Performance commitment in M&As and stock price crash risk

Jingjing LI; Yingwen GUO; Minghai WEI

doi:10.1080/21697213.2019.1695948

Performance commitment in M&As and stock price crash risk

LI, Jingjing; GUO, Yingwen; WEI, Minghai 2019-07-03 00:00:00 CHINA JOURNAL OF ACCOUNTING STUDIES 2019, VOL. 7, NO. 3, 317–344 https://doi.org/10.1080/21697213.2019.1695948 ARTICLE a b c Jingjing LI , Yingwen GUO and Minghai WEI a b School of Economics and Management, Harbin Institute of Technology, Shenzhen, China; School of Business, Nanjing University, Nanjing, China; Center for Accounting, Finance and Institutions, Sun Yat-sen University, Guangzhou, China ABSTRACT KEYWORDS Performance commitment This paper investigates the impact of performance commitment in M&A; stock price crash in M&A transactions on acquiring ﬁrms’ future crash risk. We ﬁnd risk; agency conﬂict; a positive relation between performance commitment and acquir- signalling theory; costless ing ﬁrms’ future crash risk, and this result is more pronounced in signal related-party M&A deals. These ﬁndings are consistent with our prediction that performance commitment regulation may bring about negative consequences by providing the parties with private information an opportunity to overstate the values of inferior target assets; the stock price will suddenly drop when the accumulated hidden bad news release to the market. We further systematically discuss the shortcomings in the theories and regulation related to performance commitment in M&As. We document that the current performance commitment contract is designed as a costless pro- mise with low default cost, resulting in the information insiders make use of the regulation loopholes to aggressively expropriate from less-informed minority shareholders. 1. Introduction Performance commitment regulation (PC hereafter) in mergers and acquisitions (M&As) is an innovative policy made by China authorities. The original intention of the regulation is to constrain the opportunistic behaviours of those with information advantages and improve information transparency. Ideally, performance commitment is expected to alleviate information asymmetry, thereby reducing moral hazard and adverse selection problems in M&As (Cadman, Carrizosa, & Faurel, 2014; Cain, Denis, &Denis, 2011;Lv&Han, 2014; Ragozzino & Reuer, 2009). Moreover, the PC contract is expected to increase the re-negotiation eﬃciency and reduce contract frictions caused by information incompleteness or uncertainty. However, the current performance commitment regulation in M&As has been criticised by academicians, media and investors. They claim that PC regulation provides acquirees with an opportunity to overstate the values of inferior target assets. The less-sophisticated small investors in Chinese capital market cannot distinguish ‘opportunistic’ performance CONTACT Minghai WEI mnswmh@mail.sysu.edu.cn Center for Accounting, Finance and Institutions, Sun Yat-sen University, Guangzhou, China Paper accepted by Kangtao Ye. This article has been republished with minor changes. These changes do not impact the academic content of the article. © 2019 Accounting Society of China 318 J. LI ET AL. commitments from reliable performance commitments, which may result in a stock price crash in the future when the accumulated bad news is ﬁnally released. Most prior studies on performance commitment contracts focus on the short-term eﬀects of performance commitment. Lv and Han (2014) found a positive cumulative abnormal return (CAR) around the announcement day of performance commitment in M&As and concluded that the PC contract works as an eﬀective signalling mechanism in M&A deals. Pan, Qiu, and Yang (2017) investigated the eﬀects of performance commitments on the ﬁnancial performances of target ﬁrms in the ﬁrst year after acquisition. Given most of the existing studies examine the short- term eﬀect of performance commitment, it is diﬃcult to distinguish the investor irrationality from the signalling eﬀect of performance commitment. In this paper,weexaminetheimpactof performance commitment on future stock price crash risk of acquiring ﬁrms, using the sample from 2008 to 2017. We ﬁnd that the performance commitment increases the acquiring ﬁrms’ future crash risk. We explore the theoretical and institutional reasons why the performance commitment regulation causes negative consequences in the Chinese capital mar- ket. The ﬁrst explanation is based on signalling theory (Spence, 1973). Signalling theory implies that the eﬀective signal should be costly. However, the default cost of performance commitment in M&As is nothighenoughto constrain thepromi- sers’ opportunistic behaviours. In other words, the performance commitment in M&As may become a costless commitment or cheap talk. Secondly, with respect to an institutional environment, in M&A transactions, major shareholders of listed companies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the agency conﬂicts between them and minority shareholders. Therefore, performance commitment does not work as a signalling mechanism to reduce information