qian, 2011) and ultimately helps the ﬁrm to obtain more economic resources such as subsidies and loan credits (Dai et al., 2014;Gao, Faﬀ,&Navissi, 2012; Jia & Zhang, 2010;Li et al., 2015; Li, Song, & Wu, 2015; Li, Tang, & Lian, 2016; Li, Lu, & Song, 2012; Su & He, 2010; Wang, Pan, & Huang, 2016; Wang & Qian, 2011; Zhang, Ma, & Zhang, 2013). Overall, there is a consensus in the literature that political motivation is very impor- tant for ﬁrms to conduct corporate philanthropy, especially in transitional economies like China, where many private enterprises are facing with ‘original sin’ suspicion. However, there is little research regarding whether and how the suspicion of ‘original sin’ may aﬀect the corporate behaviour of private enterprises. This paper intends to investigate the possible impact of the ‘original sin’ suspicion of private enterprises on corporate philanthropy, thereby contributes to the literature by deepening our understanding of the economic consequences of the ‘original sin’ suspicion. CHINA JOURNAL OF ACCOUNTING STUDIES 123 2.2. Hypotheses development Under the threat that enterprise may not be formally recognized and eﬀectively pro- tected because of the suspicion of ‘original sin’, a rational entrepreneur will certainly take some measures to alleviate this threat and any disadvantages resulting from it. We argue that corporate philanthropy is likely to be a strategic tool for those private enterprises suspected of ‘original sin’ to obtain and enhance their legitimacy. Firstly, enterprises can obtain the recognition and protection from government departments through charitable donations. As a safe way for enterprises to obtain their legitimacy, corporate philanthropy can help enterprises suspected of ‘original sin’ win favor and recognition from the government at a faster speed, and achieve the purpose of obtaining and improving their own legitimacy and avoiding risks by seeking political supports and protection (Dai et al., 2014; Hornstein & Zhao, 2018; Jia, Xiang, & Zhang, 2015; Sánchez, 2000; Su & He, 2010; Tang et al., 2014; Wang & Qian, 2011). Secondly, corporate philanthropy is also conducive to improving the public’s recognition of enterprises. In China, the public has a long-standing prejudice against private enter- prises with suspicion of ‘original sin’, namely the phenomenon of ‘hatred for the rich’. Besides improving social welfare, corporate philanthropy can help to establish a good image of the enterprise in the public, so as to alleviate or even eliminate the prejudice of the public towards the private enterprises (Guo, 2012; Li & Xie, 2014; Wang & Pan, 2018; Wang & Qian, 2011). Therefore, it is helpful for the private enterprises to enhance the public’s recognition of the legitimacy of enterprises. In addition, corporate philanthropy is conducive for enterprises to accumulate moral capital and plays a role of ‘reputation insurance’ (Fu & Ji, 2017; Gao, Chen, & Zhang, 2012; Shiu & Yang, 2017), which con- tributes to reduce the risks and the regulatory and social sanctions caused by the suspicion of ‘original sin’. Base on the theoretical analysis above, we argue that private enterprises with suspi- cion of ‘original sin’ will have stronger motivation to engage in corporate philanthropy so as to alleviate or eliminate the doubts on the legitimacy of enterprises raised by the public and government departments. Thus, we put forward our ﬁrst hypothesis. Hypothesis 1: Ceteris paribus, ‘original sin’ suspicion will increase the level of charitable donations of private enterprises. In fact, the suspicion of ‘original sin’ is mainly due to the imperfect of formal institutions in the process of China’s economic transformation. It means that the weak institutional environment leads to the suspicion of ‘original sin’ and brings about a series of economic consequences. Therefore, we argue that the external institutional environment is an important potential factor regulating the relationship between suspicion of ‘original sin’ of private enterprises and their corporate philan- thropy. In China, although enterprises face the same legal system and political system, there are great diﬀerences in the institutional environment among diﬀerent regions (Fan, Wang, & Zhu, 2011). Generally, a more developed institutional environment means lower government intervention and better legal protection (Li, Tang, & Lian, 2016; Tang et al., 2017). For private enterprises suspected of ‘original sin’ located in 124 X. LI ET AL. regions with more developed institutional environment, the situation that their legiti- macy cannot be formally recognized and their property right protection will be improved to a large extent. At this time, the motivation of enterprises to minimize the adverse eﬀect of ‘original sin’ suspicion through charitable donations will be signiﬁcantly weakened. However, for private enterprises located in regions with less developed institutions, due to weak protection of enterprises by law and more administrative intervention of the government on enterprises (Luo, Xiang, & Zhu, 2017; Tang et al., 2017), the legitimacy and property rights of enterprises are diﬃcult to be oﬃcially recognized and eﬀectively protected by law. Thus, private enterprises with ‘original sin’ suspicion have stronger motivation to engage in corporate philan- thropy to obtain and strengthen their legitimacy status accordingly. Besides, for ﬁrms located in regions with more developed formal institutions, they would obtain the initial property of their enterprises in a market-oriented way, a more transparent and standard way (Tang et al., 2017). It means that these enterprises would gain wider recognition from the public and they would be less likely to be suspected with ‘original sin’. Therefore, the enterprises’ strategic motives for charitable giving would be largely weakened. Thus, we propose our second hypothesis. Hypothesis 2: Institutional environment will weaken the positive eﬀect of ‘original sin’ suspicion on corporate philanthropy of private enterprises. 3. Research design 3.1. Sample and data Our simple includes 1,288 private listed ﬁrms controlled by individual or family from 2004 to 2016. We choose 2004 as the beginning of our simple because that the China Security Regulatory Commission required listed ﬁrms to report details about the ultimate controllers since 2003, and some variables used in the subsequent research of our study (such as ﬁrms’ growth opportunities measured by the growth rate of operating income) need to use the data on one preceding year. After excluding 23 listed ﬁrms whose ultimate controller transferred from the government to family or individual and seven ﬁrms whose ultimate controller are diﬃcult to judge and those are fail to disclose company predecessor, our initial sample comprises 1,258 ﬁrms and 8,173 observations. To alleviate the potential inﬂuence of abnormal samples, we exclude ﬁrms that are specially treated (ST or *ST) (72 observations), that are in ﬁnancial industries (11 observations), that issue debt exceeding asset value (4 observations), that are cross listed (29 observations), that have missing data (1,542 observations). Our ﬁnal sample includes 6,515 ﬁrm-year observations. Table 1 provides the distribution of industry and year of our sample. Our data on ﬁrms’ original property rights were obtained manually from historical evolution of the ﬁrms and the information of ultimate controllers shown in ﬁrms’ prospectus which were disclosed on CNINFO (www.cninfo.com.cn). The ﬁnancial data and corporate governance data are from China Stock Market and Accounting Research (CSMAR) database. The data we use to calculate formal institutional environment was from National Bureau of Statistics (www.stats.gov.cn). CHINA JOURNAL OF ACCOUNTING STUDIES 125 Table 1. Sample distribution by industry and year. Industry Year Total 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 A 0234567 12 14 16 17 17 17 120 B 010011268 88 9 9 53 C0 1112559 18 28 31 32 37 37 207 C1 2 8 8 10 12 13 17 25 29 32 32 36 36 260 C2 012223346 77 8 8 53 C3 021569 13 15 18 19 19 20 20 147 C4 4 8 8 12 17 23 31 70 100 108 111 119 117 728 C5 2369 15 18 22 46 60 67 68 71 72 459 C6 1335 13 16 20 46 58 66 68 73 72 444 C7 5 16 18 25 41 51 84 166 230 264 276 302 301 1,779 C8 1 8 8 12 14 17 27 48 61 65 66 71 69 467 C9 000112356 68 9 8 49 D 000001133 66 6 6 32 E 123567 10 18 25 27 28 29 28 189 F 000001345 56 7 7 38 G 3 5 5 11 22 24 51 88 115 131 139 144 141 879 H 0333369 17 22 24 24 26 25 165 J 122255788 77 7 8 69 K 011199 16 27 33 35 37 42 41 252 L 0002348 14 16 18 19 21 20 125 Total 21 66 72 111 180 221 343 640 845 942 978 1,054 1,042 6,515 Notes: Industry code A-M correspond, respectively: Agriculture, Forestry, Husbandry and Fishery (A); Mining (B); Food and Drink (C0); Textile, Clothing and Fur (C1); Wood and Furniture (C2); Papermaking and Printing (C3); Petroleum, Chemical and Plastic (C4); Electronics (C5); Metal and Nonmetallic Industry (C6); Machinery, Equipment and Instrument (C7); Medicine and Biological products (C8); Other manufacturing Industry (C9); Electricity, Gas and Water (D); Construction Industry (E); Transportation and Warehousing (F); Information and Technology (G); Wholesale and Retail (H); Real Estate (J); Social Services (K); Communication and Culture Industry (L); Comprehensive Industry (M). 3.2. Measurement 3.2.1. Dependent variables Following prior literature (Brown, Helland, & Smith, 2006; Du et al., 2010;Xu& Li, 2016;Zhang et al., 2013), we adopt the two proxies, Donation1 and Donation2, to measure the dependent variable, corporate philanthropy. Donation1 is the amount of donation divided by operating revenue, Donation2 is the amount of donation divided by total asset. Especially, referring to existed literature (Dai et al., 2014; Du et al., 2010;Masulis &Reza, 2015)the missing value of Donation1 and Donation2 is assigned to zero. 3.2.2. Independent variable As to the deﬁnition of ‘original sin’, the existing literature has not reached a consensus (Ye et al., 2012). In this paper, ‘original sin’ mainly refers to that private enterprises may be subjected to the suspicion of illegal and inappropriate deeds during the initial privatization process. In other words, the ‘ﬁrst pot of gold’ that the private enterprises need to take oﬀ may not be lawfully acquired. Therefore, following Tang et al. (2017), we use original evolution of property rights of private enterprises as the proxy of ‘original sin’ suspicion. Speciﬁcally, we construct the variable OriginalSin. According to the information about company history disclosed in the prospectus of listed companies, if a private enterprise was formerly state-owned and later privatized through methods like equity transfer before they were listed, then it is deﬁned as the privatized ﬁrms. Otherwise, if a private company was founded by a natural person at the beginning of its establishment, it is identiﬁed as de novo private ﬁrms. The variable OriginalSin equals to 1 for privatized ﬁrms and 0 for de novo 126 X. LI ET AL. private ﬁrms. Compared with the de novo private ﬁrms, the privatized ﬁrms may suﬀer suspicion of original sin from public due to the opaque transformation process from state- owned property rights to privately owned property rights, resulting severer suspicion from the public and the government concerning that they may gain their property rights through illegal or inappropriate ways that may have led to loss of state assets. 3.2.3. Moderating variables The moderating variable is the development of formal institutions. In the most existing literature, many studies took the NEBI’s marketization index to measure it (Fan et al., 2011). However, the marketization index issued by Fan et al. (2011) only updated to 2009, and the marketization index of regions from 2008–2014 issued by Wang, Fan, and Yu (2017)has alarge diﬀerence with Fan et al. (2011). As a result, these two indexes of diﬀerent sources cannot be combined simply to measure the development of formal institutions. Given that, following previous research (Wang & Qian, 2011), we computed the ratio of a province’sGDP to its government’s budget to measure the development of formal institutions (MKT). In addition, we also took NEBI’s marketization index for the sake of a robustness check, in which the value of NEBI’s marketization indexes between 2010 and 2016 was replaced by the index value of 2009. 3.2.4. Control variables As with previous corporate philanthropy research (Du, Jian, Du, Feng, & Zeng, 2014;Gao et al., 2012; Li et al., 2015;Luo et al., 2017; Wang & Qian, 2011;Xu & Li, 2016), we controlled other factors that may systematically relate to corporate philanthropy. We included MKT (the ratio of a province’s GDP to its government’s budget), political connection (PC, a dummy variable that takes value of one if the chairman/CEO of the ﬁrms is or was agovernment oﬃcial, delegate of the people’s congress, or member of the political consultative conferences, and zero otherwise), ﬁrm size (SIZE, measured as the natural logarithm of the total assets of the t year), ﬁnancial leverage (LEV, measured as the ratio of total debt to total assets of the t year), return on assets (ROA, measured as the ratio of net income divided by the average total assets of the t year), sales growth (GROWTH, measured as the change ratio in operating income t year to t-1 year), cash ﬂow from operating (CFO, measured as the ratio of operating cash ﬂow to operating income of the t year), ﬁrm age (AGE, measured as the natural logarithm of the time span from the established date of the ﬁrm plus 1), ownership of ultimate owners (SHARE, measured as the sum of the products of all equity stakes held by ultimate owners along the control chains), the separation between ultimate owners control and ownership (WEDGE, measured as the ratio of ultimate owners voting rights to cash ﬂow rights). Finally, we created 21 industry dummies (eliminating the ﬁnancial industry) according to CSRC’s Guidelines for Classiﬁcation of Listed Companies to control industry eﬀects and 13 year-dummies for 2004 to 2017 to control year eﬀects. 3.3. Regression models We estimate the following models: X X X Donation ¼ α þ α Original Sin þ Control þ Industry þ Year þ ε (1) t 0 1 t1 t CHINA JOURNAL OF ACCOUNTING STUDIES 127 Donation ¼ β þ β OriginalSin þ β OriginalSin MKT þ Control t t1 t1 0 1 2 X X (2) þ Industry þ Year þ δ where Donation refers to corporate charitable giving, that is, Donation1 and Donation2. OriginSin refers to ‘original sin’ suspicion. Control refers to a serious of variables that may aﬀect corporate philanthropy. We mainly use the Tobit regression model in our analysis because the dependent variable Donation was censored at zero. Model (1) is designed to examine the eﬀect of ‘original sin’ suspicion on corporate philanthropy. The coeﬃcient of OriginalSin in model (1) measures the eﬀect of ‘original sin’ suspicion on corporate philanthropy. According to Hypothesis 1, α is expected to be signiﬁcantly positive. Model (2) is designed to examine the moderating eﬀect of formal institution on the relationship between ‘original sin’ suspicion and corporate philanthropy. According to Hypothesis 2, β is expected to be signiﬁcantly negative. All continuous variables are winsorized at the 1% and 99% level. 4. Results 4.1. Descriptive statistics Table 2 reports the summary statistics for the main variables. The mean value of Donation1 (Donation2) is 0.045 (0.022), and the minimum value and maximum value are 0 (0) and 0.556 (0.233) respectively, indicating that there is a large gap in the participation degree of charitable donations among diﬀerent enterprises. OriginalSin has a mean value of 0.268, indicating that 26.8% of the samples obtained their initial property rights through state- owned transformation, while the remaining 73.2% of our samples were founded mainly by the ultimate controllers themselves. The standard deviation of MKT is 1.770, and the maximum and minimum values are 10.799 and 2.751, respectively. It can be seen that the formal institution in diﬀerent regions of China is quite diﬀerent. The mean value of PC is 0.411, which means that more than 40% of the samples have political connections. The average SHARE value is 0.362, indicating that the ownership of ultimate controllers in private listed companies is generally high. The mean value of WEDGE is 1.200, indicating that private listed companies adopt pyramid controlling structure to enhance their control over com- panies. In addition, the rest control variables are reasonably distributed. Table 2. Descriptive statistics. Variables N Mean S.D. Min P25 Median P75 Max Donation1 6,515 0.045 0.090 0 0 0.000 0.045 0.556 Donation2 6,515 0.022 0.041 0 0 0.005 0.024 0.233 OriginalSin 6,515 0.268 0.443 0 0 0 1 1 MKT 6,515 6.679 1.770 2.751 5.045 7.238 7.893 10.799 PC 6,515 0.411 0.492 0 0 0 1 1 SIZE 6,515 21.449 0.87 19.773 20.808 21.352 22.011 23.934 LEV 6,515 0.340 0.184 0.035 0.190 0.321 0.472 0.799 ROA 6,515 0.055 0.051 −0.110 0.024 0.051 0.082 0.227 GROWTH 6,515 0.220 0.367 −0.436 0.013 0.157 0.334 2.079 CFO 6,515 0.080 0.149 −0.433 0.006 0.077 0.157 0.542 AGE 6,515 2.441 0.417 1.386 2.197 2.485 2.708 3.219 SHARE 6,515 0.362 0.160 0.067 0.239 0.346 0.475 0.756 WEDGE 6,515 1.200 0.482 0.486 1 1 1.241 3.677 128 X. LI ET AL. 4.2. Pearson correlation coeﬃcients Table 3 reports the Pearson correlation coeﬃcient among the main variables. Donation2 is positively related to OriginalSin at the 1% signiﬁcance level, supporting Hypothesis 1, while there is no positive and signiﬁcant correlation between Donation1 and suspicion of ‘original sin’. ROA, CFO, MKT and PC are all signiﬁcantly positively correlated with Donation1/Donation2, indicating that private enterprises with better performance, more abundant cash ﬂow, political connections and located in regions with more developed institutions tend to make more charitable donations. There is a positive and signiﬁcant correlation between OriginalSin and PC, which means that private enterprises with suspicion of ‘original sin’ are more inclined to establish political relations with the government in order to obtain government support to cope with external uncertainties and improve their legitimacy. OriginalSin is signiﬁ- cantly positively related to SHARE,andsigniﬁcantly negatively related to WEDGE, suggesting that the ﬁrms with suspicion of ‘original sin’ tend to use the separation of ownership and control to disperse the economic risks caused by the suspicion of ‘original sin’. In addition, the correlation coeﬃcient between two pairs of other variables is lower than 0.5, which means that introducing them into the regression model will not cause serious multicollinearity problems. 4.3. Univariate comparisons Table 4 shows the univariate comparisons of the main variables between enterprises with (OriginalSin = 1) and without (OriginalSin = 0) suspicion of ‘original sin’ in several ﬁrm characteristics. As Table 4 shows, ﬁrms with ‘original sin’ suspicion commit less resource to corporate philanthropy compared with their counterparts, which generally Table 3. Pearson correlation matrix. Variables 1234567 Donation1 1 Donation2 0.889*** 1 OriginalSin 0.009 0.032*** 1 MKT 0.043*** 0.110*** 0.087*** 1 PC 0.053*** 0.079*** 0.079*** 0.052*** 1 SIZE −0.054*** −0.031** 0.071*** −0.118*** 0.093*** 1 LEV −0.110*** −0.056*** 0.065*** 0.104*** 0.061*** 0.486*** 1 ROA 0.079*** 0.165*** −0.030** 0.005 −0.010 0.007 −0.309*** GROWTH −0.010 0.019 −0.057*** −0.040*** −0.042*** 0.135*** 0.108*** CFO 0.061*** 0.048*** 0.034*** −0.016 −0.003 −0.031** −0.241*** AGE −0.023* −0.041*** 0.053*** −0.230*** −0.039*** 0.173*** 0.020 SHARE 0.037*** 0.040*** −0.194*** −0.075*** 0.008 −0.004 −0.122*** WEDGE −0.041*** −0.016 0.214*** 0.198*** −0.035*** −0.012 0.097*** Variables 8 9 10 11 12 13 ROA 1 GROWTH 0.286*** 1 CFO 0.304*** −0.032*** 1 AGE −0.052*** −0.030** 0.043*** 1 SHARE 0.113*** 0.010 0.004 −0.065*** 1 WEDGE −0.047*** −0.040*** 0.003 0.002 −0.498*** 1 Notes: ***, **, and * indicate signiﬁcant at 1%, 5%, 10%, respectively, two-tailed. CHINA JOURNAL OF ACCOUNTING STUDIES 129 Table 4. Univariate comparisons. Variables OriginalSin =1 (N = 1747) OriginalSin =0 (N = 4768) T value Z value Mean Median SD Mean Median SD Donation1 0.046 0.013 0.087 0.045 0.008 0.092 0.723 5.108*** Donation2 0.025 0.008 0.042 0.022 0.004 0.041 2.584*** 6.093*** MKT 6.933 7.616 1.864 6.586 6.663 1.726 7.022*** 7.187*** PC 0.476 0 0.500 0.388 0 0.487 6.421*** 6.401*** SIZE 21.551 21.451 0.854 21.411 21.314 0.872 5.749*** 6.122*** LEV 0.360 0.344 0.187 0.333 0.314 0.183 5.271*** 4.980*** ROA 0.053 0.049 0.049 0.056 0.052 0.052 −2.439** −2.686*** GROWTH 0.186 0.130 0.349 0.233 0.168 0.373 −4.604*** −5.752*** CFO 0.089 0.083 0.142 0.077 0.075 0.152 2.707*** 2.911*** AGE 2.477 2.565 0.435 2.427 2.485 0.409 4.313*** 4.777*** SHARE 0.311 0.294 0.157 0.381 0.370 0.157 −15.931*** −16.173*** WEDGE 1.371 1.001 0.620 1.138 1 0.404 17.646*** 12.733*** Notes: ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed. support our Hypothesis. Firms with suspicion of ‘original sin’ are more likely to establish political connection compared with those without suspicion of ‘original sin’, which is consistent with the results in Table 3. In addition, there are signiﬁcant diﬀerences in the size, debt, proﬁtability, growth, cash ﬂow level, ﬁrm age, and ownership structure between ﬁrms with suspicion of ‘original sin’ and their counterparts. Considering that the ﬁrm characteristics can aﬀect the corporate philanthropy, it is important to introduce these variables into the regression model in subsequent multiple regression analysis. 4.4. Multivariate regression tests Table 5 reports the regression results of our hypotheses. As Model 1 and Model 3 of Table 5 shows, OriginalSin gets positive and signiﬁcant coeﬃcient (Model 1: β = 0.008, T = 2.436; Model 3: β =0.005, T = 3.070), indicating that private enterprises with suspicion of ‘original sin’ would prefer to donate more in order to gain support from government, and improve the public recognition to them, thus obtain and enhance their legitimacy. In short, Hypothesis 1 is supported. Compared with their counterparts, ﬁrms with suspicion of ‘original sin’ contribute 0.8% higher amount of revenue to corporate philanthropy on average, accounting for 17.78% (= 0.008/0.045) of overall amount of charitable donations in our samples. It highlights the economic signiﬁcance of the inﬂuences of suspicion of ‘original sin’ on corporate philanthropy. In Model 2 and Model 4, the coeﬃcients of the interaction OriginalSin*MKT are signiﬁcantly negative at least at the 10% level (Model 2: β = −0.004, T = −2.170; Model 4: β = 0.001, T = 1.709). It means that the enterprises are more likely to be recognized and protected with the improvement of the formal institutions. Thus, private enterprises with suspicion of ‘original sin’ have weaker motivation to gain legitimacy through charitable contributions. Thus, Hypothesis 2 is supported. In terms of control variables, the coeﬃcients of PC and SIZE are positive and signiﬁcant, consistent with the results of Gao et al. (2012) and Luo et al. (2017), indicating that larger ﬁrms and ﬁrms with political connections have more corporate philanthropy. However, the coeﬃcients of LEV and WEDGE are negative and signiﬁcant, suggesting that private enter- prises with higher leverage and separation of ownership and control are less inclined to conduct charitable donations, which is basically consistent with our theoretical expectation. 130 X. LI ET AL. Table 5. Results of multivariate regression. Variables Donation1 Donation2 Model 1 Model 2 Model 3 Model 4 OriginalSin 0.008** 0.035*** 0.005*** 0.014** [2.436] [2.723] [3.070] [2.505] OriginalSin × MKT −0.004** −0.001* [−2.170] [−1.709] MKT 0.003** 0.004*** 0.002*** 0.003*** [2.464] [3.046] [4.552] [4.703] PC 0.014*** 0.014*** 0.007*** 0.007*** [4.725] [4.565] [5.564] [5.442] SIZE 0.014*** 0.014*** 0.005*** 0.005*** [6.478] [6.505] [5.625] [5.647] LEV −0.065*** −0.064*** −0.010** −0.009** [−5.977] [−5.847] [−2.073] [−1.968] ROA 0.049 0.048 0.133*** 0.132*** [1.350] [1.333] [7.891] [7.881] GROWTH −0.002 −0.002 −0.001 −0.001 [−0.342] [−0.355] [−0.563] [−0.575] CFO 0.011 0.011 −0.003 −0.003 [0.907] [0.870] [−0.699] [−0.737] AGE 0.008** 0.009** 0.005*** 0.005*** [2.120] [2.283] [2.808] [2.938] SHARE 0.003 0.003 0.005 0.005 [0.265] [0.323] [1.035] [1.081] WEDGE −0.016*** −0.016*** −0.007*** −0.007*** [−4.349] [−4.292] [−3.719] [−3.671] Constant −0.302*** −0.314*** −0.148*** −0.152*** [−6.318] [−6.523] [−6.783] [−6.934] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 6,515 6,515 6,515 6,515 F value 10.654 10.415 14.015 13.718 Log likelihood 2607.669 2610.170 6418.163 6419.779 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. 4.5. Robustness test 4.5.1. Robustness test of the measurement of formal institution In the above analysis, we use the ratio of GDP to government budget to measure the formal institution. To avoid the possible eﬀect of the measurement of variables on the results of regression analysis, we also use the marketization index issued by Fan et al. (2011) as the proxy variable of the formal institution. The results are shown in Table 6.As seen in Table 6, both coeﬃcients on OriginalSin in Model 1 and Model 3 are positive at 1% signiﬁcance level. In Model 2, the interaction OriginalSin*MKT gets a negative coeﬃ- cient at 10% signiﬁcance level, and the interaction OriginalSin*MKT in Model 4 is marginally signiﬁcant and negative (p value = 0.16). Our hypotheses are still well supported and the results are robust. 4.5.2. Robustness test of endogenous problems 4.5.2.1. Propensity score matching. This study examines the eﬀect of the suspicion of ‘original sin’ on corporate philanthropy of private enterprises. However, there may be a problem of sample selection bias if we only directly make a comparison between private enterprises with suspicion of ‘original sin’ and their counterparts. Thus, we CHINA JOURNAL OF ACCOUNTING STUDIES 131 Table 6. Regression results for the alternative measurement of formal institutions. Donation1 Donation2 Variables Model 1 Model 2 Model 3 Model 4 OriginalSin 0.008*** 0.040** 0.005*** 0.017* [2.599] [2.099] [3.340] [1.955] OriginalSin × MKT −0.003* −0.001 [−1.716] [−1.406] MKT 0.003*** 0.004*** 0.002*** 0.002*** [2.931] [3.303] [4.862] [4.797] PC 0.015*** 0.014*** 0.008*** 0.008*** [4.864] [4.729] [5.787] [5.682] SIZE 0.013*** 0.013*** 0.005*** 0.005*** [6.418] [6.369] [5.534] [5.492] LEV −0.065*** −0.063*** −0.009* −0.009* [−5.927] [−5.777] [−1.958] [−1.831] ROA 0.044 0.043 0.129*** 0.128*** [1.210] [1.168] [7.652] [7.611] GROWTH −0.002 −0.002 −0.001 −0.001 [−0.390] [−0.421] [−0.662] [−0.689] CFO 0.011 0.01 −0.003 −0.003 [0.885] [0.832] [−0.718] [−0.771] AGE 0.007* 0.008* 0.004** 0.004** [1.887] [1.952] [2.423] [2.472] SHARE 0.002 0.003 0.005 0.005 [0.240] [0.337] [1.001] [1.083] WEDGE −0.016*** −0.016*** −0.006*** −0.006*** [−4.294] [−4.297] [−3.569] [−3.571] Constant −0.312*** −0.323*** −0.154*** −0.158*** [−6.493] [−6.694] [−7.062] [−7.238] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 6,515 6,515 6,515 6,515 F value 10.842 10.618 14.243 14.005 Log likelihood 2609.305 2610.992 6419.985 6421.100 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. further adopt propensity score matching method to address this problem. Speciﬁcally, we conduct PSM method on ‘original sin’ suspicion to examine whether ﬁrms with ‘original sin’ suspicion donate more. In the ﬁrst stage of propensity score matching, we choose SIZE, LEV, ROA, AGE, GROWTH, SHARE and WEDGE, PC, MKT, Industry and Year as matching variables to calculate propensity scores. Considering that the some eco- nomic factors in the region where the enterprise is located may also aﬀect the possibility of their ‘original sin’ suspicion, we further add local government ﬁscal deﬁcit rate and unemployment rate t as exogenous matching variables. To ensure the reliability of the matching results, we conduct an equilibrium test and a common support hypothesis test for the matching variables. Table 7 shows the result of the equilibrium test for PSM. As is shown in Table 7, there is a signiﬁcant diﬀerence between the treated group and the control group before matching. After matching, the diﬀerences between two groups are no longer signiﬁcant, indicating that our matching variables are balanced between the treated group and the control group after matching and these variables no longer have an explanatory power for the possibility of ‘original sin’ suspicion. The matching satisﬁes the equilibrium hypothesis. In order to test whether the common support hypothesis is satisﬁed between the treated group and the control group, we further report the probability density distribution of the 132 X. LI ET AL. Table 7. Test for common support during the procedure of PSM. Variable Sample Mean Deviation Rate Deviation reduction T test ratio(%) Treated group Control group T value p>|t| SIZE Before 21.551 21.408 16.5 82.8 5.86 0.000 After 21.551 21.526 2.8 0.84 0.402 LEV Before 0.360 0.333 14.8 79.3 5.30 0.000 After 0.360 0.366 −3.1 −0.90 0.369 ROA Before 0.053 0.056 −6.9 82.6 −2.41 0.016 After 0.0537 0.052 1.2 0.36 0.719 AGE Before 2.477 2.428 11.6 81.9 4.21 0.000 After 2.477 2.469 2.1 0.64 0.520 GROWTH Before 0.185 0.233 −13.2 82.1 −4.64 0.000 After 0.185 0.177 2.4 0.77 0.443 SHARE Before 0.311 0.381 −44.9 92.4 −16.05 0.000 After 0.311 0.316 −3.4 −1.01 0.315 WEDGE Before 1.371 1.136 44.9 84.9 17.75 0.000 After 1.371 1.406 −6.8 −1.63 0.104 MKT Before 6.933 6.5726 20.0 91.0 7.28 0.000 After 6.933 6.900 1.8 0.53 0.594 PC Before 0.476 0.386 18.2 59.8 6.53 0.000 After 0.476 0.440 7.3 2.14 0.032 DEFICIT Before 0.533 0.453 17.2 86.0 6.28 0.000 After 0.533 0.544 −2.4 −0.68 0.495 JOBLESS Before 0.032 0.030 35.4 94.4 11.84 0.000 After 0.032 0.032 2.0 0.68 0.499 0 .2 .4 .6 .8 0 .2 .4 .6 .8 Propensity Score Propensity Score Treat Control Treat Control (1) Before Matching (2) After Matching Figure 1. Probability density distribution of propensity scores of the sample. propensity scores of the treated group and the control group before and after matching. As is shown in Figure 1, the propensity score of the treated group before matching is signiﬁcantly larger than that of the control group. After matching, the propensity score distribution of the treated group shifts to the right, and the distribution pattern of the treated group and the control group are matching well, indicating that PSM corrects the distribution bias of propensity scores of two groups. The matching satisﬁes the common support hypothesis. Table 8 reports the results of the regression analysis after using PSM. As Table 8 shows, the regression coeﬃcient on OriginalSin is positive at 1% signiﬁcance level in Model 1 and Model 3. (Model 1: β = 0.012, T = 2.920; Model 3: β = 0.006, T = 3.538). In Model 2 and Model 4, the regression coeﬃcients on the interaction OriginalSin*MKT are Density 0 1 2 3 Density 0 .5 1 1.5 2 2.5 CHINA JOURNAL OF ACCOUNTING STUDIES 133 Table 8. Regression results of PSM method. Donation1 Donation2 Variables Model 1 Model 2 Model 3 Model 4 OriginalSin 0.012*** 0.055*** 0.006*** 0.023*** [2.920] [3.384] [3.538] [3.018] OriginalSin × MKT −0.006*** −0.002** [−2.694] [−2.146] MKT 0.002 0.006*** 0.002*** 0.003*** [1.172] [2.730] [2.765] [3.458] PC 0.018*** 0.017*** 0.009*** 0.009*** [4.252] [4.081] [4.887] [4.736] SIZE 0.012*** 0.011*** 0.005*** 0.004*** [4.003] [3.971] [3.345] [3.320] LEV −0.038** −0.035** −0.000 0.001 [−2.511] [−2.315] [−0.024] [0.137] ROA 0.112** 0.113** 0.153*** 0.153*** [2.140] [2.162] [6.117] [6.142] GROWTH −0.003 −0.004 −0.002 −0.002 [−0.486] [−0.587] [−0.607] [−0.692] CFO 0.016 0.014 −0.003 −0.003 [0.930] [0.847] [−0.408] [−0.499] AGE 0.006 0.007 0.005** 0.005** [1.116] [1.321] [2.107] [2.274] SHARE 0.016 0.017 0.010 0.010 [1.053] [1.087] [1.427] [1.452] WEDGE −0.011*** −0.011*** −0.004** −0.004** [−2.684] [−2.714] [−2.033] [−2.062] Constant −0.229*** −0.257*** −0.122*** −0.132*** [−3.502] [−3.908] [−3.893] [−4.207] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 2,918 2,918 2,918 2,918 F value 5.972 5.846 7.875 7.786 Log likelihood 1479.63 1483.721 3170.134 3172.768 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. negative at least at 5% signiﬁcance level (Model 2: β = −0.006, T = −2.694; Model 4: β = −0.002, T = −2.146). These regression results are highly consistent with the results in Table 5, our results are robust. 4.5.2.2. Heckman two-stage model. The Heckman two-stage selection model is also used to further control the possible eﬀect of sample selection bias. In the ﬁrst stage, the paper constructs a Probit model to shape the suspicion of ‘original sin’ of private enterprises and calculate the inverse Mills ratio (Invmills). Then, the Invmills is added to the regression model to correct the problem of omitted variables. Speciﬁcally, in the ﬁrst stage of Heckman model, we choose OriginalSin_ind, the annual average of the industry of enterprises with suspicion of ‘original sin’, natural logarithm of 1997 GDP of the place where the company was registered (GDP1997) as our instrumental variables, and all control variables in Model 1 to calculate the inverse Mills ratio. The reason why we chose these two instrumental variables is as followings. First, in 1997, the report of 15th National congress of the CPC claimed the policies of ‘grasp the big and let go the small’, ‘advance and retreat’, ‘do something and do nothing’ for state-owned enterprises. Thus, the wave of privatization for small-sized and medium-sized state-owned enter- prises was launched all over the country, especially for those provinces with ﬁnancial 134 X. LI ET AL. diﬃculties. Besides, 1997 was 7 years earlier than the starting year of our sample. Thus, the variable GDP1997 is exogenous well. In addition, due to the existence of the peer eﬀect in the industry, enterprises in the same industry will learn from each other. Therefore, the instrumental variable OriginalSin_ind can alleviate the problem of insuﬃ- cient identiﬁcation of the instrumental variables. Table 9 reports the results. Model 1 shows the results of Heckman’s approach analysis in the ﬁrst stage, and Model 2–5 shows the results of regression analysis in the second stage. In Model 1, both instrumental variables get signiﬁcant coeﬃcients. It means that these two instrumental variables are signiﬁcantly correlated with the variable of suspi- cion of ‘original sin’, indicating these two variables satisfy the requirement of being instrumental variables. The coeﬃcients on the variable OriginalSin in Modle 2 and Modle 4 are both statistically signiﬁcant and positive (Model 2: β = 0.007, t = 2.057; Model 4: β = 0.004, t = 2.605). In Model 3 and Model 5, the interaction term OriginalSin*MKT get signiﬁcant and negative coeﬃcients at the 10% level (Model 3: β = −0.004, t = −2.167; Table 9. Regression results of Heckman two-stage model. Variables OriginalSin Donation1 Donaiton2 Model 1 Model 2 Model 3 Model 4 Model 5 OriginalSin_ind 0.706*** [5.622] GDP1997 −0.334*** [−6.852] OriginalSin 0.007** 0.033*** 0.004*** 0.014** [2.057] [2.625] [2.605] [2.395] OriginalSin × MKT −0.004** −0.001* [−2.167] [−1.716] MKT 0.138*** 0.001 0.002* 0.001** 0.002*** [7.182] [0.929] [1.688] [2.554] [2.985] PC 0.156*** 0.007** 0.007* 0.004** 0.003** [4.165] [1.984] [1.840] [2.249] [2.132] SIZE 0.076*** 0.010*** 0.010*** 0.003*** 0.003*** [2.811] [4.551] [4.557] [3.444] [3.450] LEV −0.102 −0.060*** −0.058*** −0.007 −0.006 [−0.783] [−5.460] [−5.328] [−1.453] [−1.346] ROA −0.423 0.068* 0.067* 0.143*** 0.142*** [−1.004] [1.848] [1.836] [8.441] [8.436] GROWTH −0.121** 0.003 0.003 0.001 0.001 [−2.066] [0.646] [0.646] [0.695] [0.695] CFO 0.483*** −0.005 −0.006 −0.012** −0.013** [3.655] [−0.403] [−0.455] [−2.491] [−2.545] AGE 0.219*** 0.000 0.001 0.001 0.001 [4.292] [0.062] [0.201] [0.298] [0.411] SHARE −1.296*** 0.052*** 0.053*** 0.032*** 0.032*** [−9.064] [3.240] [3.311] [4.448] [4.504] WEDGE 0.323*** −0.026*** −0.026*** −0.012*** −0.012*** [7.605] [−5.444] [−5.426] [−5.335] [−5.324] Invmills −0.048*** −0.049*** −0.027*** −0.027*** [−3.650] [−3.689] [−4.545] [−4.577] Constant −0.596 −0.144** −0.154*** −0.062** −0.066** [−0.871] [−2.430] [−2.580] [−2.329] [−2.453] Year ﬁxed eﬀect Yes Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes Yes No. of observations 6,462 6,462 6,462 6,462 6,462 F value 835.129 10.509 10.275 13.896 13.607 Log likelihood −3314.500 2576.916 2579.401 6357.601 6359.225 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. CHINA JOURNAL OF ACCOUNTING STUDIES 135 Model 4: β = −0.001, t = −1.716). These results also support the hypothesis of this paper, indicating that our ﬁndings are still robust after adopting the Heckman two-stage approach to address the sample selection bias. 4.5.2.3. Placebo test. We run the following placebo test to ensure our results are not plagued by mechanical correlation and unobserved variables. Speciﬁcally, we randomly assign a pseudo company for each private enterprise with ‘original sin’ suspicion, and then rerun our regression models. Figure 2 shows the distribution of t-values of the coeﬃcient of the indicator OriginalSin_Random after 1,000 placebo tests. In Graph (1), the dependent variable is Donation1.We ﬁnd that the distribution of t-values of the coeﬃcient estimates of the variable OriginalSin_random lies around zero and the t-value of the true eﬀect based on Model 1 of Table 5 (t = 2.436) lies outside the 95% range. In Graph (2), we also ﬁnd similar results for dependent variable Donation2. Those results suggest that random assignment of enterprises with ‘original sin’ suspicion does not have a signiﬁcant impact on corporate philanthropy. Therefore, our results are not aﬀected by unobserved variables and mechan- ical correlation and the conclusions are robust. 