asymmetry, but drive the parties with private information to hide the true value of the target assets; the stock price will suddenly drop when the accumulated hidden bad news release to the market. We contribute to the literature in three aspects. First, to our best knowledge, this is the ﬁrst paper to systematically discuss the shortcomings in the theories and regulation related to performance commitment in M&As. Secondly, there is a growing research examining the determinants of ﬁrms’ stock price crash risk. This paper extends previous studies by providing the evidence that PC in M&As could increase the acquiring ﬁrms’ future crash risk. Third, policymakers and regulators may be interested in our ﬁndings. We argue that the current performance commitment regulation may not be suitable for the China market in which type II agency conﬂict is dominated. Our ﬁndings suggest that policy design and implementation need to ﬁt with China institutional environment. Our results shed some light on how to revise the PC regulation in M&As. 2. Institutional backgrounds In 2008, China Securities Regulatory Commission (CSRC) issued ‘The Administration Measures for Signiﬁcant Asset Restructuring of Listed Companies’ (hereafter The CHINA JOURNAL OF ACCOUNTING STUDIES 319 Administration Measures) to alleviate the information asymmetry and valuation risk in M&As. For the M&A transactions are valuated based on discounted future- expected cash ﬂows, CSRC explicitly stipulated that the acquirees should sign performance commitment contracts with acquirers so that acquirers can reliably evaluate the values of targets based on targets’ future-expected performances. In addition, the target shareholders need to sign a compensation agreement with the acquirer. The target shareholders need to pay back the cash or stocks if the realised performance of the target misses their promised performance goal. During the pre-acquisition period, to get a high premium in acquisitions, target shareholders have strong incentives to camouﬂage their real situations, leading to valuation risk. The authorities expect the performance commitment regulation to work as a mechanism to reduce valuation risk and protect minority shareholders from low-quality M&A deals. The ‘Q&A of Listed Companies’ Mergers and Acquisitions’ (p.128) compiled by the Shenzhen Stock Exchange clearly states that ‘. . . high valuations need to be based on a matched high-performance commitment . . .’. Target shareholders need to make a tradeoﬀ between the compensation risk and valuation gain. However, the current performance commitment regulation in M&A has considerate theoretical ‘loopholes’, which may drive the target shareholders to camouﬂage their real situations at the expense of triggering the compensation clause. We analyse the ‘loopholes’ of PC regulation in four aspects. First, in terms of legal nature, performance commitment contract in M&As is a kind of anonymous contract without name prescribed in Contract Law. As a nameless contract, it does not have compulsory enforcement force, and thus it cannot provide a strong protection to the acquirer shareholders. Second, based on signalling theory, we argue that performance commitment contract in M&As is a costless commitment, rather than costly signal. According to current regulation, when the acquiree fails to realise the performance goal, they are only required to compensate for the diﬀerence between the actual realised earnings and the promised earnings, rather the diﬀerence in valuations. The default cost is too low to constrain acquirees’ incentive to overstate their performance forecast. Therefore, the design of performance commitment regulation ignores costly signal- ling theory which indicates that eﬀective signals must be costly (Spence, 1973). The theoretical loopholes of current performance commitment regulation may result in opportunistic behaviours of the parties with information advantages. Third, with respect to agency conﬂict, agency conﬂicts are intensiﬁed by the concentrated ownership structure in China market (Allen, Qian, & Qian, 2005;Ang, Cole, & Lin, 2000; Jiang,Lee,&Yue, 2010). The major shareholders of listed compa- nies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the aggravated agency conﬂicts between them and min- ority shareholders. Fourth, the existing regulatory policies are insuﬃcient, thereby giving the parties with private information a space for opportunistic behaviour. There is a limited restriction on the behaviours of acquiring ﬁrms’ original (controlling) shareholders 320 J. LI ET AL. and targets’ original shareholders in the post-commitment period. Therefore, these information insiders are capable to obtain excess beneﬁtbymakinguse of those regulatory defects. Earnout is one of the debt-like claims proposed by the investors to address adverse selection issues in western countries, such as the UK and US. Previous literature suggests that the earnout contract can reduce adverse selection and moral hazard issues in M&As (Cain et al., 2011), and earnout contracts perform better in M&A deals with severer information asymmetry (Datar, Frankel, & Wolfson, 2001; Ragozzino & Reuer, 2009). PC contract is a kind of debt-like contract which is similar to earnout. However, PC contracts diﬀer from earnouts in term of payment and default cost. With respect to payment, in earnout, the acquirer will sign up a contract with the target which divides the payment into two parts, the ﬁrst part is an upfront payment while the second is the key contingent component based on the target’s future perfor- mance. But in PCs, acquirers are required to immediately pay for the acquisitions when acquisition contracts are signed; target shareholders need to pay back the cash or stocks if their future performances fail to meet the expectation. With respect to default cost, earnout contract links the default cost to the valuation of target, while performance commitment contract links the default cost to the diﬀerence between realised proﬁt and promised proﬁt. The limited default cost of PC contracts gives the parities with private information a motivation to release opportunistic commitment. 3. Related literature and hypotheses In this section, we further clarify the theoretical framework of the PC regulation in M&As and provides two competing hypotheses: eﬃciency prediction and agency conﬂict prediction. On one hand, based on signalling theory, performance commit- ment regulation requires the acquirees to disclose their performance forecasts for target assets and make a compensation agreement in case they fail to meet the performance. One can expect that it may work as a signalling mechanism, reduce the information asymmetry and valuation risk. This is an eﬃciency prediction. On the other hand, performance commitment in M&As may further worsen the type II agency conﬂict. Making use of the theoretical and implementation loopholes of PC regulation, controlling shareholders of listed companies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the agency conﬂicts between them and minority shareholders. That is, agency conﬂict prediction of PC in M&As. 3.1. Eﬃciency prediction Rey and Salanié (1996) theoretically analysed the value of commitment in the contract under the circumstance of high asymmetric information. They ﬁnd commitment can reduce the adverse selection problems. Xu, Zhang, and Wu (2008) found the additional CHINA JOURNAL OF ACCOUNTING STUDIES 321 commitments in the share-split reform have signalling eﬀects. Using share-split reform setting, Hou, Jin, Yang, Yuan, and Zhang (2015) found performance commitments can help controlling shareholders to reduce the share compensation that they have to pay. Ideally, performance commitment contracts are expected to reduce adverse selection problems caused by information asymmetry. Moreover, the performance commitment contract may reduce post-merger re-negotiation cost. Meanwhile, the compensation agreement seems to align the interest of acquirees with the acquirer, and thus facilitate alleviating the moral hazard problem and motivating targets to meet their performance forecast. 3.2. Agency conﬂict prediction In mergers and acquisitions, information asymmetry leads to lots of problems. Moreover, agency conﬂicts are intensiﬁed by the concentrated ownership structure (Allen et al., 2005; Ang et al., 2000; Jiang et al., 2010). For the M&As in the Chinese capital market, the large shareholders of listed companies have information advantages, and the minority shareholders have limited access to information. Large shareholders have incentive and ability to expropriate from minority shareholders via related-party M&As (Ji, Wei, & Liu, 2010; Liu, Zhong, & Jin, 2007;Wei,Cai,&Cheng, 2016). For the M&As with performance commitment, the type II agency conﬂict may be further aggravated. PC contract may make the type II agency conﬂict in M&A transactions changed from the traditional conﬂict between large shareholder and minority share- holders to the aggravated conﬂict between ‘Large shareholder+ Commitment parties’ and minority shareholders. Under ideal conditions, performance promises in M&As could reduce information asymmetries by pre-disclosing the target assets performance forecast and clearly establishing a commitment. However, due to the theoretical ﬂaws and institutional environment, performance commitment may not serve as a signal mechanism; instead, the parties with private information may use it as a tool for concealing bad news and engaging in opportunistic behaviour. 