4.5.2.4. Robustness test of estimation method. Our conclusions are based on analysis of unbalanced panel data. Since there is the possibility for panel data to suﬀer from time-serial autocorrelation, we conducted the following clustering regression by year to alleviate this problem. The results are presented in Table 10. As shown in Table 10, the coeﬃcients of OriginalSin are positive and signiﬁcant in Model 1 and Model 3 (Model 1: β = 0.008, t = 3.583; Model 3: β = 0.005, t = 4.029), while in Model 2 and Model 4, the coeﬃcients of OriginalSin are both negative and signiﬁcant at the 10% level (Model 2: β = −0.004, t = −2.130; Model 4: β = −0.001, t = −1.826). Those results are consistent with results shown in Table 5 and our conclusions are robust to alternative estimation method. (1) Donation1 as dependent variable (2) Donation2 as dependent variable Figure 2. Distribution of t-values of the coeﬃcient estimates of the independent variable originalsin_random. 136 X. LI ET AL. Table 10. Regression results by clustering at year level. Variables Donation1 Donation2 Model 1 Model 2 Model 3 Model 4 OriginalSin 0.008*** 0.035*** 0.005*** 0.014*** [3.585] [2.838] [4.029] [2.783] OriginalSin × MKT −0.004** −0.001* [−2.130] [−1.826] MKT 0.003*** 0.004*** 0.002*** 0.003*** [2.709] [2.893] [4.923] [4.558] PC 0.014*** 0.014*** 0.007*** 0.007*** [6.424] [6.177] [6.537] [6.304] SIZE 0.014*** 0.014*** 0.005*** 0.005*** [7.959] [8.034] [7.079] [7.120] LEV −0.065*** −0.064*** −0.010** −0.009* [−5.374] [−5.407] [−2.024] [−1.958] ROA 0.049* 0.048* 0.133*** 0.132*** [1.670] [1.672] [7.296] [7.368] GROWTH −0.002 −0.002 −0.001 −0.001 [−0.627] [−0.647] [−0.880] [−0.906] CFO 0.011 0.011 −0.003 −0.003 [1.018] [0.983] [−0.702] [−0.751] AGE 0.008*** 0.009*** 0.005*** 0.005*** [3.442] [3.496] [3.186] [3.199] SHARE 0.003 0.003 0.005 0.005 [0.238] [0.292] [0.833] [0.874] WEDGE −0.016*** −0.016*** −0.007*** −0.007*** [−6.528] [−6.499] [−5.451] [−5.443] Constant −0.302*** −0.314*** −0.148*** −0.152*** [−9.287] [−9.155] [−9.017] [−8.848] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 6,515 6,515 6,515 6,515 Log likelihood 2607.669 2610.17 6418.163 6419.779 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. 4.6. Further analysis 4.6.1. Results of the eﬀects of political connections Due to the fact such as history, economy and institution, the private property rights of private enterprises may not be well protected by formal institution. As a result, there are more uncertainties and risks faced with private enterprises in China. As to private entrepreneurs, they may adopt various ways to alleviate those risks. Political connec- tions, as an alternative mechanism of legal protection, can protect private enterprises in a less developed formal institutional environment with insuﬃcient protection of prop- erty rights (Chen & Wang, 2013; Park & Luo, 2001; Wu, Wu, & Liu, 2008; Yang, Feng, & Song, 2014). Therefore, private enterprises generally tend to actively establish and maintain political ties with the government and the oﬃcials. Since political connections contribute to meet the needs of private entrepreneurs to cope with various uncertain- ties, will the inﬂuence of ‘original sin’ suspicion on corporate philanthropy of private enterprises change with the diﬀerence of political connections? To answer this question, we further analyze the potential eﬀect of political connections on the relationship between the suspicion of ‘original sin’ and corpo- rate philanthropy. Table 11 reports the results. As seen in Table 11, in the non- political connections subsample of Model 1 and Model 3, the regression coeﬃcients CHINA JOURNAL OF ACCOUNTING STUDIES 137 Table 11. Results of cross-section analysis according to political connections. Variable Donation1 Donation2 PC =0 PC =1 PC =0 PC =1 Model 1 Model 2 Model 3 Model 4 OriginalSin 0.012** 0.001 0.006*** 0.003 [2.560] [0.315] [2.760] [1.293] SIZE 0.018*** 0.006** 0.006*** 0.003* [6.124] [2.044] [4.964] [1.801] LEV −0.067*** −0.062*** −0.009 −0.009 [−4.931] [−3.277] [−1.575] [−1.135] ROA 0.026 0.083 0.109*** 0.169*** [0.559] [1.440] [5.120] [6.201] GROWTH −0.001 −0.003 −0.001 −0.002 [−0.183] [−0.433] [−0.318] [−0.523] CFO 0.001 0.019 −0.005 −0.003 [0.075] [0.962] [−0.915] [−0.425] AGE 0.008 0.016*** 0.002 0.011*** [1.503] [2.718] [0.888] [4.225] SHARE 0.004 0.001 0.003 0.008 [0.292] [0.047] [0.417] [1.175] WEDGE −0.016*** −0.014** −0.005** −0.008** [−3.548] [−2.026] [−2.502] [−2.294] MKT 0.005*** 0.000 0.003*** 0.002*** [3.558] [0.170] [4.305] [2.680] Intercept −0.449*** −0.049 −0.182*** −0.059* [−6.691] [−0.693] [−6.216] [−1.792] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 3,836 2,679 3,836 2,679 F value 5.750 6.009 7.567 7.639 Log likelihood 1174.910 1505.632 3321.334 3161.327 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. on OriginalSin are positive at least at 5% level (Model 1: β = 0.012, t = 2.560, Model 3: β = 0.006, t = 2.760). However, in the political connections subsample of Model 2 and Model 4, the regression coeﬃcients on OriginalSin are positive but not sig- niﬁcant (Model 2: β = 0.001, t = 0.315; Model 2: β = 0.003, t = 1.293). Further, we use Fisher’s Permutation test method to test the diﬀerence of the coeﬃcients between two groups. The coeﬃcient on OriginalSin in the non-political connections subsample (PC = 0)issigniﬁcantly larger than that in the political connections subsample (PC = 1) at 10% signiﬁcance level (p = 0.066) if Donation1 wasasthe dependent variable. The coeﬃcient on OriginalSin in the non-political connections subsample (PC = 0) is larger than that in the political connections subsample (PC = 1) at 17% signiﬁcance level (p = 0.168) at the marginal signiﬁcance level if Donation2 was as the dependent variable. The above results show that the positive eﬀect on corporate philanthropy of private enterprises mainly occurs in ﬁrms without political connections. Compared with private entrepreneurs with suspicion of ‘original sin’ that have political connections, those ﬁrms which do not have political connections face worse property rights protection and greater uncertainties. Therefore, entrepreneurs of these enterprises have a stronger incentive to establish and improve the legitimacy of the enterprises through strategic charitable donations. 138 X. LI ET AL. 4.6.2. Analysis of the impact of entrepreneurs showing up on the hurun rich list Enterprises with suspicion of ‘original sin’ make more charitable donations to reduce or eliminate the risks and regulatory and social sanctions brought by the public and government departments’ doubts on its legitimacy. The enterprises’ degree of the legitimacy which the public and the government question are positively correlated with the degree of attention received by the enterprises. Thus, we expect that the enterprises with suspicion of ‘original sin’ would make more charitable donations if they receive more attention from the public. To examine the potential eﬀect of public attention, we focus on whether entrepreneurs show up on the Hurun Rich List. Through analyzing the changes of charitable donations between enterprises with suspicion of ‘original sin’ and within suspicion of ‘original sin’ after they ﬁrst show up on the Hurun Rich List, we could verify our assumption of this paper. The reason why we chose showing up on Hurun Rich List as the proxy variable of public attention is that the Hurun Rich List can attract the numerous media reports. In addition, billionaires and their controlling enterprises can receive extensive attention from the public, which further triggers social public discussion about their wealth, leading to a sharp rise in the risks and costs of the enterprises faced (Ye et al., 2012). Comparing diﬀerences between the charitable donation behaviours of the sample companies before (LIST =0) and after (LIST = 1) getting on the list, we analyze the changes of the charitable donation behaviours of enterprises with suspicion of ‘original sin’ when there is more public attention. Table 12 shows regression results. As is shown in Table 12, the coeﬃcient of OriginalSin in Model 2 is signiﬁcantly positive in Model 2 after showing up on the rich list (LIST = 1), while the coeﬃcient of OriginalSin is not signiﬁcant in Model 1 before showing up on the rich list (LIST =0) (Model 1: β = −0.006, t = −0.725; Model 2: β = 0.008, t = 1.774). Similarly, the coeﬃcient of OriginalSin in Model 4 is marginally signiﬁcantly positive, while the coeﬃcient of OriginalSin is not signiﬁcant in Model 3 (Model 3: β = 0.001, t = 0.151; Model 4: β = 0.004, t = 1.561). These results show that the enterprises with suspicion of ‘original sin’ are faced with more questions from public, which triggers more risks and costs of political, regulatory and social sanctions when receiving more public attention. Thus, entrepre- neurs tend to donate more after showing up on the rich list to reduce the negative impact on the suspicion of ‘original sin’. 5. Conclusions and implications Due to institutional defections, Chinese private enterprises have always been faced with the problem of ‘original sin’ suspicion, which poses a threat on the legitimacy of these private enterprises. Being faced with this problem, what method will private entrepreneurs take to mitigate the potential negative eﬀect arising from ‘original sin’ suspicion? We argued that corporate philanthropy is likely to be an eﬀective strategic tool for private enterprises with ‘original sin’ suspicion to obtain and improve their legitimacy from the public and the government. In other words, ‘original sin’ suspi- cion will aﬀect the philanthropic behaviour of private enterprises. We used hand- collected data to test whether ‘original sin’ suspicion aﬀects the corporate philan- thropy of Chinese listed private enterprises and investigated the moderating eﬀect of formal institutions. After various robustness checks, we ﬁnd that there is a positive CHINA JOURNAL OF ACCOUNTING STUDIES 139 Table 12. Results of cross-section analysis according to entrepreneurs showing up on the Hurun rich list. Variables Donation1 Donation2 Model 1 Model 2 Model 3 Model 4 LIST =0 LIST =1 LIST =0 LIST =1 OriginalSin −0.006 0.008* 0.001 0.004 [−0.725] [1.774] [0.151] [1.561] MKT 0.005* 0.002 0.003*** 0.003*** [1.951] [1.361] [2.788] [3.845] PC 0.027*** 0.010*** 0.010*** 0.007*** [3.551] [2.590] [3.176] [3.686] SIZE 0.003 0.017*** −0.002 0.007*** [0.395] [5.388] [−0.750] [4.576] LEV −0.084*** −0.075*** −0.007 −0.017** [−3.058] [−5.154] [−0.586] [−2.400] ROA −0.038 0.087** 0.108*** 0.155*** [−0.419] [2.033] [2.808] [7.515] GROWTH −0.000 −0.008 −0.001 −0.004 [−0.028] [−1.526] [−0.205] [−1.404] CFO 0.017 −0.003 0.001 −0.008 [0.636] [−0.189] [0.046] [−1.206] AGE −0.002 0.014** 0.002 0.006** [−0.234] [2.577] [0.595] [2.222] SHARE −0.023 −0.013 −0.01 −0.003 [−0.805] [−0.867] [−0.838] [−0.441] WEDGE −0.037*** −0.012** −0.020*** −0.006** [−3.289] [−2.349] [−4.061] [−2.294] Constant 0.059 −0.396*** 0.076 −0.195*** [0.375] [−5.615] [1.142] [−5.763] Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 1,296 2,943 1,296 2,943 Chi2 142.999 329.311 186.517 472.701 Log likelihood 361.820 1597.495 1139.415 3291.784 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. relationship between ‘original sin’ suspicion of private enterprises and corporate philanthropy. Formal institutions could attenuate the positive eﬀect of ‘original sin’ suspicion on corporate philanthropy. In addition, we also ﬁnd that the positive eﬀect of ‘original sin’ suspicion on corporate philanthropy is more pronounced for private enterprises without political connections and after the entrepreneurs showing up on the Hurun Rich List. Our study has several implications for practices and policy-makers. First, under the background that the ‘original sin’ problem is a very important and realistic problem faced by private enterprises, the government and the public should face up to the ‘original sin’ problem of private enterprises, and thus reduce the instrumental chari- table donations made by private enterprises and make corporate philanthropy play a role in improving social welfare. Second, on 27 November 2016, the central committee of the communist party of China issued ‘the state council approved the opinions of perfecting the system of property rights in accordance with the law to protect property’. It stressed ‘to face up to the problems accompanied with the development of the private enterprises and handle them according to the law since the reform and opening up’, which alleviate concerns for vast number of 140 X. LI ET AL. enterprises, especially for those who may be regarded with ‘original sin’. Our study indicates that facing up to and properly handling such problem is of great signiﬁ- cance to the long-term development of private enterprises, which to some extent conﬁrms the important positive value of the promulgation of the opinions. Furthermore, to optimize the environment of enterprise management system and reduce the uncertainty of environment that private enterprises face by constraining government behaviour and reducing administrative intervention form the govern- ment, will improve the stability and sustainability of the operating environment faced by the private enterprises as well as weaken their instrumental motivation of corpo- rate philanthropy. Finally, we have to acknowledge that our study has some limitations that warrant further research. Although corruption and loss of state-owned in the process of state- owned enterprise restructuring and privatization are concrete and practical dimensions to describe suspicion of ‘original sin’, there are still some other problems we need to further recognize. In reality, because of the imperfect system in the early development stage of the economy, there are still many obscure questions about ‘original sin’, such as playing ‘edge ball’ and seek loopholes in the formal institutions. These are not involved in our study and need to further explore. At the same time, the microeconomic consequences of the suspicion of ‘original sin’ need to be paid more attention to in the future. Acknowledgments We greatly thank for valuable comments and suggestions from the Senior Editor Prof. Kangtao Ye and two anonymous reviewers. All remaining errors are our own. Disclosure statement No potential conﬂict of interest was reported by the authors. Funding We acknowledge ﬁnancial support from the National Natural Science Foundation of China [Grant No. 71790602 and 71572160]. References Brammer, S., & Millington, A. (2005). Corporate reputation and philanthropy: An empirical analysis. Journal of Business Ethics, 61(1), 29–44. Brown, W.O., Helland, E., & Smith, J.K. (2006). Corporate philanthropic practices. Journal of Corporate Finance, 12(5), 855–877. Campbell, L., Gulas, C.S., & Gruca, T.S. (1999). Corporate giving behavior and decision-maker social consciousness. Journal of Business Ethics, 19(4), 375–383. Chen, & Wang. (2013). Family involvement, political connection and institutional environment: Evidence from Chinese private enterprises. Management World,(10), 130–141. (in Chinese). Chen, J.C., Patten, D.M., & Roberts, R.W. (2008). Corporate charitable contributions: A corporate social performance or legitimacy strategy? Journal of Business Ethics, 82(1), 131–144. CHINA JOURNAL OF ACCOUNTING STUDIES 141 Dai, Y., Pan, Y., & Feng, S. (2014). Are Chinese enterprises’ charitable donations “political contribu- tions”? Evidence from the replacements of the municipal party secretaries. Economic Research, 49(2), 74–86. (in Chinese). Dai, Y., Peng, Z., & Pan, Y. (2016). Corporate charitable donations: Self-Redemption under the risk of litigation. Journal of Xiamen University (Arts & Social Sciences),(2), 122–131. (in Chinese). Du, X. (2015). Is corporate philanthropy used as environmental misconduct dressing? Evidence from Chinese family-owned ﬁrms. Journal of Business Ethics, 129(2), 341–361. Du, X., Guo, J., & Lei, Y. (2010). Political connections and donation of private listed companies: Measurement methods and empirical evidence. Finance and Trade Research, 21(1), 89–99. (in Chinese). Du,X.,Jian,W.,Du,Y.,Feng,W., & Zeng,Q. (2014). Religion, the nature of ultimate owner, and corporate philanthropic giving: Evidence from China. Journal of Business Ethics, 123(2), 235–256. Fan, G., Wang, X., & Zhu, H. (2011). NERI index of marketization of China’s Provinces: The report on the relative process of marketization of each region in China. Beijing: Economic Science Press. (in Chinese). Feng, J., & Cheng, W. (2010). Economic motivations and determinants of corporate philanthropy: Overview and implications. Economic Perspectives,(9), 138–141. (in Chinese). Fu, C., & Ji, L. (2017). Litigation risk and corporate charitable giving: An explanation from the perspective of reputation insurance. Nankai Business Review, 20(2), 108–121. (in Chinese). Galaskiewicz, J. (1997). An urban grants economy revisited: Corporate charitable contributions in the Twin Cities, 1979–81, 1987–89. Administrative Science Quarterly, 42(3), 445–471. Gao, F., Faﬀ, R., & Navissi, F. (2012). Corporate philanthropy: Insights from the 2008 Wenchuan earthquake in China. Paciﬁc-Basin Finance Journal, 20(3), 363–377. Gao, Y., Chen, Y., & Zhang, Y. (2012). “Red scarf” or “green scarf”: Research on the motivations of the charitable donation from private enterprise. Management World, 38(8), 106–114. (in Chinese). Gardberg, N.A., & Fombrun, C.J. (2006). Corporate citizenship: Creating intangible assets across institutional environments. Academy of Management Review, 31(2), 329–346. Gautier, A., & Pache, A.C. (2015). Research on corporate philanthropy: A review and assessment. Journal of Business Ethics, 126(3), 343–369. Guo, J. (2012). Corporate donation in China: Voluntary or requested? Evidence from listed companies. Journal of Finance and Economics,(8), 50–60. (in Chinese). Haley, U.C. (1991). Corporate contributions as managerial masques: Reframing corporate contribu- tions as strategies to inﬂuence society. Journal of Management Studies, 28(5), 485–510. Hornstein, A.S., & Zhao, M. (2018). Reaching through the fog: Institutional environment and cross- border giving of corporate foundations. Strategic Management Journal, 39(10), 2666–2690. Jensen, M.C., & Meckling, W.H. (1976). Theory of the ﬁrm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. Jia, M., Xiang, Y., & Zhang, Z. (2015). The reconstruction of business and government relationship: The usage of business corruption or philanthropic giving?. Nankai Business Review, 18(5), 4–17. (in Chinese). Jia, M., & Zhang, Z. (2010). Does political connection inﬂuence corporate philanthropy?. Management World,(4), 99–113. (in Chinese). Koehn, D., & Ueng, J. (2010). Is philanthropy being used by corporate wrongdoers to buy good will? Journal of Management & Governance, 14(1), 1–16. Lev, B., Petrovits, C., & Radhakrishnan, S. (2010). Is doing good good for you? How corporate charitable contributions enhance revenue growth. Strategic Management Journal, 31(2), 182–200. Li, S., Chen, X., & Song, X. (2016). The generosity of the poor: A study of strategic motivation of corporate philanthropy. Management World,(5), 116–127. (in Chinese). Li, S., Lu, Q., & Song, X. (2012). Corporate donation among China’s money-losing enterprises. China Industrial Economics,(8), 148–160. (in Chinese). 142 X. LI ET AL. Li, S., Song, X., & Wu, H. (2015). Political connection, ownership structure, and corporate philan- thropy in china: A strategic-political perspective. Journal of Business Ethics, 129(2), 399–411. Li, S., & Xie, X. (2014). Corporate social responsibility, political relationship and debt ﬁnancing of private enterprises: Evidence from Chinese capital market. Nankai Business Review, 17(6), 30–40. Li, W., Wang, P., & Xu, Y. (2015). Philanthropy, political connection and debt ﬁnance: Reciprocal behavior of governments and private enterprises. Nankai Business Review, 18(1), 4–14. (in Chinese). Li, Z., Tang, X., & Lian, Y. (2016). The puzzle of Chinese private enterprises’ corporate social responsibility. Management World,(9), 136–148. (in Chinese). Liu, H., Lin, L., & Huang, S. (2017). Oﬃcial turnover and the rise and fall of local enterprises: An evidence from China’s prefecture-level cities. China Industrial Economics,(1), 62–80. (in Chinese). Luo, D., Liao, J., & Wang, J. (2016). Local oﬃcials, turnover and corporate risk: Evidence from Chinese listed ﬁrms. Economic Research, 51(5), 130–142. (in Chinese). Luo, J., Kaul, A., & Seo, H. (2018). Winning us with triﬂes: Adverse selection in the use of philanthropy as insurance. Strategic Management Journal, 39(10), 2591–2617. Luo, J.H., Xiang, Y., & Zhu, R. (2017). Military top executives and corporate philanthropy: Evidence from China. Asia Paciﬁc Journal of Management, 34(3), 725–755. Masulis, R.W., & Reza, S.W. (2015). Agency problems of corporate philanthropy. Review of Financial Studies, 28(2), 592–636. Pan, Y., Wen, R., & Liu, S. (2017). The selﬁsh goodwill: New evidence from corporate philanthropy in typhoon. China Industrial Economics,(05), 135–153. (in Chinese). Park, S.H., & Luo, Y. (2001). Guanxi and organizational dynamics: Organizational networking in Chinese ﬁrms. Strategic Management Journal, 22(5), 455–477. Peng, F., & Fan, Z. (2016). Tax incentives, donation costs and enterprises’ charitable donations. The Journal of World Economy, 39(7), 144–167. (in Chinese). Petrenko, O.V., Aime, F., Ridge, J., & Hill, A. (2016). Corporate social responsibility or CEO narcis- sism? CSR motivations and organizational performance. Strategic Management Journal, 37(2), 262–279. Porter, M.E., & Kramer, M.R. (2002). The competitive advantage of corporate philanthropy. Harvard Business Review, 80(12), 56–68. Qian, X., & Xu, Y. (2014). Oﬃcial turnover, political identity and the risk taking of private listed ﬁrms. China Economic Quarterly, 13(4), 1437–1460. (in Chinese). Sánchez, C.M. (2000). Motives for corporate philanthropy in El Salvador: Altruism and political legitimacy. Journal of Business Ethics, 27(4), 363–375. Shan, L., Gan, L., & Zheng, T. (2008). Corporate donations and economic incentives: An empirical study based on corporate donations following the 5.12 earthquake in China. Economic Research, 43(11), 51–61. (in Chinese). Shiu, Y.M., & Yang, S.L. (2017). Does engagement in corporate social responsibility provide strategic insurance-like eﬀects? Strategic Management Journal, 38(2), 455–470. Su, J., & He, J. (2010). Does giving lead to getting? Evidence from Chinese private enterprises. Journal of Business Ethics, 93(1), 73–90. Tan, W., & Xu, L. (2015). Agency cost, corporate governance, and corporate donations. Economic Management, 37(9), 51–62. (in Chinese). Tang, S., Wen, D., & Sun, Z. (2017). “original sin” suspicion and the earnings quality of private listed ﬁrms. Management World,(8), 106–122. (in Chinese). Tang, Y., Zuo, J., & Li, H. (2014). The impact of institutional environment transition on corporate philanthropic behavior. Economic Research, 49(2), 61–73. (in Chinese). Wang, H., & Qian, C. (2011). Corporate philanthropy and corporate ﬁnancial performance: The roles of stakeholder response and political access. Academy of Management Journal, 54(6), 1159–1181. Wang, J., & Pan, X. (2018). The inﬂuence of charitable donation on audit opinion: An empirical analysis based on the selﬁsh motives of charitable donations. Auditing Research,(3), 87–94. (in Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 143 Wang, K., Jin, Z., & Jiao, J. (2014). Strategic group membership, corporate charitable contribution and performance: Evidence from Chinese real estate industry. Nankai Business Review, 17(6), 53–62. (in Chinese). Wang, X., Fan, G., & Yu, J. (2017). NERI index of marketization of China’s Provinces. Beijing: Economic Science Press (in Chinese). Wang, X., Li, Y., & Li, F. (2015). Research on corporate social responsibility and the eﬀectiveness of executive compensation incentive: Strategic motivation or shirking excuse. Accounting Research, (10), 51–58. (in Chinese). Wang, Y., Pan, Y., & Huang, L. (2016). Corporate charitable donations: Public behavior or other intentions. Finance and Trade Research, 27(1), 133–141. (in Chinese). Wu, W., Wu, C., & Liu, X. (2008). Political connection and market valuation: Evidence from China individual-controlled listed ﬁrms. Economic Research,(7), 130–141. (in Chinese). Xiu, Z., & Zhou, Z. (2016). Regional happiness, social capital and corporate philanthropic giving. Journal of Management Science, 29(2), 146–160. (in Chinese). Xu, N., & Li, Z. (2016). CEOs’ poverty experience and corporate philanthropy. Economic Research, 51 (12), 133–146. (in Chinese). Yang, Z., Feng, L., & Song, Z. (2014). The political connection eﬀect between the institutional environment and state-ownership. Review of Investment Studies, 33(7), 144–157. (in Chinese). Ye, Q., Li, Z., & Li, G. (2012). Does the rich list aﬀect the quality of accounting information?: Perspective of political cost. Management World,(1), 104–120. (in Chinese). Zhang, M., Ma, L., & Zhang, W. (2013). The eﬀect of ﬁrm-government bonding on corporate charitable donations: Based on empirical evidence of listed companies in China. Management World,(7), 163–171. (in Chinese). Zhang, R., Rezaee, Z., & Zhu, J. (2010). Corporate philanthropic disaster response and ownership type: Evidence from Chinese ﬁrms’ response to the Sichuan earthquake. Journal of Business Ethics, 91(1), 51–63. Zhang, R., Zhu, J., Yue, H., & Zhu, C. (2010). Corporate philanthropic giving, advertising intensity, and industry competition level. Journal of Business Ethics, 94(1), 39–52. Zheng, G., & Xu, Y. (2011). Corporate philanthropy, corporate governance and shareholder wealth. Nankai Business Review, 14(2), 92–101. (in Chinese). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png China Journal of Accounting Studies Taylor & Francis http://www.deepdyve.com/lp/taylor-francis/original-sin-suspicion-institutional-environment-and-corporate-8ZpQj4Edvr

Loading next page...

References (65)

Heli Wang, C. Qian (2011)
Corporate Philanthropy and Corporate Financial Performance: The Roles of Stakeholder Response and Political Access
Academy of Management Journal, 54
Usha Haley (1991)
Corporate Contributions As Managerial Masques: Reframing Corporate Contributions As Strategies to Influence Society*
Journal of Management Studies, 28
(2016)
Corporate charitable donations: Public behavior or other intentions
M. Porter, M. Kramer (2002)
The competitive advantage of corporate philanthropy.
Harvard business review, 80 12
Xing-qiang Du (2014)
Is Corporate Philanthropy Used as Environmental Misconduct Dressing? Evidence from Chinese Family-Owned Firms
Journal of Business Ethics, 129
(2014)
The impact of institutional environment transition on corporate philanthropic behavior
(2017)
Litigation risk and corporate charitable giving: An explanation from the perspective of reputation insurance
O. Petrenko, Federico Aime, Jason Ridge, A. Hill (2016)
Corporate social responsibility or CEO narcissism? CSR motivations and organizational performance
Strategic Management Journal, 37
(2014)
Strategic group membership, corporate charitable contribution and performance: Evidence from Chinese real estate industry
S. Park, Yadong Luo (2001)
Guanxi and organizational dynamics: organizational networking in Chinese firms
Southern Medical Journal, 22
Ronald Masulis, S. Reza (2014)
Agency Problems of Corporate Philanthropy
Corporate Governance: Social Responsibility & Social Impact eJournal
(2014)
Corporate social responsibility, political relationship and debt financing of private enterprises: Evidence from Chinese capital market
Xing-qiang Du, Wei Jian, Yingjie Du, Wentao Feng, Quan Zeng (2014)
Religion, the Nature of Ultimate Owner, and Corporate Philanthropic Giving: Evidence from China
Journal of Business Ethics, 123
(2014)
Official turnover, political identity and the risk taking of private listed firms
W. Brown, Eric Helland, Janet Smith (2006)
Corporate philanthropic practices
Journal of Corporate Finance, 12
Jun Su, Jia He (2009)
Does Giving Lead to Getting? Evidence from Chinese Private Enterprises
Journal of Business Ethics, 93
Jin‐hui Luo, Yuangao Xiang, Ruichao Zhu (2017)
Military top executives and corporate philanthropy: Evidence from China
Asia Pacific Journal of Management, 34
(2016)
Tax incentives, donation costs and enterprises’ charitable donations
(2008)
Political connection and market valuation: Evidence from China individual-controlled listed firms
J. Galaskiewicz (1997)
An urban grants economy revisited: Corporate charitable contributions in the twin cities, 1979-81, 1987-89
Administrative Science Quarterly, 42
(2014)
Are Chinese enterprises’ charitable donations “political contributions”? Evidence from the replacements of the municipal party secretaries
(2016)
Corporate charitable donations: Self-Redemption under the risk of litigation
F. Gao, R. Faff, F. Navissi (2012)
Corporate philanthropy: Insights from the 2008 Wenchuan Earthquake in China
Pacific-basin Finance Journal, 20
(2016)
Regional happiness, social capital and corporate philanthropic giving
(2008)
Corporate philanthropy: Insights from the
Naomi Gardberg, C. Fombrun (2006)
Corporate Citizenship: Creating Intangible Assets Across Institutional Environments
Academy of Management Review, 31
(2017)
original sin” suspicion and the earnings quality of private listed firms
Jennifer Chen, Dennis Patten, R. Roberts (2008)
Corporate Charitable Contributions: A Corporate Social Performance or Legitimacy Strategy?
Journal of Business Ethics, 82
(2012)
Does the rich list affect the quality of accounting information?: Perspective of political cost
(2016)
Local officials, turnover and corporate risk: Evidence from Chinese listed firms
(2016)
The puzzle of Chinese private enterprises’ corporate social responsibility
(2014)
The political connection effect between the institutional environment and state-ownership
(2017)
Official turnover and the rise and fall of local enterprises: An evidence from China’s prefecture-level cities
(2011)
Corporate philanthropy, corporate governance and shareholder wealth
(2013)
Family involvement, political connection and institutional environment: Evidence from Chinese private enterprises
Leland Campbell, Charles Gulas, T. Gruca (1999)
Corporate Giving Behavior and Decision-Maker Social Consciousness
Journal of Business Ethics, 19
Sihai Li, Xianzhong Song, Huiying Wu (2015)
Political Connection, Ownership Structure, and Corporate Philanthropy in China: A Strategic-Political Perspective
Journal of Business Ethics, 129
Song Xian-zhong (2012)
Corporate Donation among China's Money-losing Enterprises
China Industrial Economy
Abigail Hornstein, Minyuan Zhao (2018)
Reaching through the fog: Institutional environment and cross‐border giving of corporate foundations
Strategic Management Journal
Guo Jian-hua (2012)
Corporate Donation in China:Voluntary or Requested? Evidence from Listed Companies
Journal of finance and economics
Yung-Ming Shiu, Shou-Lin Yang (2017)
Does engagement in corporate social responsibility provide strategic insurance-like effects?
Southern Medical Journal, 38
B. Lev, Christine Petrovits, S. Radhakrishnan (2008)
Is Doing Good Good for You? How Corporate Charitable Contributions Enhance Revenue Growth
Public Economics: Taxation
(2015)
Agency cost, corporate governance, and corporate donations
(2016)
The generosity of the poor: A study of strategic motivation of corporate philanthropy
Ran Zhang, Z. Rezaee, Jigao Zhu (2009)
Corporate Philanthropic Disaster Response and Ownership Type: Evidence from Chinese Firms’ Response to the Sichuan Earthquake
Journal of Business Ethics, 91
王小鲁, 朱恒鹏 (2001)
中国市场化指数 : 各地区市场化相对进程报告 = NERI Index of Marketization of China's Provinces
Ran Zhang, Jigao Zhu, Heng Yue, Chunyan Zhu (2010)
Corporate Philanthropic Giving, Advertising Intensity, and Industry Competition Level
Journal of Business Ethics, 94
Michael Jensen, W. Meckling (1976)
Theory of the Firm
(2008)
Corporate donations and economic incentives: An empirical study based on corporate donations following the 5.12 earthquake in China
Arthur Gautier, Anne-Claire Pache (2015)
Research on Corporate Philanthropy: A Review and Assessment
Journal of Business Ethics, 126
(2010)
Economic motivations and determinants of corporate philanthropy: Overview and implications
(2015)
The reconstruction of business and government relationship: The usage of business corruption or philanthropic giving
(2015)
Research on corporate social responsibility and the effectiveness of executive compensation incentive: Strategic motivation or shirking excuse
Jiao Luo, A. Kaul, Haram Seo (2018)
Winning us with trifles: Adverse selection in the use of philanthropy as insurance
Strategic Management Journal
Wei’an Li, Pengcheng Wang (2016)
Philanthropy, political connection and debt finance: Reciprocal behavior of governments and private enterprises
Nankai Business Review International, 7
Lei Yu (2010)
Political Connections and Donation of Private Listed Companies:Measurement Methods and Empirical Evidence
Finance and Trade Research
(2016)
CEOs’ poverty experience and corporate philanthropy
(2018)
The influence of charitable donation on audit opinion: An empirical analysis based on the selfish motives of charitable donations
(2012)
Red scarf” or “green scarf”: Research on the motivations of the charitable donation from private enterprise
Stephen Brammer, A. Millington (2005)
Corporate Reputation and Philanthropy: An Empirical Analysis
Journal of Business Ethics, 61
(2010)
Does political connection influence corporate philanthropy
D. Koehn, Joe Ueng (2010)
Is philanthropy being used by corporate wrongdoers to buy good will?