3.3. Research hypothesis development Lv and Han (2014) found a positive cumulative abnormal return (CAR) around the announcement day of performance commitment in M&As and concluded that the per- formance commitment contract works as an eﬀective signal mechanism in M&A deals. However, their result could be explained in an alternative way. Due to the immature institutional environment and dominance of small individual investors in the Chinese stock market, the positive CARs eﬀect could be explained by investor irrationality. To address the unresolved research question that whether performance commitment works as a signalling mechanism in M&A deals, we study on the impact of performance commitments on future stock price crash risk of acquiring ﬁrms, rather than short-term eﬀect. The previous studies ﬁnd that the managers have the incentive to withhold bad news, and bad news hoarding can lead to stock price crashes (Chu & Fang, 2016; 322 J. LI ET AL. Hutton, Marcus, & Tehranian, 2009; Jin & Myers, 2006;Kim,Li, &Zhang, 2011a, 2011b; Xu, Jiang, Yi, & Xu, 2012;Xu,Yu,&Yi, 2013). Ideally, M&A PC regulation requires the acquirees to disclose target asset performance forecast for certain post-merger years and clearly make a commitment, and a compensation agreement to compensate acquirers’ loss in the case that the acquirees cannot fulﬁl the commitment. This regulation may reduce acquirees’ incentive to hide bad news related to target quality, and thus reduce the stock price crash risk of acquiring ﬁrms. However, based on the above analysis of the theoretical and implementation defects of the current PC regulation, acquirees may make use of performance commitment regula- tion to conceal the true value of the inferior target asset, and in turn increase the future stock price crash risk of acquiring ﬁrms. Above analyses lead to the following competing hypotheses: H1a: Performance commitments in M&As reduce the acquiring ﬁrms’ future stock price crash risk. H1b: Performance commitments in M&As increase the acquiring ﬁrms’ future stock price crash risk. Many Chinese ﬁrms have a high divergence between the control and cash ﬂow rights of controlling shareholders (Jiang et al., 2010). Previous literature suggests that control- ling shareholders tend to expropriate from minority shareholders through related-party M&As in China (Deng, Zeng, & He, 2011; Li, Yu, & Wang, 2005; Tang & Han, 2018). Li et al. (2005) ﬁnd a negative market reaction to related-party M&As. Ji and Ma (2016) used a case analysis to indicate that controlling shareholder tends to manipulate performance com- mitment contracts in related-party M&A transactions and minority shareholders suﬀer the losses eventually. Compared with non-related-party M&As, the performance commitments signed in related-party M&As are more likely to be manipulated by the parties with private information (e.g. controlling shareholders). When the negative news related to target performance accumulate to a certain extent and release to the market, stock price crashes. This leads to the following hypothesis: H2: Performance commitments in related-party M&As have a stronger impact on stock price crash risk of acquiring ﬁrms. Theprimary strand of voluntarydisclosureliteraturedocuments that managers disclose their private information because rational market participants would other- wise interpret non-disclosure as unfavourable news and consequently discount the value of the ﬁrm’s assets. Voluntary disclosure mitigates the adverse selection problem in capital markets by reducing information asymmetry between the ﬁrm and investors. Previous literature ﬁnds that voluntary disclosure could reduce ﬁrms’ cost of capital and increase their stock liquidity (Cheynel, 2013;Francis,Nanda,& Olsson, 2008). We are interested to investigate whether voluntarily signed perfor- mance commitments have a better performance than mandatory commitments. Based on prior analyses, voluntary commitment may not be a kind of high-quality CHINA JOURNAL OF ACCOUNTING STUDIES 323 voluntary disclosure, but an opportunistic tool used by the parties with information advantages. Given that, voluntarily commitment may not be a better signal of high- quality acquisition than mandatory commitment. Above analyses lead us to develop the following hypothesis. H3: The impact of voluntary performance commitments on acquiring ﬁrms’ future stock price crash risk is not diﬀerent from mandatory ones. 4. Sample and research design 4.1. Sample and data source We get the mergers and acquisitions (M&As) data from CSMAR database. The performance commitment data and ﬁrm ﬁnancial data are obtained from WIND database. We only keep the M&A events in which acquiring ﬁrms are listed non- ﬁnancial ﬁrms. We also drop the M&A transactions which are less than 5 million yuan. Our sample period starts from the year of 2008 and ends in 2017. Our ﬁnal sample includes 8137 ﬁrm-year observations. Table 1 shows the distribution of performance commitments in M&As by year and industry. Panel A of Table1 shows the distribution of performance commit- ments in M&As by year. The performance commitment regulation is issued in 2008. In our sample, there are rare performance commitments in the beginning 3 years of our sample, the number of performance commitment in M&As is growing since the year of 2011. Overall, 35.69% of M&A deals during our sample period contain performance commitments. Part B of Table 1 describes the distribution of performance commitments in M&As by industry. The top three industries with the most amount of M&A transactions are (1) Manufacturing industry, (2) Information transmission, software and information technology services industry, and (3) Wholesale and retail industry. The top three industries with the highest proportion of performance commitments in M&A deals are (1) Information transmission, software and information technology services indus- try, (2) Culture, sports and entertainment industry, and (3) Leasing and business services industry. The proportion of M&A transactions with performance commit- ments in these three industries was 56.02%, 50.67% and 48.57%, respectively. The above-mentioned statistics show that the presence of M&A performance commit- ments varies among industries, and thus we control the industry-ﬁxed eﬀects in the subsequent empirical regressions. 