Journal of Management & Governance, 14
(2017)
The selfish goodwill: New evidence from corporate philanthropy in typhoon
Carol Sánchez (2000)
Motives for Corporate Philanthropy in El Salvador: Altruism and Political Legitimacy
Journal of Business Ethics, 27
(2013)
The effect of firm-government bonding on corporate charitable donations: Based on empirical evidence of listed companies in China

Publisher: Taylor & Francis
Copyright: © 2019 Accounting Society of China
ISSN: 2169-7221
eISSN: 2169-7213
DOI: 10.1080/21697213.2019.1643068
Publisher site: See Article on Publisher Site

Abstract

CHINA JOURNAL OF ACCOUNTING STUDIES 2019, VOL. 7, NO. 1, 119–143 https://doi.org/10.1080/21697213.2019.1643068 ARTICLE ‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises Xue Li, Jin-hui Luo and Zeyue Huang School of Management, Xiamen University, Xiamen, China ABSTRACT KEYWORDS ‘Original sin’ suspicion; The suspicion of ‘original sin’ is an unavoidable problem in the devel- private enterprises; opment of China’s private enterprises. Using a sample of A-share corporate philanthropy; private listed ﬁrms from 2004–2016, we investigate whether and institutional environment how the suspicion of ‘original sin’ of private enterprises aﬀects their corporate philanthropy. We ﬁnd that private enterprises with the suspicion of ‘original sin’ have stronger incentives to take corporate philanthropy as a strategic tool to reduce the threat arising from the ‘original sin’. High development levels of institutional environment can attenuate the positive eﬀect of ‘original sin’ suspicion on private enterprises’ corporate philanthropy. Furthermore, we ﬁnd that the positive association between the ‘original sin’ suspicion of private enterprises and their corporate philanthropy is more pronounced when they do not have political connections and/or after their ulti- mate controllers are showed up on the Hurun Rich List. 1. Introduction Since the Reform and Opening policy, the private sector has developed vigorously from scratch and contributes greatly to China’s economic growth. But until nowadays, private enterprises are still faced with the suspicion of ‘original sin’. ‘Origin sin’ means that some private enterprises are suspected of engaging in some ‘unruly’ behaviours when they ﬁrst obtained their initial capital. Those private enterprises may not be formally recog- nized by the public (Tang, Wen, & Sun, 2017; Ye, Li, & Li, 2012). In some extreme cases, ‘original sin’ suspicion even leads to the private entrepreneurs being jailed (for example, the founder of WuMart, Zhang Wenzhong and the eﬀective controller of Greencool, Gu Chujun). Faced with the threat of not being recognized and protected formally in legitimacy, how to improve their legitimacy has become a strategic problem for these private enterprises for a long time. In other words, it has become a very important question that how those private entrepreneurs can deal with the lack of legitimacy due to ‘original sin’ suspicion. Existing studies show that the suspicion of ‘original sin’ of private enterprises will bring higher political costs to enterprises. In order to decrease those political costs, these enterprises tend to lower their transparency (Tang et al., 2017; Ye et al., 2012). CONTACT Jin-hui Luo jinhuiluo@xmu.edu.cn School of Management, Xiamen University, No. 422, Siming South Road, Xiamen, Fujian 361005, P. R. China Paper accepted by Kangtao Ye. © 2019 Accounting Society of China 120 X. LI ET AL. Speciﬁcally, the ‘original sin’ suspicion leads to that the legitimacy of the private companies cannot be recognized by the government and the public. A large number of studies in recent years have shown that, as one of demonstrations of social respon- sibility, corporate philanthropy is often strategically used by companies to achieve their political goals, such as to promote the legitimacy and gain goodwill from the public (Dai, Pan, & Feng, 2014; Du, 2015; Hornstein & Zhao, 2018; Li, Chen, & Song, 2016; Li, Wang, & Xu, 2015; Sánchez, 2000; Wang & Qian, 2011). Therefore, this paper is intended to investigate that whether corporate philanthropy is employed by private enterprises with ‘original sin’ suspicion as a strategic tool to alleviate the impact of ‘original sin’ and gain legitimacy. The following research will be conducted from two aspects: (1) how will the suspicion of ‘original sin’ aﬀect the philanthropic behaviour of private enter- prises? (2) can formal institutional environment have an impact on the relationship between ‘original sin’ suspicion and corporate philanthropy? Using a sample of A-share private listed ﬁrms between 2004 and 2016, we measure private enterprises’‘original sin’ according to whether their original property rights are transferred from state ownership to private ownership, and examine the relationship between the ‘original sin’ suspicion of private enterprises and their corporate philan- thropy. We ﬁnd that: (1) the suspicion of ‘original sin’ of ﬁrms has a signiﬁcantly positive eﬀect on corporate philanthropy. Due to public and government’s doubts about the legitimacy of enterprises and even the threat of their private property rights being takeover caused by the suspicion of ‘original sin’, private entrepreneurs with ‘original sin’ suspicion have stronger motivation to reduce or eliminate the adverse eﬀect through corporate philanthropy; (2) the positive eﬀect of ‘original sin’ suspicion on charitable donations of private enterprises is found to be weaker for ﬁrms located in regions with more developed institutions. In poorer institutional environment, private enterprises with suspicion of ‘original sin’ may face more serious threats of being liquidated and the lack of protection of their legitimacy, which may further strengthen their motivation to improve corporate legitimacy through corporate philanthropy. Our results are robust after adopting alternative variable measurements and tacking the potential endogeneity concerns. Furthermore, we also ﬁnd that the positive eﬀect of ‘original sin’ suspicion of private enterprises on corporate is more pronounced in those ﬁrms without political connections and when the entrepreneur was listed on the Hurun rich list. Therefore, our ﬁndings indicate that private enterprises with suspicion of ‘original sin’ have strong motivation to enhance the legitimacy of enterprises and protect their property rights through corporate philanthropy. Our study contributes to the literature in several ways. First, this paper contributes to the literature about ‘original sin’ suspicion of private enterprises in developing econo- mies like China. The suspicion of ‘original sin’ is an unavoidable problem in the devel- opment of private enterprises, and has triggered widespread discussion throughout the country several times. We empirically examine the impact of the suspicion of ‘original sin’ of private enterprises on corporate philanthropy and ﬁnd that the private enterprises with the suspicion of ‘original sin’ tend to strategically donate more to alleviate and oﬀset the costs caused by the lack of legitimacy, which can deepen our understanding of the outcome of suspicion of ‘original sin’ problem faced with private enterprises. Second, this paper also contributes to the literature regarding the determi- nants of corporate philanthropy. To the best of our knowledge, we are among the ﬁrst CHINA JOURNAL OF ACCOUNTING STUDIES 121 to examine the non-altruistic motivations of corporate philanthropic behaviours under the transitional institutional background from the perspective of ‘original sin’ suspicion of private enterprises. Our ﬁnding conﬁrms the strategic motivation of corporate philan- thropy, thus providing incremental empirical evidence for the research about corporate philanthropy. Third, we further analyze the moderating eﬀect of formal institutional development on the relationship between the suspicion of ‘original sin’ of private enterprises and corporate philanthropy, showing that the formal institutions can have a profound eﬀect on behaviours of the enterprises, especially private enterprises. 2. Literature review and hypotheses development 2.1. Literature review 2.1.1. ‘Original sin’ suspicion of private enterprises Under the background of China’s emerging and transitional economy, the transforma- tion of institutions and the establishment of new rules are a normal situation faced by enterprises. Under such background, due to the imperfect institution system and weak implementation, some private enterprises inevitably have some unruly or even illegal operational behaviours in the development of their enterprises, which may lead to ‘original sin’ suspicion from the public and the government. Such suspicion would have a profound impact on enterprise behaviours. ‘Original sin’ suspicion mainly refers to the suspicion from social public to the compliance and legality of private entrepre- neurs’ initial wealth accumulation. Such suspicion could lead to the problem of private enterprise legitimacy, and thus they could not get the acceptance of the public and the government department and legal protection. Therefore, the private entrepreneurs suﬀer the threats of higher liquidation, even the deprivation of property rights and legal sanction (Tang et al., 2017; Ye et al., 2012). In recent years, some scholars have paid attention to the possible economic con- sequences of the suspicion of ‘original sin’ of private enterprises. Using the event of the entrepreneurs of the listed ﬁrms ranking on the Hurun Rich List, Ye et al. (2012) suggested that the suspicion of ‘original sin’ would bring more attention and question from public to the rich, and they may suﬀer higher threat of being liquidated resulting from ‘original sin’ suspicion. It makes the rich face a sudden increase in political costs after being listed. So the billionaires with ‘original sin’ have stronger motivation to reduce the political cost by reducing the quality of accounting information. Similarly, from perspective of political costs, Tang et al. (2017) argued that the private enterprises with ‘original sin’ suspicion have stronger motivation to reduce the political costs they bear by reducing the enterprise information transparency. 2.1.2. Motivations of corporate philanthropy As an important and explicit form of corporate social responsibility, corporate philanthropy has always attracted academic attention. According to FASB, corporate philanthropy is an uncon- ditional, voluntary and nonreciprocal transfers of cash and other assets to other entities such as non-proﬁt organizations and government agencies. This deﬁnition is in line with the altruistic motivation of corporate philanthropy identiﬁed by previous literature (Campbell, Gulas, & 122 X. LI ET AL. Gruca, 1999;Xiu &Zhou, 2016). In addition to altruistic motives, there are also self-interested motivations for corporate philanthropic giving, which can be classiﬁed into three types: (1) economic motivation, (2) managerial self-interest motivation, and (3) political and institutional motivation (Feng & Cheng, 2010;Lietal., 2016;Petrenko, Aime,Ridge, & Hill, 2016;Wang, Li, & Li, 2015; Zhang, Rezaee, & Zhu, 2010; Zhang, Zhu, Yue, & Zhu, 2010). Many studies have supported the economic motivation of corporate philanthropy, which suggests that one purpose of corporate philanthropy is to maximize economic beneﬁts. Philanthropic giving can create an advertising eﬀect that is helpful for the ﬁrm to establish positive image and reputation, leading to better achievement of both economic and social goals (Brammer & Millington, 2005; Lev, Petrovits, & Radhakrishnan, 2010; Pan, Wen, & Liu, 2017; Peng & Fan, 2016; Porter & Kramer, 2002; Shan, Gan, & Zheng, 2008; Wang, Jin, & Jiao, 2014; Zhang et al., 2010; Zheng & Xu, 2011). In addition, corporate philanthropy can be employed as a means to cope with negative events by shifting public attention away from bad news (Shiu & Yang, 2017), window- dressing poor performance from another aspects of corporate social responsibility (Chen, Patten, & Roberts, 2008; Du, 2015; Gao, Chen, & Zhang, 2012; Luo, Kaul, & Seo, 2018), attenuating risk perception of shareholder about performance decline, mitigating negative market reaction for earning restatements (Koehn & Ueng, 2010; Li, Chen, & Song, 2016), reducing litigation risks (Dai, Peng, & Pan, 2016; Fu & Ji, 2017), and even helping the ﬁrm to obtain a more favourable audit opinion (Wang & Pan, 2018). Second, corporate philanthropy can also be driven by managerial self-interest motivation (Haley, 1991). Jensen and Meckling (1976) suggested that corporate philanthropy can provide private beneﬁts to corporate managers, even though those beneﬁts may be at the expense of shareholders’ interest. Those private beneﬁts can be both pecuniary, such as higher compensation, and non-pecuniary, such as social recognition and higher reputa- tional status (Galaskiewicz, 1997; Jia & Zhang, 2010; Petrenko et al., 2016). Opportunistic behaviours of managers can also lead to higher philanthropic contributions (Petrenko et al., 2016;Tan & Xu, 2015). Finally, corporate philanthropy may also be politically driven, especially in developing economies with higher policy uncertainty (Du, Guo, & Lei, 2010; Gautier & Pache, 2015; Li, Lu, &Song, 2012; Tang, Zuo, & Li, 2014; Wang & Qian, 2011). By coordinating with the government’s scheme and sharing the costs, companies can quickly gain goodwill and trust from the government, which may help ﬁrms to gain more political support and legitimacy (Hornstein & Zhao, 2018;Sánchez, 2000; Su & He, 2010; Tang et al., 2014;Wang &Qian, 2011) and ultimately helps the ﬁrm to obtain more economic resources such as subsidies and loan credits (Dai et al., 2014;Gao, Faﬀ,&Navissi, 2012; Jia & Zhang, 2010;Li et al., 2015; Li, Song, & Wu, 2015; Li, Tang, & Lian, 2016; Li, Lu, & Song, 2012; Su & He, 2010; Wang, Pan, & Huang, 2016; Wang & Qian, 2011; Zhang, Ma, & Zhang, 2013). Overall, there is a consensus in the literature that political motivation is very impor- tant for ﬁrms to conduct corporate philanthropy, especially in transitional economies like China, where many private enterprises are facing with ‘original sin’ suspicion. However, there is little research regarding whether and how the suspicion of ‘original sin’ may aﬀect the corporate behaviour of private enterprises. This paper intends to investigate the possible impact of the ‘original sin’ suspicion of private enterprises on corporate philanthropy, thereby contributes to the literature by deepening our understanding of the economic consequences of the ‘original sin’ suspicion. CHINA JOURNAL OF ACCOUNTING STUDIES 123 2.2. Hypotheses development Under the threat that enterprise may not be formally recognized and eﬀectively pro- tected because of the suspicion of ‘original sin’, a rational entrepreneur will certainly take some measures to alleviate this threat and any disadvantages resulting from it. We argue that corporate philanthropy is likely to be a strategic tool for those private enterprises suspected of ‘original sin’ to obtain and enhance their legitimacy. Firstly, enterprises can obtain the recognition and protection from government departments through charitable donations. As a safe way for enterprises to obtain their legitimacy, corporate philanthropy can help enterprises suspected of ‘original sin’ win favor and recognition from the government at a faster speed, and achieve the purpose of obtaining and improving their own legitimacy and avoiding risks by seeking political supports and protection (Dai et al., 2014; Hornstein & Zhao, 2018; Jia, Xiang, & Zhang, 2015; Sánchez, 2000; Su & He, 2010; Tang et al., 2014; Wang & Qian, 2011). Secondly, corporate philanthropy is also conducive to improving the public’s recognition of enterprises. In China, the public has a long-standing prejudice against private enter- prises with suspicion of ‘original sin’, namely the phenomenon of ‘hatred for the rich’. Besides improving social welfare, corporate philanthropy can help to establish a good image of the enterprise in the public, so as to alleviate or even eliminate the prejudice of the public towards the private enterprises (Guo, 2012; Li & Xie, 2014; Wang & Pan, 2018; Wang & Qian, 2011). Therefore, it is helpful for the private enterprises to enhance the public’s recognition of the legitimacy of enterprises. In addition, corporate philanthropy is conducive for enterprises to accumulate moral capital and plays a role of ‘reputation insurance’ (Fu & Ji, 2017; Gao, Chen, & Zhang, 2012; Shiu & Yang, 2017), which con- tributes to reduce the risks and the regulatory and social sanctions caused by the suspicion of ‘original sin’. Base on the theoretical analysis above, we argue that private enterprises with suspi- cion of ‘original sin’ will have stronger motivation to engage in corporate philanthropy so as to alleviate or eliminate the doubts on the legitimacy of enterprises raised by the public and government departments. Thus, we put forward our ﬁrst hypothesis. Hypothesis 1: Ceteris paribus, ‘original sin’ suspicion will increase the level of charitable donations of private enterprises. In fact, the suspicion of ‘original sin’ is mainly due to the imperfect of formal institutions in the process of China’s economic transformation. It means that the weak institutional environment leads to the suspicion of ‘original sin’ and brings about a series of economic consequences. Therefore, we argue that the external institutional environment is an important potential factor regulating the relationship between suspicion of ‘original sin’ of private enterprises and their corporate philan- thropy. In China, although enterprises face the same legal system and political system, there are great diﬀerences in the institutional environment among diﬀerent regions (Fan, Wang, & Zhu, 2011). Generally, a more developed institutional environment means lower government intervention and better legal protection (Li, Tang, & Lian, 2016; Tang et al., 2017). For private enterprises suspected of ‘original sin’ located in 124 X. LI ET AL. regions with more developed institutional environment, the situation that their legiti- macy cannot be formally recognized and their property right protection will be improved to a large extent. At this time, the motivation of enterprises to minimize the adverse eﬀect of ‘original sin’ suspicion through charitable donations will be signiﬁcantly weakened. However, for private enterprises located in regions with less developed institutions, due to weak protection of enterprises by law and more administrative intervention of the government on enterprises (Luo, Xiang, & Zhu, 2017; Tang et al., 2017), the legitimacy and property rights of enterprises are diﬃcult to be oﬃcially recognized and eﬀectively protected by law. Thus, private enterprises with ‘original sin’ suspicion have stronger motivation to engage in corporate philan- thropy to obtain and strengthen their legitimacy status accordingly. Besides, for ﬁrms located in regions with more developed formal institutions, they would obtain the initial property of their enterprises in a market-oriented way, a more transparent and standard way (Tang et al., 2017). It means that these enterprises would gain wider recognition from the public and they would be less likely to be suspected with ‘original sin’. Therefore, the enterprises’ strategic motives for charitable giving would be largely weakened. Thus, we propose our second hypothesis. Hypothesis 2: Institutional environment will weaken the positive eﬀect of ‘original sin’ suspicion on corporate philanthropy of private enterprises. 