4.2. Measuring ﬁrm-speciﬁc crash risk Following prior work (e.g., Chen, Hong, & Stein, 2001; Kim et al., 2011a), we use two measures of crash risk. The ﬁrst measure is negative conditional return skewness (NCSKEW) measure. Speciﬁcally, we compute NCSKEW for each ﬁrm i in ﬁscal year t as follows: 324 J. LI ET AL. Table 1. Sample distribution. Year Total acquisitions Acquisitions with PCs Percentage (PCs) Panel A: the distribution of acquisition events by year 2008 130 0 0.00% 2009 128 1 0.78% 2010 106 0 0.00% 2011 157 10 6.40% 2012 234 16 6.84% 2013 323 99 30.65% 2014 465 238 51.18% 2015 641 339 52.89% 2016 558 236 42.30% 2017 462 204 44.16% Total 3203 1143 35.69% Panel B:the distribution of acquisitions events according to the acquiring ﬁrms’ industry A 38 8 21.05 B 85 18 21.17 C 1811 669 36.94 D 147 29 19.73 E 75 26 34.67 F 180 41 22.78 G 87 16 18.39 H 14 1 7.14 I 332 186 56.02 K 159 25 15.72 L 70 34 48.57 M 38 18 47.37 N 49 19 38.78 P5 2 40 Q 16 7 43.75 R 75 38 50.67 S 22 6 27.27 Total 3203 1143 35.69 We show the numbers and proportions of acquisitions with PCs in Table 1. Panel A represents the annual distribution of acquisitions announced by Chinese listed companies between 2008 and 2017. It can be found that during the sample period, acquisitions with PC contracts are increasing year by year. Panel B shows that the distribution of acquisition announced by Chinese listed companies between 2008 and 2017 according to the acquiring ﬁrms’ industry. We follow the guidance on the industry category of listed companies issued by CSRC, where A = agriculture, B = mining, C = manufacturing, D = electricity, gas, and water, E = building and construction, F = wholesale and retail business, G = transportation, warehousing and post, H = accommodation and catering, I = information transportation, software and information technology service, K = real estate, L = leasing and business services, M = scientiﬁc research, N = water conservancy and public service, P = education, Q = health and community service, R = culture, sports and entertain- ment, S = the others. We ﬁrst account for the general market eﬀect on crash risk by estimating ﬁrm-speciﬁc weekly returns, denoted as W, as the natural logarithm of one plus the residual return from the expanded market model regression for each ﬁrm and year: r ¼ α þ β r þ β r þ β r þ β r þ β r þ ε (1) i;t m;t2 m;t1 m;t m;tþ1 m;tþ2 i;t 1;i 2;i 3;i 4;i 5;i where r is the return on stock i in week t and r is the value-weighted A-share market i;t m;t return in week t. The ﬁrm-speciﬁc weekly returns for ﬁrm i in week t are represented by W ¼ ln 1 þ ε , where ε is the residual in Equation (1). We use the negative coeﬃcient i;t i;t i;t of skewness, NCSKEW, to measure crash risk: CHINA JOURNAL OF ACCOUNTING STUDIES 325 hi 3 P nðn 1Þ w i;t NCSKEW ¼ (2) i;t ðÞ n 1ðÞ n 2 w i;t where n is the number of observations of ﬁrm-speciﬁc weekly returns of ﬁrm i - during year t. Our second measure of crash risk is the down-to-up volatility measure (DUVOL) of the crash likelihood, which captures asymmetric volatilities between negative and positive ﬁrm-speciﬁc weekly returns. For each ﬁrm i in ﬁscal year t, DUVOL is computed as the natural logarithm of the ratio of the standard deviation of down weeks to the standard deviation of up weeks: hi 8 9 < = ðÞ n 1 w down i;t hi DUVOL ¼ ln (3) i;t : ; ðÞ n 1 w up i:t where n andn are, respectively, the number of up weeks and the number of down weeks. u d 4.3. Research design To test our main hypothesis, we estimate the following two regressions that link our measures of crash risk of acquiring ﬁrms in year t to performance commitment in year t-1 and to a set of control variables in year t-1: NCSKEW ¼ α þ β COMMIT þ ControlVariables þ ε (4) i;t i; t1 t1 i; t DUVOL ¼ α þ β COMMIT þ ControlVariables þ ε (5) i; t i; t1 t1 i; t where NCSKEW is the negative skewness of ﬁrm-speciﬁc weekly returns calculated i,t basedonEquation(2); DUVOL is a down-to-up volatility measure calculated based i,t on Equation (3).COMMIT is adummy variable with avalue of 1ifthe merged ﬁrm i,t-1 is in the post-performance commitment period, and 0 otherwise. Both Equations (5) and (6) are estimated using ordinary least squares (OLS) regressions. The set of control variables includes DTURN , NCSKEW (DUVOL ), SIGMA , i,t-1 i,t-1 i,t-1 i,t-1 RET , SIZE ,MB , LEV , ROA , and ACCM , which are taken from Chen i,t-1 i,t-1 i,t-1 i,t-1 i,t-1 i,t-1 et al. (2001) and Hutton et al. (2009). The variable DTURN is the detrended average i, t-1 monthly stock turnover in year t-1. The variable SIGMA is the standard deviation of i, t-1 ﬁrm-speciﬁc weekly returns over the ﬁscal year period t-1. The variable RET is the i, t-1 arithmetic average of ﬁrm-speciﬁc weekly returns in the ﬁscal year period t-1. The SIZE is deﬁned as the log of total assets in year t-1. The variable MB is the market i, t-1 i, t-1 value of equity divided by the book value of equity in year t-1. The variable LEV is i, t-1 the total liability divided by total assets. The variable ROA is deﬁned as net income i, t-1 divided by lagged total assets. The variable ACCM is the measure of accrual i, t-1 manipulation, which is measured by the absolute discretionary accruals. We use the modiﬁed Jones model to estimate the discretionary accruals. 326 J. LI ET AL. To test H2, we split our sample into two subsamples. One is related-party M&A transactions group. The other one is non-related-party M&As group. We test the impact of performance commitment on acquiring ﬁrms’ stock price crash risk in each subsample. To test H3, we construct two variables. V_COMMIT is deﬁned as 1 if the performance commitment is voluntarily made by deal parties, and 0 otherwise. M_COMMIT is deﬁned as 1 if the performance commitment is mandatory, and 0 otherwise. We drop the COMMIT variable from Equations (4) and (5) and augment Equations (4) and (5) with the variables V_COMMIT and M_COMMIT as follows: NCSKEW ¼ α þ β V COMMIT þ β M COMMIT þ ControlVariables þ ε (6) i; t i; t1 i; t1 t1 i; t 1 2 DUVOL ¼ α þ β V COMMIT þ β M COMMIT þ ControlVariables þ ε (7) i; t i; t1 i; t1 t1 i; t 1 2 5. Empirical results 5.1. Descriptive statistics Table 2 presents descriptive statistics for all the variables used in the regression analyses, based on the sample of ﬁrm years with non-missing control variables. As seen in Table 2, the mean value of NCSKEW is −0.284, the mean value of DUVOL is −0.191. This result is similar to that of Chu et al. (2016). The mean value of COMMIT is 0.24, suggesting 24% of the ﬁrm-years are in the post-commitment period. Table 3 presents the correlations for all the variables used in the regression analyses. Table 3 shows that the two crash risk measures (i.e. NCSKEW and DUVOL) are highly correlated with a ratio of 0.871. More importantly, both measures of future crash risk are positively correlated with COMMIT, which is consistent with our predictions that performance commitments in M&As increase the acquiring ﬁrms’ future crash risk. Table 2. Descriptive statistics. Variables N Mean Std Min Max NCSKEW 8137 −0.284 0.653 −2.194 1.496 DUVOL 8137 −0.192 0.471 −1.319 1.050 COMMIT 8137 0.240 0.426 0 1 t-1 SIZE 8137 22.364 1.237 18.932 25.672 t-1 LEV 8137 0.491 0.213 0.049 1.360 t-1 MB 8137 4.159 3.986 −1.963 27.718 t-1 ROA 8137 0.033 0.056 −0.309 0.205 t-1 DTURN 8137 −0.033 0.389 −1.718 0.943 t-1 RET 8137 0.340 0.767 −0.713 3.331 t-1 SIGMA 8137 0.056 0.024 0.018 0.192 t-1 ACCM 8137 0.058 0.060 0.001 0.329 t-1 NCSKEW 8137 −0.315 0.636 −2.1942 1.495 t-1 DUVOL 8137 −0.221 0.462 −1.319 1.050 t-1 RELATED 8137 0.683 0.465 0 1 MAJOR 8137 0.315 0.464 0 1 This table reports summary statistics of the main variables used in this study. We provide the detailed deﬁnitions of all the variables in Appendix 1. CHINA JOURNAL OF ACCOUNTING STUDIES 327 Table 3. Correlations. AB C D E F G H I G K NCSKEW A1 DUVOL B 0.871*** 1 COMMIT C 0.088*** 0.079*** 1 t-1 SIZE D −0.122*** −0.114*** −0.115*** 1 t-1 LEV E −0.059*** −0.050*** −0.197*** 0.436*** 1 t-1 MB F 0.111*** 0.098*** 0.154*** −0.452*** −0.033*** 1 t-1 ROA G 0.002 −0.005 0.060*** 0.029** −0.356*** 0.007 1 t-1 DTURN H −0.082*** −0.092*** 0.004 0.058*** 0.053*** 0.077*** −0.074*** 1 t-1 RET I 0.045*** 0.022* 0.187*** −0.148*** −0.081*** 0.409*** 0.057*** 0.410*** 1 t-1 SIGMA G 0.036*** 0.009 0.291*** −0.180*** −0.080*** 0.419*** −0.044*** 0.362*** 0.607*** 1 t-1 ACCM K 0.022* 0.013 0.003 −0.060*** 0.108*** 0.104*** −0.093*** −0.005 0.020* 0.069*** 1 t-1 This table presents the correlations of the main variables used in this study. The superscripts ***, **, and *indicate statistical signiﬁcance at the 1%, 5%, and 10% levels, respectively. Deﬁnitions of the variables are given in Appendix 1. 328 J. LI ET AL. Table 4. Univariate analysis. (1) (2) COMMIT =1 COMMIT = 0 The diﬀerence t-1 t-1 Variables (N = 1949) (N = 6188) (1)-(2) NCSKEW −0.199 −0.311 0.111*** DUVOL −0.132 −0.211 0.078*** SIZE 22.033 22.298 −0.265*** t-1 LEV 0.415 0.515 −0.100*** t-1 MB 5.229 3.822 1.407*** t-1 ROA 0.038 0.032 0.006*** t-1 DTURN −0.043 −0.030 −0.013 t-1 RET 0.421 0.315 0.106*** t-1 SIGMA 0.067 0.053 0.014*** t-1 ACCM 0.057 0.058 −0.002 t-1 RELATED 1.029 0.980 0.048** t-1 MAJOR 0.856 0.209 0.647*** t-1 NCSKEW −0.271 −0.329 0.059*** t-1 DUVOL −0.195 −0.230 0.035*** t-1 This table provides the univariate analysis. The superscripts ***, **, and * indicate statistical signiﬁcance at the 1%, 5%, and 10% levels, respectively. Deﬁnitions of the variables are given in Appendix 1. Table 5. The impact of PCs on acquiring ﬁrms’ stock price crash risk (H1). (1) (2) NCSKEW DUVOL t t COMMIT 0.0411** 0.0269* t-1 (2.1129) (1.8827) SIZE −0.0389*** −0.0321*** t-1 (−4.7432) (−5.6766) LEV −0.00650 0.0276 t-1 (−0.1402) (0.8489) MB 0.0063** 0.0034* t-1 (2.3811) (1.8866) ROA 0.0439 −0.00170 t-1 (0.3109) (−0.0157) DTURN −0.0352 −0.0239 t-1 (−1.3443) (−1.2631) RET 0.0817*** 0.0603*** t-1 (4.9663) (5.0167) SIGMA 0.534 −0.131 t-1 (1.2972) (−0.4372) ACCM 0.0533 −0.0224 t-1 (0.4296) (−0.2472) NCSKEW 0.0661*** t-1 (5.5629) DUVOL 0.0677*** t-1 (5.9395) Intercept 0.5047*** 0.5114*** (2.8006) (4.1057) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0570 0.0580 This table presents the results of the impact of PCs on acquiring ﬁrms’ stock price crash risk. The sample contains ﬁrm-year observations from 2008 to 2017 with non-missing values for all the control variables. The t-values are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate statistical signiﬁcance at 10%, 5%, and 1% levels, respectively. Variables are deﬁned in Appendix 1. CHINA JOURNAL OF ACCOUNTING STUDIES 329 Table 6. Related-party M&As, PC contracts and acquiring ﬁrms’ stock price crash risk (H2). Related-party M&As Non-related-party M&As (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t COMMIT 0.0461* 0.0312* 0.0251 0.0153 t-1 (1.9560) (1.7863) (0.7199) (0.6104) SIZE −0.0394*** −0.0329*** −0.0388** −0.0303*** t-1 (−3.9947) (−4.8131) (−2.3765) (−2.7762) LEV 0.0295 0.0389 −0.0886 0.00130 t-1 (0.5226) (0.9506) (−1.0150) (0.0236) MB 0.0084*** 0.0051** 0.000200 −0.00170 t-1 (2.7086) (2.3689) (0.0374) (−0.5745) ROA 0.0702 −0.0126 0.0180 0.0655 t-1 (0.4043) (−0.0940) (0.0750) (0.3369) DTURN −0.0363 −0.0310 −0.0334 −0.0133 t-1 (−1.0193) (−1.2144) (−0.9141) (−0.4880) RET 0.0604*** 0.0493*** 0.1215*** 0.0779*** t-1 (3.0820) (3.4548) (4.0203) (3.5982) SIGMA 0.704 −0.0813 0.236 −0.134 t-1 (1.4515) (−0.2301) (0.3039) (−0.2393) ACCM 0.138 0.0811 −0.161 −0.255 t-1 (0.9157) (0.7458) (−0.7272) (−1.4971) NCSKEW 0.0565*** 0.0830*** t-1 (3.9303) (3.9181) DUVOL 0.0453*** 0.1006*** t-1 (3.0635) (4.8084) Interpret 0.4867** 0.444 0.5117*** 0.343 (2.2299) (1.2361) (3.3763) (1.4459) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 5564 5564 2573 2573 Adjusted R 0.0610 0.0620 0.0480 0.0550 This table presents the impact of related-party M&As on the relation between PC contracts and acquiring ﬁrms’ stock price crash risk. The sample contains ﬁrm-year observations from 2008 to 2017 with non-missing values for all the control variables. We split the sample into related-party M&As group and non-related-party M&As group. The t-values are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate statistical signiﬁcance at 10%, 5%, and 1% levels, respectively. Variables are deﬁned in Appendix 1. Table 4 provides the univariable analyses. The results are generally consistent with our previous prediction. As seen in Table 4, the mean values of NCSKEW and DUVOL for the two groups in which COMMIT equals 0 or 1 are signiﬁcantly diﬀerent at 1% level. 5.2. Multivariate test of H1 Table 5 presents the multivariate regression analyses for testing H1. As shown, acquiring ﬁrms future crash risk in year t measured by both NCSKEW and DUVOL are positively related to COMMIT in year t-1. These ﬁndings support hypothesis H1b, indicating that performance commitment increases the acquiring ﬁrms’ future crash risk. 5.3. Test of H2 If the positive relation between performance commitment in M&As and acquiring ﬁrms’ future crash risk is due to performance commitment facilitating opportunistic 330 J. LI ET AL. Table 7. Voluntarily signed PCs and acquiring ﬁrms’ stock price crash risk (H3). (1) (2) NCSKEW DUVOL t t V_Commit 0.0866*** 0.0638*** t-1 (4.0522) (4.1257) M_Commit 0.0884*** 0.0755*** t-1 (3.2655) (3.7642) SIZE −0.0306*** −0.0218*** t-1 (−3.9846) (−3.9863) LEV −0.0429 −0.0183 t-1 (−0.9336) (−0.5509) MB 0.0095*** 0.0066*** t-1 (3.7546) (3.7711) ROA −0.0289 −0.0926 t-1 (−0.2043) (−0.8341) DTURN −0.1529*** −0.1093*** t-1 (−6.6893) (−6.6528) RET 0.0917*** 0.0632*** t-1 (7.7174) (7.1459) SIGMA −0.7922** −1.0018*** t-1 (−2.1792) (−3.8333) ACCM 0.101 −0.00270 t-1 (0.8116) (−0.0292) NCSKEW 0.0483*** t-1 (4.0799) DUVOL 0.0496*** t-1 (4.3202) Interpret 0.3486** 0.2867** (2.0718) (2.3866) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0310 0.0280 Coef. of V_Commit = Coef. M_Commit. F-value 0.00 0.27 This table reports the impact of voluntarily signed PCs in M&A transac- tions and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and ***indicate signiﬁcance at10%, 5%, and 1% levels, respectively. behaviours of information insiders, such ashiding the true valueoftargetassets and provide opportunistic high-performance forecasts, one can expect that the strength of the relation to be stronger for ﬁrms involved in related-party M&As, as hypothesised in H2. We report the result for related-party M&As in column 1 and 2 of Table 6.As reported in column 1and 2of Table 6, when future crash risk is measured by NCSKEW and DUVOL,respectively, thecoeﬃcients of COMMIT are highly signiﬁcant with an expected sign. However, as shown in column 3 and 4 of Table 6,the coeﬃcients of COMMIT are not signiﬁcant in the subsample of non-related party M&As. These results suggest that the positive impact of performance CHINA JOURNAL OF ACCOUNTING STUDIES 