3. Research design 3.1. Sample and data Our simple includes 1,288 private listed ﬁrms controlled by individual or family from 2004 to 2016. We choose 2004 as the beginning of our simple because that the China Security Regulatory Commission required listed ﬁrms to report details about the ultimate controllers since 2003, and some variables used in the subsequent research of our study (such as ﬁrms’ growth opportunities measured by the growth rate of operating income) need to use the data on one preceding year. After excluding 23 listed ﬁrms whose ultimate controller transferred from the government to family or individual and seven ﬁrms whose ultimate controller are diﬃcult to judge and those are fail to disclose company predecessor, our initial sample comprises 1,258 ﬁrms and 8,173 observations. To alleviate the potential inﬂuence of abnormal samples, we exclude ﬁrms that are specially treated (ST or *ST) (72 observations), that are in ﬁnancial industries (11 observations), that issue debt exceeding asset value (4 observations), that are cross listed (29 observations), that have missing data (1,542 observations). Our ﬁnal sample includes 6,515 ﬁrm-year observations. Table 1 provides the distribution of industry and year of our sample. Our data on ﬁrms’ original property rights were obtained manually from historical evolution of the ﬁrms and the information of ultimate controllers shown in ﬁrms’ prospectus which were disclosed on CNINFO (www.cninfo.com.cn). The ﬁnancial data and corporate governance data are from China Stock Market and Accounting Research (CSMAR) database. The data we use to calculate formal institutional environment was from National Bureau of Statistics (www.stats.gov.cn). CHINA JOURNAL OF ACCOUNTING STUDIES 125 Table 1. Sample distribution by industry and year. Industry Year Total 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 A 0234567 12 14 16 17 17 17 120 B 010011268 88 9 9 53 C0 1112559 18 28 31 32 37 37 207 C1 2 8 8 10 12 13 17 25 29 32 32 36 36 260 C2 012223346 77 8 8 53 C3 021569 13 15 18 19 19 20 20 147 C4 4 8 8 12 17 23 31 70 100 108 111 119 117 728 C5 2369 15 18 22 46 60 67 68 71 72 459 C6 1335 13 16 20 46 58 66 68 73 72 444 C7 5 16 18 25 41 51 84 166 230 264 276 302 301 1,779 C8 1 8 8 12 14 17 27 48 61 65 66 71 69 467 C9 000112356 68 9 8 49 D 000001133 66 6 6 32 E 123567 10 18 25 27 28 29 28 189 F 000001345 56 7 7 38 G 3 5 5 11 22 24 51 88 115 131 139 144 141 879 H 0333369 17 22 24 24 26 25 165 J 122255788 77 7 8 69 K 011199 16 27 33 35 37 42 41 252 L 0002348 14 16 18 19 21 20 125 Total 21 66 72 111 180 221 343 640 845 942 978 1,054 1,042 6,515 Notes: Industry code A-M correspond, respectively: Agriculture, Forestry, Husbandry and Fishery (A); Mining (B); Food and Drink (C0); Textile, Clothing and Fur (C1); Wood and Furniture (C2); Papermaking and Printing (C3); Petroleum, Chemical and Plastic (C4); Electronics (C5); Metal and Nonmetallic Industry (C6); Machinery, Equipment and Instrument (C7); Medicine and Biological products (C8); Other manufacturing Industry (C9); Electricity, Gas and Water (D); Construction Industry (E); Transportation and Warehousing (F); Information and Technology (G); Wholesale and Retail (H); Real Estate (J); Social Services (K); Communication and Culture Industry (L); Comprehensive Industry (M). 3.2. Measurement 3.2.1. Dependent variables Following prior literature (Brown, Helland, & Smith, 2006; Du et al., 2010;Xu& Li, 2016;Zhang et al., 2013), we adopt the two proxies, Donation1 and Donation2, to measure the dependent variable, corporate philanthropy. Donation1 is the amount of donation divided by operating revenue, Donation2 is the amount of donation divided by total asset. Especially, referring to existed literature (Dai et al., 2014; Du et al., 2010;Masulis &Reza, 2015)the missing value of Donation1 and Donation2 is assigned to zero. 3.2.2. Independent variable As to the deﬁnition of ‘original sin’, the existing literature has not reached a consensus (Ye et al., 2012). In this paper, ‘original sin’ mainly refers to that private enterprises may be subjected to the suspicion of illegal and inappropriate deeds during the initial privatization process. In other words, the ‘ﬁrst pot of gold’ that the private enterprises need to take oﬀ may not be lawfully acquired. Therefore, following Tang et al. (2017), we use original evolution of property rights of private enterprises as the proxy of ‘original sin’ suspicion. Speciﬁcally, we construct the variable OriginalSin. According to the information about company history disclosed in the prospectus of listed companies, if a private enterprise was formerly state-owned and later privatized through methods like equity transfer before they were listed, then it is deﬁned as the privatized ﬁrms. Otherwise, if a private company was founded by a natural person at the beginning of its establishment, it is identiﬁed as de novo private ﬁrms. The variable OriginalSin equals to 1 for privatized ﬁrms and 0 for de novo 126 X. LI ET AL. private ﬁrms. Compared with the de novo private ﬁrms, the privatized ﬁrms may suﬀer suspicion of original sin from public due to the opaque transformation process from state- owned property rights to privately owned property rights, resulting severer suspicion from the public and the government concerning that they may gain their property rights through illegal or inappropriate ways that may have led to loss of state assets. 3.2.3. Moderating variables The moderating variable is the development of formal institutions. In the most existing literature, many studies took the NEBI’s marketization index to measure it (Fan et al., 2011). However, the marketization index issued by Fan et al. (2011) only updated to 2009, and the marketization index of regions from 2008–2014 issued by Wang, Fan, and Yu (2017)has alarge diﬀerence with Fan et al. (2011). As a result, these two indexes of diﬀerent sources cannot be combined simply to measure the development of formal institutions. Given that, following previous research (Wang & Qian, 2011), we computed the ratio of a province’sGDP to its government’s budget to measure the development of formal institutions (MKT). In addition, we also took NEBI’s marketization index for the sake of a robustness check, in which the value of NEBI’s marketization indexes between 2010 and 2016 was replaced by the index value of 2009. 3.2.4. Control variables As with previous corporate philanthropy research (Du, Jian, Du, Feng, & Zeng, 2014;Gao et al., 2012; Li et al., 2015;Luo et al., 2017; Wang & Qian, 2011;Xu & Li, 2016), we controlled other factors that may systematically relate to corporate philanthropy. We included MKT (the ratio of a province’s GDP to its government’s budget), political connection (PC, a dummy variable that takes value of one if the chairman/CEO of the ﬁrms is or was agovernment oﬃcial, delegate of the people’s congress, or member of the political consultative conferences, and zero otherwise), ﬁrm size (SIZE, measured as the natural logarithm of the total assets of the t year), ﬁnancial leverage (LEV, measured as the ratio of total debt to total assets of the t year), return on assets (ROA, measured as the ratio of net income divided by the average total assets of the t year), sales growth (GROWTH, measured as the change ratio in operating income t year to t-1 year), cash ﬂow from operating (CFO, measured as the ratio of operating cash ﬂow to operating income of the t year), ﬁrm age (AGE, measured as the natural logarithm of the time span from the established date of the ﬁrm plus 1), ownership of ultimate owners (SHARE, measured as the sum of the products of all equity stakes held by ultimate owners along the control chains), the separation between ultimate owners control and ownership (WEDGE, measured as the ratio of ultimate owners voting rights to cash ﬂow rights). Finally, we created 21 industry dummies (eliminating the ﬁnancial industry) according to CSRC’s Guidelines for Classiﬁcation of Listed Companies to control industry eﬀects and 13 year-dummies for 2004 to 2017 to control year eﬀects. 3.3. Regression models We estimate the following models: X X X Donation ¼ α þ α Original Sin þ Control þ Industry þ Year þ ε (1) t 0 1 t1 t CHINA JOURNAL OF ACCOUNTING STUDIES 127 Donation ¼ β þ β OriginalSin þ β OriginalSin MKT þ Control t t1 t1 0 1 2 X X (2) þ Industry þ Year þ δ where Donation refers to corporate charitable giving, that is, Donation1 and Donation2. OriginSin refers to ‘original sin’ suspicion. Control refers to a serious of variables that may aﬀect corporate philanthropy. We mainly use the Tobit regression model in our analysis because the dependent variable Donation was censored at zero. Model (1) is designed to examine the eﬀect of ‘original sin’ suspicion on corporate philanthropy. The coeﬃcient of OriginalSin in model (1) measures the eﬀect of ‘original sin’ suspicion on corporate philanthropy. According to Hypothesis 1, α is expected to be signiﬁcantly positive. Model (2) is designed to examine the moderating eﬀect of formal institution on the relationship between ‘original sin’ suspicion and corporate philanthropy. According to Hypothesis 2, β is expected to be signiﬁcantly negative. All continuous variables are winsorized at the 1% and 99% level. 4. Results 4.1. Descriptive statistics Table 2 reports the summary statistics for the main variables. The mean value of Donation1 (Donation2) is 0.045 (0.022), and the minimum value and maximum value are 0 (0) and 0.556 (0.233) respectively, indicating that there is a large gap in the participation degree of charitable donations among diﬀerent enterprises. OriginalSin has a mean value of 0.268, indicating that 26.8% of the samples obtained their initial property rights through state- owned transformation, while the remaining 73.2% of our samples were founded mainly by the ultimate controllers themselves. The standard deviation of MKT is 1.770, and the maximum and minimum values are 10.799 and 2.751, respectively. It can be seen that the formal institution in diﬀerent regions of China is quite diﬀerent. The mean value of PC is 0.411, which means that more than 40% of the samples have political connections. The average SHARE value is 0.362, indicating that the ownership of ultimate controllers in private listed companies is generally high. The mean value of WEDGE is 1.200, indicating that private listed companies adopt pyramid controlling structure to enhance their control over com- panies. In addition, the rest control variables are reasonably distributed. Table 2. Descriptive statistics. Variables N Mean S.D. Min P25 Median P75 Max Donation1 6,515 0.045 0.090 0 0 0.000 0.045 0.556 Donation2 6,515 0.022 0.041 0 0 0.005 0.024 0.233 OriginalSin 6,515 0.268 0.443 0 0 0 1 1 MKT 6,515 6.679 1.770 2.751 5.045 7.238 7.893 10.799 PC 6,515 0.411 0.492 0 0 0 1 1 SIZE 6,515 21.449 0.87 19.773 20.808 21.352 22.011 23.934 LEV 6,515 0.340 0.184 0.035 0.190 0.321 0.472 0.799 ROA 6,515 0.055 0.051 −0.110 0.024 0.051 0.082 0.227 GROWTH 6,515 0.220 0.367 −0.436 0.013 0.157 0.334 2.079 CFO 6,515 0.080 0.149 −0.433 0.006 0.077 0.157 0.542 AGE 6,515 2.441 0.417 1.386 2.197 2.485 2.708 3.219 SHARE 6,515 0.362 0.160 0.067 0.239 0.346 0.475 0.756 WEDGE 6,515 1.200 0.482 0.486 1 1 1.241 3.677 128 X. LI ET AL. 4.2. Pearson correlation coeﬃcients Table 3 reports the Pearson correlation coeﬃcient among the main variables. Donation2 is positively related to OriginalSin at the 1% signiﬁcance level, supporting Hypothesis 1, while there is no positive and signiﬁcant correlation between Donation1 and suspicion of ‘original sin’. ROA, CFO, MKT and PC are all signiﬁcantly positively correlated with Donation1/Donation2, indicating that private enterprises with better performance, more abundant cash ﬂow, political connections and located in regions with more developed institutions tend to make more charitable donations. There is a positive and signiﬁcant correlation between OriginalSin and PC, which means that private enterprises with suspicion of ‘original sin’ are more inclined to establish political relations with the government in order to obtain government support to cope with external uncertainties and improve their legitimacy. OriginalSin is signiﬁ- cantly positively related to SHARE,andsigniﬁcantly negatively related to WEDGE, suggesting that the ﬁrms with suspicion of ‘original sin’ tend to use the separation of ownership and control to disperse the economic risks caused by the suspicion of ‘original sin’. In addition, the correlation coeﬃcient between two pairs of other variables is lower than 0.5, which means that introducing them into the regression model will not cause serious multicollinearity problems. 4.3. Univariate comparisons Table 4 shows the univariate comparisons of the main variables between enterprises with (OriginalSin = 1) and without (OriginalSin = 0) suspicion of ‘original sin’ in several ﬁrm characteristics. As Table 4 shows, ﬁrms with ‘original sin’ suspicion commit less resource to corporate philanthropy compared with their counterparts, which generally Table 3. Pearson correlation matrix. Variables 1234567 Donation1 1 Donation2 0.889*** 1 OriginalSin 0.009 0.032*** 1 MKT 0.043*** 0.110*** 0.087*** 1 PC 0.053*** 0.079*** 0.079*** 0.052*** 1 SIZE −0.054*** −0.031** 0.071*** −0.118*** 0.093*** 1 LEV −0.110*** −0.056*** 0.065*** 0.104*** 0.061*** 0.486*** 1 ROA 0.079*** 0.165*** −0.030** 0.005 −0.010 0.007 −0.309*** GROWTH −0.010 0.019 −0.057*** −0.040*** −0.042*** 0.135*** 0.108*** CFO 0.061*** 0.048*** 0.034*** −0.016 −0.003 −0.031** −0.241*** AGE −0.023* −0.041*** 0.053*** −0.230*** −0.039*** 0.173*** 0.020 SHARE 0.037*** 0.040*** −0.194*** −0.075*** 0.008 −0.004 −0.122*** WEDGE −0.041*** −0.016 0.214*** 0.198*** −0.035*** −0.012 0.097*** Variables 8 9 10 11 12 13 ROA 1 GROWTH 0.286*** 1 CFO 0.304*** −0.032*** 1 AGE −0.052*** −0.030** 0.043*** 1 SHARE 0.113*** 0.010 0.004 −0.065*** 1 WEDGE −0.047*** −0.040*** 0.003 0.002 −0.498*** 1 Notes: ***, **, and * indicate signiﬁcant at 1%, 5%, 10%, respectively, two-tailed. CHINA JOURNAL OF ACCOUNTING STUDIES 129 Table 4. Univariate comparisons. Variables OriginalSin =1 (N = 1747) OriginalSin =0 (N = 4768) T value Z value Mean Median SD Mean Median SD Donation1 0.046 0.013 0.087 0.045 0.008 0.092 0.723 5.108*** Donation2 0.025 0.008 0.042 0.022 0.004 0.041 2.584*** 6.093*** MKT 6.933 7.616 1.864 6.586 6.663 1.726 7.022*** 7.187*** PC 0.476 0 0.500 0.388 0 0.487 6.421*** 6.401*** SIZE 21.551 21.451 0.854 21.411 21.314 0.872 5.749*** 6.122*** LEV 0.360 0.344 0.187 0.333 0.314 0.183 5.271*** 4.980*** ROA 0.053 0.049 0.049 0.056 0.052 0.052 −2.439** −2.686*** GROWTH 0.186 0.130 0.349 0.233 0.168 0.373 −4.604*** −5.752*** CFO 0.089 0.083 0.142 0.077 0.075 0.152 2.707*** 2.911*** AGE 2.477 2.565 0.435 2.427 2.485 0.409 4.313*** 4.777*** SHARE 0.311 0.294 0.157 0.381 0.370 0.157 −15.931*** −16.173*** WEDGE 1.371 1.001 0.620 1.138 1 0.404 17.646*** 12.733*** Notes: ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed. support our Hypothesis. Firms with suspicion of ‘original sin’ are more likely to establish political connection compared with those without suspicion of ‘original sin’, which is consistent with the results in Table 3. In addition, there are signiﬁcant diﬀerences in the size, debt, proﬁtability, growth, cash ﬂow level, ﬁrm age, and ownership structure between ﬁrms with suspicion of ‘original sin’ and their counterparts. Considering that the ﬁrm characteristics can aﬀect the corporate philanthropy, it is important to introduce these variables into the regression model in subsequent multiple regression analysis. 4.4. Multivariate regression tests Table 5 reports the regression results of our hypotheses. As Model 1 and Model 3 of Table 5 shows, OriginalSin gets positive and signiﬁcant coeﬃcient (Model 1: β = 0.008, T = 2.436; Model 3: β =0.005, T = 3.070), indicating that private enterprises with suspicion of ‘original sin’ would prefer to donate more in order to gain support from government, and improve the public recognition to them, thus obtain and enhance their legitimacy. In short, Hypothesis 1 is supported. Compared with their counterparts, ﬁrms with suspicion of ‘original sin’ contribute 0.8% higher amount of revenue to corporate philanthropy on average, accounting for 17.78% (= 0.008/0.045) of overall amount of charitable donations in our samples. It highlights the economic signiﬁcance of the inﬂuences of suspicion of ‘original sin’ on corporate philanthropy. In Model 2 and Model 4, the coeﬃcients of the interaction OriginalSin*MKT are signiﬁcantly negative at least at the 10% level (Model 2: β = −0.004, T = −2.170; Model 4: β = 0.001, T = 1.709). It means that the enterprises are more likely to be recognized and protected with the improvement of the formal institutions. Thus, private enterprises with suspicion of ‘original sin’ have weaker motivation to gain legitimacy through charitable contributions. Thus, Hypothesis 2 is supported. In terms of control variables, the coeﬃcients of PC and SIZE are positive and signiﬁcant, consistent with the results of Gao et al. (2012) and Luo et al. (2017), indicating that larger ﬁrms and ﬁrms with political connections have more corporate philanthropy. However, the coeﬃcients of LEV and WEDGE are negative and signiﬁcant, suggesting that private enter- prises with higher leverage and separation of ownership and control are less inclined to conduct charitable donations, which is basically consistent with our theoretical expectation. 130 X. LI ET AL. Table 5. Results of multivariate regression. Variables Donation1 Donation2 Model 1 Model 2 Model 3 Model 4 OriginalSin 0.008** 0.035*** 0.005*** 0.014** [2.436] [2.723] [3.070] [2.505] OriginalSin × MKT −0.004** −0.001* [−2.170] [−1.709] MKT 0.003** 0.004*** 0.002*** 0.003*** [2.464] [3.046] [4.552] [4.703] PC 0.014*** 0.014*** 0.007*** 0.007*** [4.725] [4.565] [5.564] [5.442] SIZE 0.014*** 0.014*** 0.005*** 0.005*** [6.478] [6.505] [5.625] [5.647] LEV −0.065*** −0.064*** −0.010** −0.009** [−5.977] [−5.847] [−2.073] [−1.968] ROA 0.049 0.048 0.133*** 0.132*** [1.350] [1.333] [7.891] [7.881] GROWTH −0.002 −0.002 −0.001 −0.001 [−0.342] [−0.355] [−0.563] [−0.575] CFO 0.011 0.011 −0.003 −0.003 [0.907] [0.870] [−0.699] [−0.737] AGE 0.008** 0.009** 0.005*** 0.005*** [2.120] [2.283] [2.808] [2.938] SHARE 0.003 0.003 0.005 0.005 [0.265] [0.323] [1.035] [1.081] WEDGE −0.016*** −0.016*** −0.007*** −0.007*** [−4.349] [−4.292] [−3.719] [−3.671] Constant −0.302*** −0.314*** −0.148*** −0.152*** [−6.318] [−6.523] [−6.783] [−6.934] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 6,515 6,515 6,515 6,515 F value 10.654 10.415 14.015 13.718 Log likelihood 2607.669 2610.170 6418.163 6419.779 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. 4.5. Robustness test 4.5.1. Robustness test of the measurement of formal institution In the above analysis, we use the ratio of GDP to government budget to measure the formal institution. To avoid the possible eﬀect of the measurement of variables on the results of regression analysis, we also use the marketization index issued by Fan et al. (2011) as the proxy variable of the formal institution. The results are shown in Table 6.As seen in Table 6, both coeﬃcients on OriginalSin in Model 1 and Model 3 are positive at 1% signiﬁcance level. In Model 2, the interaction OriginalSin*MKT gets a negative coeﬃ- cient at 10% signiﬁcance level, and the interaction OriginalSin*MKT in Model 4 is marginally signiﬁcant and negative (p value = 0.16). Our hypotheses are still well supported and the results are robust. 4.5.2. Robustness test of endogenous problems 4.5.2.1. Propensity score matching. This study examines the eﬀect of the suspicion of ‘original sin’ on corporate philanthropy of private enterprises. However, there may be a problem of sample selection bias if we only directly make a comparison between private enterprises with suspicion of ‘original sin’ and their counterparts. Thus, we CHINA JOURNAL OF ACCOUNTING STUDIES 131 Table 6. Regression results for the alternative measurement of formal institutions. Donation1 Donation2 Variables Model 1 Model 2 Model 3 Model 4 OriginalSin 0.008*** 0.040** 0.005*** 0.017* [2.599] [2.099] [3.340] [1.955] OriginalSin × MKT −0.003* −0.001 [−1.716] [−1.406] MKT 0.003*** 0.004*** 0.002*** 0.002*** [2.931] [3.303] [4.862] [4.797] PC 0.015*** 0.014*** 0.008*** 0.008*** [4.864] [4.729] [5.787] [5.682] SIZE 0.013*** 0.013*** 0.005*** 0.005*** [6.418] [6.369] [5.534] [5.492] LEV −0.065*** −0.063*** −0.009* −0.009* [−5.927] [−5.777] [−1.958] [−1.831] ROA 0.044 0.043 0.129*** 0.128*** [1.210] [1.168] [7.652] [7.611] GROWTH −0.002 −0.002 −0.001 −0.001 [−0.390] [−0.421] [−0.662] [−0.689] CFO 0.011 0.01 −0.003 −0.003 [0.885] [0.832] [−0.718] [−0.771] AGE 0.007* 0.008* 0.004** 0.004** [1.887] [1.952] [2.423] [2.472] SHARE 0.002 0.003 0.005 0.005 [0.240] [0.337] [1.001] [1.083] WEDGE −0.016*** −0.016*** −0.006*** −0.006*** [−4.294] [−4.297] [−3.569] [−3.571] Constant −0.312*** −0.323*** −0.154*** −0.158*** [−6.493] [−6.694] [−7.062] [−7.238] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 6,515 6,515 6,515 6,515 F value 10.842 10.618 14.243 14.005 Log likelihood 2609.305 2610.992 6419.985 6421.100 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. further adopt propensity score matching method to address this problem. Speciﬁcally, we conduct PSM method on ‘original sin’ suspicion to examine whether ﬁrms with ‘original sin’ suspicion donate more. In the ﬁrst stage of propensity score matching, we choose SIZE, LEV, ROA, AGE, GROWTH, SHARE and WEDGE, PC, MKT, Industry and Year as matching variables to calculate propensity scores. Considering that the some eco- nomic factors in the region where the enterprise is located may also aﬀect the possibility of their ‘original sin’ suspicion, we further add local government ﬁscal deﬁcit rate and unemployment rate t as exogenous matching variables. To ensure the reliability of the matching results, we conduct an equilibrium test and a common support hypothesis test for the matching variables. Table 7 shows the result of the equilibrium test for PSM. As is shown in Table 7, there is a signiﬁcant diﬀerence between the treated group and the control group before matching. After matching, the diﬀerences between two groups are no longer signiﬁcant, indicating that our matching variables are balanced between the treated group and the control group after matching and these variables no longer have an explanatory power for the possibility of ‘original sin’ suspicion. The matching satisﬁes the equilibrium hypothesis. In order to test whether the common support hypothesis is satisﬁed between the treated group and the control group, we further report the probability density distribution of the 132 X. LI ET AL. Table 7. Test for common support during the procedure of PSM. Variable Sample Mean Deviation Rate Deviation reduction T test ratio(%) Treated group Control group T value p>|t| SIZE Before 21.551 21.408 16.5 82.8 5.86 0.000 After 21.551 21.526 2.8 0.84 0.402 LEV Before 0.360 0.333 14.8 79.3 5.30 0.000 After 0.360 0.366 −3.1 −0.90 0.369 ROA Before 0.053 0.056 −6.9 82.6 −2.41 0.016 After 0.0537 0.052 1.2 0.36 0.719 AGE Before 2.477 2.428 11.6 81.9 4.21 0.000 After 2.477 2.469 2.1 0.64 0.520 GROWTH Before 0.185 0.233 −13.2 82.1 −4.64 0.000 After 0.185 0.177 2.4 0.77 0.443 SHARE Before 0.311 0.381 −44.9 92.4 −16.05 0.000 After 0.311 0.316 −3.4 −1.01 0.315 WEDGE Before 1.371 1.136 44.9 84.9 17.75 0.000 After 1.371 1.406 −6.8 −1.63 0.104 MKT Before 6.933 6.5726 20.0 91.0 7.28 0.000 After 6.933 6.900 1.8 0.53 0.594 PC Before 0.476 0.386 18.2 59.8 6.53 0.000 After 0.476 0.440 7.3 2.14 0.032 DEFICIT Before 0.533 0.453 17.2 86.0 6.28 0.000 After 0.533 0.544 −2.4 −0.68 0.495 JOBLESS Before 0.032 0.030 35.4 94.4 11.84 0.000 After 0.032 0.032 2.0 0.68 0.499 0 .2 .4 .6 .8 0 .2 .4 .6 .8 Propensity Score Propensity Score Treat Control Treat Control (1) Before Matching (2) After Matching Figure 1. Probability density distribution of propensity scores of the sample. propensity scores of the treated group and the control group before and after matching. As is shown in Figure 1, the propensity score of the treated group before matching is signiﬁcantly larger than that of the control group. After matching, the propensity score distribution of the treated group shifts to the right, and the distribution pattern of the treated group and the control group are matching well, indicating that PSM corrects the distribution bias of propensity scores of two groups. The matching satisﬁes the common support hypothesis. Table 8 reports the results of the regression analysis after using PSM. As Table 8 shows, the regression coeﬃcient on OriginalSin is positive at 1% signiﬁcance level in Model 1 and Model 3. (Model 1: β = 0.012, T = 2.920; Model 3: β = 0.006, T = 3.538). In Model 2 and Model 4, the regression coeﬃcients on the interaction OriginalSin*MKT are Density 0 1 2 3 Density 0 .5 1 1.5 2 2.5 CHINA JOURNAL OF ACCOUNTING STUDIES 133 Table 8. Regression results of PSM method. Donation1 Donation2 Variables Model 1 Model 2 Model 3 Model 4 OriginalSin 0.012*** 0.055*** 0.006*** 0.023*** [2.920] [3.384] [3.538] [3.018] OriginalSin × MKT −0.006*** −0.002** [−2.694] [−2.146] MKT 0.002 0.006*** 0.002*** 0.003*** [1.172] [2.730] [2.765] [3.458] PC 0.018*** 0.017*** 0.009*** 0.009*** [4.252] [4.081] [4.887] [4.736] SIZE 0.012*** 0.011*** 0.005*** 0.004*** [4.003] [3.971] [3.345] [3.320] LEV −0.038** −0.035** −0.000 0.001 [−2.511] [−2.315] [−0.024] [0.137] ROA 0.112** 0.113** 0.153*** 0.153*** [2.140] [2.162] [6.117] [6.142] GROWTH −0.003 −0.004 −0.002 −0.002 [−0.486] [−0.587] [−0.607] [−0.692] CFO 0.016 0.014 −0.003 −0.003 [0.930] [0.847] [−0.408] [−0.499] AGE 0.006 0.007 0.005** 0.005** [1.116] [1.321] [2.107] [2.274] SHARE 0.016 0.017 0.010 0.010 [1.053] [1.087] [1.427] [1.452] WEDGE −0.011*** −0.011*** −0.004** −0.004** [−2.684] [−2.714] [−2.033] [−2.062] Constant −0.229*** −0.257*** −0.122*** −0.132*** [−3.502] [−3.908] [−3.893] [−4.207] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 2,918 2,918 2,918 2,918 F value 5.972 5.846 7.875 7.786 Log likelihood 1479.63 1483.721 3170.134 3172.768 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. negative at least at 5% signiﬁcance level (Model 2: β = −0.006, T = −2.694; Model 4: β = −0.002, T = −2.146). These regression results are highly consistent with the results in Table 5, our results are robust. 4.5.2.2. Heckman two-stage model. The Heckman two-stage selection model is also used to further control the possible eﬀect of sample selection bias. In the ﬁrst stage, the paper constructs a Probit model to shape the suspicion of ‘original sin’ of private enterprises and calculate the inverse Mills ratio (Invmills). Then, the Invmills is added to the regression model to correct the problem of omitted variables. Speciﬁcally, in the ﬁrst stage of Heckman model, we choose OriginalSin_ind, the annual average of the industry of enterprises with suspicion of ‘original sin’, natural logarithm of 1997 GDP of the place where the company was registered (GDP1997) as our instrumental variables, and all control variables in Model 1 to calculate the inverse Mills ratio. The reason why we chose these two instrumental variables is as followings. First, in 1997, the report of 15th National congress of the CPC claimed the policies of ‘grasp the big and let go the small’, ‘advance and retreat’, ‘do something and do nothing’ for state-owned enterprises. Thus, the wave of privatization for small-sized and medium-sized state-owned enter- prises was launched all over the country, especially for those provinces with ﬁnancial 134 X. LI ET AL. diﬃculties. Besides, 1997 was 7 years earlier than the starting year of our sample. Thus, the variable GDP1997 is exogenous well. In addition, due to the existence of the peer eﬀect in the industry, enterprises in the same industry will learn from each other. Therefore, the instrumental variable OriginalSin_ind can alleviate the problem of insuﬃ- cient identiﬁcation of the instrumental variables. Table 9 reports the results. Model 1 shows the results of Heckman’s approach analysis in the ﬁrst stage, and Model 2–5 shows the results of regression analysis in the second stage. In Model 1, both instrumental variables get signiﬁcant coeﬃcients. It means that these two instrumental variables are signiﬁcantly correlated with the variable of suspi- cion of ‘original sin’, indicating these two variables satisfy the requirement of being instrumental variables. The coeﬃcients on the variable OriginalSin in Modle 2 and Modle 4 are both statistically signiﬁcant and positive (Model 2: β = 0.007, t = 2.057; Model 4: β = 0.004, t = 2.605). In Model 3 and Model 5, the interaction term OriginalSin*MKT get signiﬁcant and negative coeﬃcients at the 10% level (Model 3: β = −0.004, t = −2.167; Table 9. Regression results of Heckman two-stage model. Variables OriginalSin Donation1 Donaiton2 Model 1 Model 2 Model 3 Model 4 Model 5 OriginalSin_ind 0.706*** [5.622] GDP1997 −0.334*** [−6.852] OriginalSin 0.007** 0.033*** 0.004*** 0.014** [2.057] [2.625] [2.605] [2.395] OriginalSin × MKT −0.004** −0.001* [−2.167] [−1.716] MKT 0.138*** 0.001 0.002* 0.001** 0.002*** [7.182] [0.929] [1.688] [2.554] [2.985] PC 0.156*** 0.007** 0.007* 0.004** 0.003** [4.165] [1.984] [1.840] [2.249] [2.132] SIZE 0.076*** 0.010*** 0.010*** 0.003*** 0.003*** [2.811] [4.551] [4.557] [3.444] [3.450] LEV −0.102 −0.060*** −0.058*** −0.007 −0.006 [−0.783] [−5.460] [−5.328] [−1.453] [−1.346] ROA −0.423 0.068* 0.067* 0.143*** 0.142*** [−1.004] [1.848] [1.836] [8.441] [8.436] GROWTH −0.121** 0.003 0.003 0.001 0.001 [−2.066] [0.646] [0.646] [0.695] [0.695] CFO 0.483*** −0.005 −0.006 −0.012** −0.013** [3.655] [−0.403] [−0.455] [−2.491] [−2.545] AGE 0.219*** 0.000 0.001 0.001 0.001 [4.292] [0.062] [0.201] [0.298] [0.411] SHARE −1.296*** 0.052*** 0.053*** 0.032*** 0.032*** [−9.064] [3.240] [3.311] [4.448] [4.504] WEDGE 0.323*** −0.026*** −0.026*** −0.012*** −0.012*** [7.605] [−5.444] [−5.426] [−5.335] [−5.324] Invmills −0.048*** −0.049*** −0.027*** −0.027*** [−3.650] [−3.689] [−4.545] [−4.577] Constant −0.596 −0.144** −0.154*** −0.062** −0.066** [−0.871] [−2.430] [−2.580] [−2.329] [−2.453] Year ﬁxed eﬀect Yes Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes Yes No. of observations 6,462 6,462 6,462 6,462 6,462 F value 835.129 10.509 10.275 13.896 13.607 Log likelihood −3314.500 2576.916 2579.401 6357.601 6359.225 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. CHINA JOURNAL OF ACCOUNTING STUDIES 135 Model 4: β = −0.001, t = −1.716). These results also support the hypothesis of this paper, indicating that our ﬁndings are still robust after adopting the Heckman two-stage approach to address the sample selection bias. 4.5.2.3. Placebo test. We run the following placebo test to ensure our results are not plagued by mechanical correlation and unobserved variables. Speciﬁcally, we randomly assign a pseudo company for each private enterprise with ‘original sin’ suspicion, and then rerun our regression models. Figure 2 shows the distribution of t-values of the coeﬃcient of the indicator OriginalSin_Random after 1,000 placebo tests. In Graph (1), the dependent variable is Donation1.We ﬁnd that the distribution of t-values of the coeﬃcient estimates of the variable OriginalSin_random lies around zero and the t-value of the true eﬀect based on Model 1 of Table 5 (t = 2.436) lies outside the 95% range. In Graph (2), we also ﬁnd similar results for dependent variable Donation2. Those results suggest that random assignment of enterprises with ‘original sin’ suspicion does not have a signiﬁcant impact on corporate philanthropy. Therefore, our results are not aﬀected by unobserved variables and mechan- ical correlation and the conclusions are robust. 4.5.2.4. Robustness test of estimation method. Our conclusions are based on analysis of unbalanced panel data. Since there is the possibility for panel data to suﬀer from time-serial autocorrelation, we conducted the following clustering regression by year to alleviate this problem. The results are presented in Table 10. As shown in Table 10, the coeﬃcients of OriginalSin are positive and signiﬁcant in Model 1 and Model 3 (Model 1: β = 0.008, t = 3.583; Model 3: β = 0.005, t = 4.029), while in Model 2 and Model 4, the coeﬃcients of OriginalSin are both negative and signiﬁcant at the 10% level (Model 2: β = −0.004, t = −2.130; Model 4: β = −0.001, t = −1.826). Those results are consistent with results shown in Table 5 and our conclusions are robust to alternative estimation method. (1) Donation1 as dependent variable (2) Donation2 as dependent variable Figure 2. Distribution of t-values of the coeﬃcient estimates of the independent variable originalsin_random. 136 X. LI ET AL. Table 10. Regression results by clustering at year level. Variables Donation1 Donation2 Model 1 Model 2 Model 3 Model 4 OriginalSin 0.008*** 0.035*** 0.005*** 0.014*** [3.585] [2.838] [4.029] [2.783] OriginalSin × MKT −0.004** −0.001* [−2.130] [−1.826] MKT 0.003*** 0.004*** 0.002*** 0.003*** [2.709] [2.893] [4.923] [4.558] PC 0.014*** 0.014*** 0.007*** 0.007*** [6.424] [6.177] [6.537] [6.304] SIZE 0.014*** 0.014*** 0.005*** 0.005*** [7.959] [8.034] [7.079] [7.120] LEV −0.065*** −0.064*** −0.010** −0.009* [−5.374] [−5.407] [−2.024] [−1.958] ROA 0.049* 0.048* 0.133*** 0.132*** [1.670] [1.672] [7.296] [7.368] GROWTH −0.002 −0.002 −0.001 −0.001 [−0.627] [−0.647] [−0.880] [−0.906] CFO 0.011 0.011 −0.003 −0.003 [1.018] [0.983] [−0.702] [−0.751] AGE 0.008*** 0.009*** 0.005*** 0.005*** [3.442] [3.496] [3.186] [3.199] SHARE 0.003 0.003 0.005 0.005 [0.238] [0.292] [0.833] [0.874] WEDGE −0.016*** −0.016*** −0.007*** −0.007*** [−6.528] [−6.499] [−5.451] [−5.443] Constant −0.302*** −0.314*** −0.148*** −0.152*** [−9.287] [−9.155] [−9.017] [−8.848] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 6,515 6,515 6,515 6,515 Log likelihood 2607.669 2610.17 6418.163 6419.779 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. 4.6. Further analysis 4.6.1. Results of the eﬀects of political connections Due to the fact such as history, economy and institution, the private property rights of private enterprises may not be well protected by formal institution. As a result, there are more uncertainties and risks faced with private enterprises in China. As to private entrepreneurs, they may adopt various ways to alleviate those risks. Political connec- tions, as an alternative mechanism of legal protection, can protect private enterprises in a less developed formal institutional environment with insuﬃcient protection of prop- erty rights (Chen & Wang, 2013; Park & Luo, 2001; Wu, Wu, & Liu, 2008; Yang, Feng, & Song, 2014). Therefore, private enterprises generally tend to actively establish and maintain political ties with the government and the oﬃcials. Since political connections contribute to meet the needs of private entrepreneurs to cope with various uncertain- ties, will the inﬂuence of ‘original sin’ suspicion on corporate philanthropy of private enterprises change with the diﬀerence of political connections? To answer this question, we further analyze the potential eﬀect of political connections on the relationship between the suspicion of ‘original sin’ and corpo- rate philanthropy. Table 11 reports the results. As seen in Table 11, in the non- political connections subsample of Model 1 and Model 3, the regression coeﬃcients CHINA JOURNAL OF ACCOUNTING STUDIES 137 Table 11. Results of cross-section analysis according to political connections. Variable Donation1 Donation2 PC =0 PC =1 PC =0 PC =1 Model 1 Model 2 Model 3 Model 4 OriginalSin 0.012** 0.001 0.006*** 0.003 [2.560] [0.315] [2.760] [1.293] SIZE 0.018*** 0.006** 0.006*** 0.003* [6.124] [2.044] [4.964] [1.801] LEV −0.067*** −0.062*** −0.009 −0.009 [−4.931] [−3.277] [−1.575] [−1.135] ROA 0.026 0.083 0.109*** 0.169*** [0.559] [1.440] [5.120] [6.201] GROWTH −0.001 −0.003 −0.001 −0.002 [−0.183] [−0.433] [−0.318] [−0.523] CFO 0.001 0.019 −0.005 −0.003 [0.075] [0.962] [−0.915] [−0.425] AGE 0.008 0.016*** 0.002 0.011*** [1.503] [2.718] [0.888] [4.225] SHARE 0.004 0.001 0.003 0.008 [0.292] [0.047] [0.417] [1.175] WEDGE −0.016*** −0.014** −0.005** −0.008** [−3.548] [−2.026] [−2.502] [−2.294] MKT 0.005*** 0.000 0.003*** 0.002*** [3.558] [0.170] [4.305] [2.680] Intercept −0.449*** −0.049 −0.182*** −0.059* [−6.691] [−0.693] [−6.216] [−1.792] Year ﬁxed eﬀect Yes Yes Yes Yes Industry ﬁxed eﬀect Yes Yes Yes Yes No. of observations 3,836 2,679 3,836 2,679 F value 5.750 6.009 7.567 7.639 Log likelihood 1174.910 1505.632 3321.334 3161.327 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. on OriginalSin are positive at least at 5% level (Model 1: β = 0.012, t = 2.560, Model 3: β = 0.006, t = 2.760). However, in the political connections subsample of Model 2 and Model 4, the regression coeﬃcients on OriginalSin are positive but not sig- niﬁcant (Model 2: β = 0.001, t = 0.315; Model 2: β = 0.003, t = 1.293). Further, we use Fisher’s Permutation test method to test the diﬀerence of the coeﬃcients between two groups. The coeﬃcient on OriginalSin in the non-political connections subsample (PC = 0)issigniﬁcantly larger than that in the political connections subsample (PC = 1) at 10% signiﬁcance level (p = 0.066) if Donation1 wasasthe dependent variable. The coeﬃcient on OriginalSin in the non-political connections subsample (PC = 0) is larger than that in the political connections subsample (PC = 1) at 17% signiﬁcance level (p = 0.168) at the marginal signiﬁcance level if Donation2 was as the dependent variable. The above results show that the positive eﬀect on corporate philanthropy of private enterprises mainly occurs in ﬁrms without political connections. Compared with private entrepreneurs with suspicion of ‘original sin’ that have political connections, those ﬁrms which do not have political connections face worse property rights protection and greater uncertainties. Therefore, entrepreneurs of these enterprises have a stronger incentive to establish and improve the legitimacy of the enterprises through strategic charitable donations. 