331 Table 8. PC contracts, target quality uncertainty and acquiring ﬁrms’ stock price crash risk. (1) (2) NCSKEW DUVOL t t COMMIT 0.0480** 0.0335** t-1 (2.1340) (2.0222) UNCERTAINTY −0.000400 −0.000200 t-1 (−0.3729) (−0.3855) COMMIT* UNCERTAINTY −0.0005 −0.000800 t-1 (−0.3104) (−0.6594) SIZE −0.0396*** −0.0312*** t-1 (−4.3605) (−4.8866) LEV −0.0245 −0.00360 t-1 (−0.4801) (−0.0992) MB 0.0051* 0.0035* t-1 (1.7306) (1.7117) ROA −0.0276 −0.0546 t-1 (−0.1748) (−0.4439) DTURN −0.0546* −0.0359* t-1 (−1.8993) (−1.7145) RET 0.0922*** 0.0655*** t-1 (4.9844) (4.8188) SIGMA 0.657 0.00270 t-1 (1.4582) (0.0081) ACCM 0.0263 −0.0635 t-1 (0.1847) (−0.6107) NCSKEW 0.0621*** t-1 (4.8127) DUVOL 0.0643*** t-1 (5.1048) Interpret 0.3872* 0.3517** (1.9528) (2.5016) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 6655 6655 Adjusted R 0.0590 0.0620 This table reports the impact of target quality uncertainty on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. commitment on acquiring ﬁrms’ future crash risk is dominated by related-party M&As transactions. The above results are consistent with H2. 5.4. Test of H3 Table 7 presents the impact of voluntary performance commitment on acquiring ﬁrms’ future crash risk. We ﬁnd the coeﬃcient of V_COMMIT is not signiﬁcantly diﬀerent from the coeﬃcient of M_COMMIT, no matter measuring stock price crash risk by NCSKEW or DUVOL. This result supports our previous prediction. In short, the results presented in prior tables, taken together, are consistent with the agency theory explanation for performance commitment regulation in M&As. 332 J. LI ET AL. Table 9. PC contracts, assessment institution reputation and acquiring ﬁrms’ stock price crash risk.at10%, (1) (2) NCSKEW DUVOL t t COMMIT 0.0773*** 0.0605*** t-1 (4.0093) (4.3390) TOP_AI 0.0204 −0.00750 t-1 (0.5211) (−0.2653) COMMIT*TOP_AI 0.0175 0.0323 t-1 (0.3253) (0.8277) SIZE −0.0322*** −0.0229*** t-1 (−4.4552) (−4.3786) LEV −0.0344 −0.0145 t-1 (−0.8378) (−0.4896) MB 0.0093*** 0.0065*** t-1 (4.1888) (4.0554) ROA −0.0274 −0.0931 t-1 (−0.1944) (−0.9126) DTURN −0.1501*** −0.1070*** t-1 (−6.8215) (−6.7231) RET 0.0905*** 0.0622*** t-1 (7.3482) (6.9792) SIGMA −0.8020** −1.0089*** t-1 (−2.1478) (−3.7345) ACCM 0.0817 −0.0194 t-1 (0.6799) (−0.2228) NCSKEW 0.0489*** t-1 (4.3153) DUVOL 0.0502*** t-1 (4.4473) Interpret 0.4092*** 0.3289*** (2.6113) (2.8991) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0300 0.0270 This table reports the impact of assessment institution (AI) reputation on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. 6. Additional tests and robustness checks 6.1. Does the target quality uncertainty matter? We measure the target quality uncertainty in two ways. First, we measure the target quality uncertainty by the diﬀerence between assessment valuation and book valua- tion of target assets. Secondly, we measure the target quality uncertainty by the quality of ﬁnancial intermediaries in M&As (i.e. ﬁnancial advisor and assessment institution). We augment Equations (4) and (5) with target quality uncertainty variables and their interactions with performance commitment variable. Tables 8–10 present results for these analyses. The results show that the coeﬃcients of interaction terms are CHINA JOURNAL OF ACCOUNTING STUDIES 333 Table 10. PC contracts, ﬁnancial advisor reputation and acquiring ﬁrms’ stock price crash risk. (1) (2) NCSKEW DUVOL t t COMMIT 0.0780*** 0.0617*** t-1 (4.2102) (4.6043) TOP_FA −0.0057 −0.0130 t-1 (−0.0737) (−0.2334) COMMIT*TOP_FA 0.0568 0.0478 t-1 (0.6320) (0.7339) SIZE −0.0322*** −0.0230*** t-1 (−4.4581) (−4.3994) LEV −0.0346 −0.0143 t-1 (−0.8426) (−0.4812) MB 0.0093*** 0.0065*** t-1 (4.1921) (4.0452) ROA −0.0291 −0.0931 t-1 (−0.2065) (−0.9130) DTURN −0.1496*** −0.1069*** t-1 (−6.7999) (−6.7132) RET 0.0903*** 0.0622*** t-1 (7.3346) (6.9734) SIGMA −0.8063** −1.0100*** t-1 (−2.1600) (−3.7402) ACCM 0.0804 −0.0202 t-1 (0.6689) (−0.2323) NCSKEW 0.0492*** t-1 (4.3448) DUVOL 0.0502*** t-1 (4.4546) Interpret 0.4108*** 0.3308*** (2.6229) (2.9181) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0300 0.0270 This table reports the impact of ﬁnancial advisor (FA) reputation on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. insigniﬁcant, suggesting performance commitment could increase the acquiring ﬁrms’ future crash risk regardless of the target quality uncertainty and quality of external monitoring. These results further support our conjecture that PC regulation per se is problematic. 6.2. Do institutional investors matter? Institutional investors are more sophistical investors who have more ability to access to private information. The institutional investors may be capable to ‘see through’ the problematic performance commitment regulation. We augment Equations (4) and (5) with institutional ownership variable and its interaction with performance commitment 334 J. LI ET AL. Table 11. PC contracts, institutional ownership and acquiring ﬁrms’ stock price crash risk. (1) (2) NCSKEW DUVOL t t COMMIT 0.1113*** 0.0999*** t-1 (3.5397) (4.4552) INST 0.0155*** 0.0097*** t-1 (7.6286) (6.5258) COMMIT* INST −0.0078* −0.0078** t-1 (−1.7414