138 X. LI ET AL. 4.6.2. Analysis of the impact of entrepreneurs showing up on the hurun rich list Enterprises with suspicion of ‘original sin’ make more charitable donations to reduce or eliminate the risks and regulatory and social sanctions brought by the public and government departments’ doubts on its legitimacy. The enterprises’ degree of the legitimacy which the public and the government question are positively correlated with the degree of attention received by the enterprises. Thus, we expect that the enterprises with suspicion of ‘original sin’ would make more charitable donations if they receive more attention from the public. To examine the potential eﬀect of public attention, we focus on whether entrepreneurs show up on the Hurun Rich List. Through analyzing the changes of charitable donations between enterprises with suspicion of ‘original sin’ and within suspicion of ‘original sin’ after they ﬁrst show up on the Hurun Rich List, we could verify our assumption of this paper. The reason why we chose showing up on Hurun Rich List as the proxy variable of public attention is that the Hurun Rich List can attract the numerous media reports. In addition, billionaires and their controlling enterprises can receive extensive attention from the public, which further triggers social public discussion about their wealth, leading to a sharp rise in the risks and costs of the enterprises faced (Ye et al., 2012). Comparing diﬀerences between the charitable donation behaviours of the sample companies before (LIST =0) and after (LIST = 1) getting on the list, we analyze the changes of the charitable donation behaviours of enterprises with suspicion of ‘original sin’ when there is more public attention. Table 12 shows regression results. As is shown in Table 12, the coeﬃcient of OriginalSin in Model 2 is signiﬁcantly positive in Model 2 after showing up on the rich list (LIST = 1), while the coeﬃcient of OriginalSin is not signiﬁcant in Model 1 before showing up on the rich list (LIST =0) (Model 1: β = −0.006, t = −0.725; Model 2: β = 0.008, t = 1.774). Similarly, the coeﬃcient of OriginalSin in Model 4 is marginally signiﬁcantly positive, while the coeﬃcient of OriginalSin is not signiﬁcant in Model 3 (Model 3: β = 0.001, t = 0.151; Model 4: β = 0.004, t = 1.561). These results show that the enterprises with suspicion of ‘original sin’ are faced with more questions from public, which triggers more risks and costs of political, regulatory and social sanctions when receiving more public attention. Thus, entrepre- neurs tend to donate more after showing up on the rich list to reduce the negative impact on the suspicion of ‘original sin’. 5. Conclusions and implications Due to institutional defections, Chinese private enterprises have always been faced with the problem of ‘original sin’ suspicion, which poses a threat on the legitimacy of these private enterprises. Being faced with this problem, what method will private entrepreneurs take to mitigate the potential negative eﬀect arising from ‘original sin’ suspicion? We argued that corporate philanthropy is likely to be an eﬀective strategic tool for private enterprises with ‘original sin’ suspicion to obtain and improve their legitimacy from the public and the government. In other words, ‘original sin’ suspi- cion will aﬀect the philanthropic behaviour of private enterprises. We used hand- collected data to test whether ‘original sin’ suspicion aﬀects the corporate philan- thropy of Chinese listed private enterprises and investigated the moderating eﬀect of formal institutions. After various robustness checks, we ﬁnd that there is a positive CHINA JOURNAL OF ACCOUNTING STUDIES 139 Table 12. Results of cross-section analysis according to entrepreneurs showing up on the Hurun rich list. Variables Donation1 Donation2 Model 1 Model 2 Model 3 Model 4 LIST =0 LIST =1 LIST =0 LIST =1 OriginalSin −0.006 0.008* 0.001 0.004 [−0.725] [1.774] [0.151] [1.561] MKT 0.005* 0.002 0.003*** 0.003*** [1.951] [1.361] [2.788] [3.845] PC 0.027*** 0.010*** 0.010*** 0.007*** [3.551] [2.590] [3.176] [3.686] SIZE 0.003 0.017*** −0.002 0.007*** [0.395] [5.388] [−0.750] [4.576] LEV −0.084*** −0.075*** −0.007 −0.017** [−3.058] [−5.154] [−0.586] [−2.400] ROA −0.038 0.087** 0.108*** 0.155*** [−0.419] [2.033] [2.808] [7.515] GROWTH −0.000 −0.008 −0.001 −0.004 [−0.028] [−1.526] [−0.205] [−1.404] CFO 0.017 −0.003 0.001 −0.008 [0.636] [−0.189] [0.046] [−1.206] AGE −0.002 0.014** 0.002 0.006** [−0.234] [2.577] [0.595] [2.222] SHARE −0.023 −0.013 −0.01 −0.003 [−0.805] [−0.867] [−0.838] [−0.441] WEDGE −0.037*** −0.012** −0.020*** −0.006** [−3.289] [−2.349] [−4.061] [−2.294] Constant 0.059 −0.396*** 0.076 −0.195*** [0.375] [−5.615] [1.142] [−5.763] Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 1,296 2,943 1,296 2,943 Chi2 142.999 329.311 186.517 472.701 Log likelihood 361.820 1597.495 1139.415 3291.784 Notes: (1) ***, **, and * indicate signiﬁcance at 1%, 5%, 10%, respectively, two-tailed; (2) t-statistics, based on heteroscedasticity adjustment, are in brackets. relationship between ‘original sin’ suspicion of private enterprises and corporate philanthropy. Formal institutions could attenuate the positive eﬀect of ‘original sin’ suspicion on corporate philanthropy. In addition, we also ﬁnd that the positive eﬀect of ‘original sin’ suspicion on corporate philanthropy is more pronounced for private enterprises without political connections and after the entrepreneurs showing up on the Hurun Rich List. Our study has several implications for practices and policy-makers. First, under the background that the ‘original sin’ problem is a very important and realistic problem faced by private enterprises, the government and the public should face up to the ‘original sin’ problem of private enterprises, and thus reduce the instrumental chari- table donations made by private enterprises and make corporate philanthropy play a role in improving social welfare. Second, on 27 November 2016, the central committee of the communist party of China issued ‘the state council approved the opinions of perfecting the system of property rights in accordance with the law to protect property’. It stressed ‘to face up to the problems accompanied with the development of the private enterprises and handle them according to the law since the reform and opening up’, which alleviate concerns for vast number of 140 X. LI ET AL. enterprises, especially for those who may be regarded with ‘original sin’. Our study indicates that facing up to and properly handling such problem is of great signiﬁ- cance to the long-term development of private enterprises, which to some extent conﬁrms the important positive value of the promulgation of the opinions. Furthermore, to optimize the environment of enterprise management system and reduce the uncertainty of environment that private enterprises face by constraining government behaviour and reducing administrative intervention form the govern- ment, will improve the stability and sustainability of the operating environment faced by the private enterprises as well as weaken their instrumental motivation of corpo- rate philanthropy. Finally, we have to acknowledge that our study has some limitations that warrant further research. Although corruption and loss of state-owned in the process of state- owned enterprise restructuring and privatization are concrete and practical dimensions to describe suspicion of ‘original sin’, there are still some other problems we need to further recognize. In reality, because of the imperfect system in the early development stage of the economy, there are still many obscure questions about ‘original sin’, such as playing ‘edge ball’ and seek loopholes in the formal institutions. These are not involved in our study and need to further explore. At the same time, the microeconomic consequences of the suspicion of ‘original sin’ need to be paid more attention to in the future. Acknowledgments We greatly thank for valuable comments and suggestions from the Senior Editor Prof. Kangtao Ye and two anonymous reviewers. All remaining errors are our own. Disclosure statement No potential conﬂict of interest was reported by the authors. Funding We acknowledge ﬁnancial support from the National Natural Science Foundation of China [Grant No. 71790602 and 71572160]. References Brammer, S., & Millington, A. (2005). Corporate reputation and philanthropy: An empirical analysis. Journal of Business Ethics, 61(1), 29–44. Brown, W.O., Helland, E., & Smith, J.K. (2006). Corporate philanthropic practices. Journal of Corporate Finance, 12(5), 855–877. Campbell, L., Gulas, C.S., & Gruca, T.S. (1999). Corporate giving behavior and decision-maker social consciousness. Journal of Business Ethics, 19(4), 375–383. Chen, & Wang. (2013). Family involvement, political connection and institutional environment: Evidence from Chinese private enterprises. Management World,(10), 130–141. (in Chinese). Chen, J.C., Patten, D.M., & Roberts, R.W. (2008). Corporate charitable contributions: A corporate social performance or legitimacy strategy? Journal of Business Ethics, 82(1), 131–144. CHINA JOURNAL OF ACCOUNTING STUDIES 141 Dai, Y., Pan, Y., & Feng, S. (2014). Are Chinese enterprises’ charitable donations “political contribu- tions”? Evidence from the replacements of the municipal party secretaries. Economic Research, 49(2), 74–86. (in Chinese). Dai, Y., Peng, Z., & Pan, Y. (2016). Corporate charitable donations: Self-Redemption under the risk of litigation. Journal of Xiamen University (Arts & Social Sciences),(2), 122–131. (in Chinese). Du, X. (2015). Is corporate philanthropy used as environmental misconduct dressing? Evidence from Chinese family-owned ﬁrms. Journal of Business Ethics, 129(2), 341–361. Du, X., Guo, J., & Lei, Y. (2010). Political connections and donation of private listed companies: Measurement methods and empirical evidence. Finance and Trade Research, 21(1), 89–99. (in Chinese). Du,X.,Jian,W.,Du,Y.,Feng,W., & Zeng,Q. (2014). Religion, the nature of ultimate owner, and corporate philanthropic giving: Evidence from China. Journal of Business Ethics, 123(2), 235–256. Fan, G., Wang, X., & Zhu, H. (2011). NERI index of marketization of China’s Provinces: The report on the relative process of marketization of each region in China. Beijing: Economic Science Press. (in Chinese). Feng, J., & Cheng, W. (2010). Economic motivations and determinants of corporate philanthropy: Overview and implications. Economic Perspectives,(9), 138–141. (in Chinese). Fu, C., & Ji, L. (2017). Litigation risk and corporate charitable giving: An explanation from the perspective of reputation insurance. Nankai Business Review, 20(2), 108–121. (in Chinese). Galaskiewicz, J. (1997). An urban grants economy revisited: Corporate charitable contributions in the Twin Cities, 1979–81, 1987–89. Administrative Science Quarterly, 42(3), 445–471. Gao, F., Faﬀ, R., & Navissi, F. (2012). Corporate philanthropy: Insights from the 2008 Wenchuan earthquake in China. Paciﬁc-Basin Finance Journal, 20(3), 363–377. Gao, Y., Chen, Y., & Zhang, Y. (2012). “Red scarf” or “green scarf”: Research on the motivations of the charitable donation from private enterprise. Management World, 38(8), 106–114. (in Chinese). Gardberg, N.A., & Fombrun, C.J. (2006). Corporate citizenship: Creating intangible assets across institutional environments. Academy of Management Review, 31(2), 329–346. Gautier, A., & Pache, A.C. (2015). Research on corporate philanthropy: A review and assessment. Journal of Business Ethics, 126(3), 343–369. Guo, J. (2012). Corporate donation in China: Voluntary or requested? Evidence from listed companies. Journal of Finance and Economics,(8), 50–60. (in Chinese). Haley, U.C. (1991). Corporate contributions as managerial masques: Reframing corporate contribu- tions as strategies to inﬂuence society. Journal of Management Studies, 28(5), 485–510. Hornstein, A.S., & Zhao, M. (2018). Reaching through the fog: Institutional environment and cross- border giving of corporate foundations. Strategic Management Journal, 39(10), 2666–2690. Jensen, M.C., & Meckling, W.H. (1976). Theory of the ﬁrm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305–360. Jia, M., Xiang, Y., & Zhang, Z. (2015). The reconstruction of business and government relationship: The usage of business corruption or philanthropic giving?. Nankai Business Review, 18(5), 4–17. (in Chinese). Jia, M., & Zhang, Z. (2010). Does political connection inﬂuence corporate philanthropy?. Management World,(4), 99–113. (in Chinese). Koehn, D., & Ueng, J. (2010). Is philanthropy being used by corporate wrongdoers to buy good will? Journal of Management & Governance, 14(1), 1–16. Lev, B., Petrovits, C., & Radhakrishnan, S. (2010). Is doing good good for you? How corporate charitable contributions enhance revenue growth. Strategic Management Journal, 31(2), 182–200. Li, S., Chen, X., & Song, X. (2016). The generosity of the poor: A study of strategic motivation of corporate philanthropy. Management World,(5), 116–127. (in Chinese). Li, S., Lu, Q., & Song, X. (2012). Corporate donation among China’s money-losing enterprises. China Industrial Economics,(8), 148–160. (in Chinese). 142 X. LI ET AL. Li, S., Song, X., & Wu, H. (2015). Political connection, ownership structure, and corporate philan- thropy in china: A strategic-political perspective. Journal of Business Ethics, 129(2), 399–411. Li, S., & Xie, X. (2014). Corporate social responsibility, political relationship and debt ﬁnancing of private enterprises: Evidence from Chinese capital market. Nankai Business Review, 17(6), 30–40. Li, W., Wang, P., & Xu, Y. (2015). Philanthropy, political connection and debt ﬁnance: Reciprocal behavior of governments and private enterprises. Nankai Business Review, 18(1), 4–14. (in Chinese). Li, Z., Tang, X., & Lian, Y. (2016). The puzzle of Chinese private enterprises’ corporate social responsibility. Management World,(9), 136–148. (in Chinese). Liu, H., Lin, L., & Huang, S. (2017). Oﬃcial turnover and the rise and fall of local enterprises: An evidence from China’s prefecture-level cities. China Industrial Economics,(1), 62–80. (in Chinese). Luo, D., Liao, J., & Wang, J. (2016). Local oﬃcials, turnover and corporate risk: Evidence from Chinese listed ﬁrms. Economic Research, 51(5), 130–142. (in Chinese). Luo, J., Kaul, A., & Seo, H. (2018). Winning us with triﬂes: Adverse selection in the use of philanthropy as insurance. Strategic Management Journal, 39(10), 2591–2617. Luo, J.H., Xiang, Y., & Zhu, R. (2017). Military top executives and corporate philanthropy: Evidence from China. Asia Paciﬁc Journal of Management, 34(3), 725–755. Masulis, R.W., & Reza, S.W. (2015). Agency problems of corporate philanthropy. Review of Financial Studies, 28(2), 592–636. Pan, Y., Wen, R., & Liu, S. (2017). The selﬁsh goodwill: New evidence from corporate philanthropy in typhoon. China Industrial Economics,(05), 135–153. (in Chinese). Park, S.H., & Luo, Y. (2001). Guanxi and organizational dynamics: Organizational networking in Chinese ﬁrms. Strategic Management Journal, 22(5), 455–477. Peng, F., & Fan, Z. (2016). Tax incentives, donation costs and enterprises’ charitable donations. The Journal of World Economy, 39(7), 144–167. (in Chinese). Petrenko, O.V., Aime, F., Ridge, J., & Hill, A. (2016). Corporate social responsibility or CEO narcis- sism? CSR motivations and organizational performance. Strategic Management Journal, 37(2), 262–279. Porter, M.E., & Kramer, M.R. (2002). The competitive advantage of corporate philanthropy. Harvard Business Review, 80(12), 56–68. Qian, X., & Xu, Y. (2014). Oﬃcial turnover, political identity and the risk taking of private listed ﬁrms. China Economic Quarterly, 13(4), 1437–1460. (in Chinese). Sánchez, C.M. (2000). Motives for corporate philanthropy in El Salvador: Altruism and political legitimacy. Journal of Business Ethics, 27(4), 363–375. Shan, L., Gan, L., & Zheng, T. (2008). Corporate donations and economic incentives: An empirical study based on corporate donations following the 5.12 earthquake in China. Economic Research, 43(11), 51–61. (in Chinese). Shiu, Y.M., & Yang, S.L. (2017). Does engagement in corporate social responsibility provide strategic insurance-like eﬀects? Strategic Management Journal, 38(2), 455–470. Su, J., & He, J. (2010). Does giving lead to getting? Evidence from Chinese private enterprises. Journal of Business Ethics, 93(1), 73–90. Tan, W., & Xu, L. (2015). Agency cost, corporate governance, and corporate donations. Economic Management, 37(9), 51–62. (in Chinese). Tang, S., Wen, D., & Sun, Z. (2017). “original sin” suspicion and the earnings quality of private listed ﬁrms. Management World,(8), 106–122. (in Chinese). Tang, Y., Zuo, J., & Li, H. (2014). The impact of institutional environment transition on corporate philanthropic behavior. Economic Research, 49(2), 61–73. (in Chinese). Wang, H., & Qian, C. (2011). Corporate philanthropy and corporate ﬁnancial performance: The roles of stakeholder response and political access. Academy of Management Journal, 54(6), 1159–1181. Wang, J., & Pan, X. (2018). The inﬂuence of charitable donation on audit opinion: An empirical analysis based on the selﬁsh motives of charitable donations. Auditing Research,(3), 87–94. (in Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 143 Wang, K., Jin, Z., & Jiao, J. (2014). Strategic group membership, corporate charitable contribution and performance: Evidence from Chinese real estate industry. Nankai Business Review, 17(6), 53–62. (in Chinese). Wang, X., Fan, G., & Yu, J. (2017). NERI index of marketization of China’s Provinces. Beijing: Economic Science Press (in Chinese). Wang, X., Li, Y., & Li, F. (2015). Research on corporate social responsibility and the eﬀectiveness of executive compensation incentive: Strategic motivation or shirking excuse. Accounting Research, (10), 51–58. (in Chinese). Wang, Y., Pan, Y., & Huang, L. (2016). Corporate charitable donations: Public behavior or other intentions. Finance and Trade Research, 27(1), 133–141. (in Chinese). Wu, W., Wu, C., & Liu, X. (2008). Political connection and market valuation: Evidence from China individual-controlled listed ﬁrms. Economic Research,(7), 130–141. (in Chinese). Xiu, Z., & Zhou, Z. (2016). Regional happiness, social capital and corporate philanthropic giving. Journal of Management Science, 29(2), 146–160. (in Chinese). Xu, N., & Li, Z. (2016). CEOs’ poverty experience and corporate philanthropy. Economic Research, 51 (12), 133–146. (in Chinese). Yang, Z., Feng, L., & Song, Z. (2014). The political connection eﬀect between the institutional environment and state-ownership. Review of Investment Studies, 33(7), 144–157. (in Chinese). Ye, Q., Li, Z., & Li, G. (2012). Does the rich list aﬀect the quality of accounting information?: Perspective of political cost. Management World,(1), 104–120. (in Chinese). Zhang, M., Ma, L., & Zhang, W. (2013). The eﬀect of ﬁrm-government bonding on corporate charitable donations: Based on empirical evidence of listed companies in China. Management World,(7), 163–171. (in Chinese). Zhang, R., Rezaee, Z., & Zhu, J. (2010). Corporate philanthropic disaster response and ownership type: Evidence from Chinese ﬁrms’ response to the Sichuan earthquake. Journal of Business Ethics, 91(1), 51–63. Zhang, R., Zhu, J., Yue, H., & Zhu, C. (2010). Corporate philanthropic giving, advertising intensity, and industry competition level. Journal of Business Ethics, 94(1), 39–52. Zheng, G., & Xu, Y. (2011). Corporate philanthropy, corporate governance and shareholder wealth. Nankai Business Review, 14(2), 92–101. (in Chinese).

Journal

China Journal of Accounting Studies – Taylor & Francis

Published: Jan 2, 2019

Keywords: ‘Original sin’ suspicion; private enterprises; corporate philanthropy; institutional environment

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises

‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises

‘Original sin’ suspicion, institutional environment, and corporate philanthropy in private enterprises

References (65)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies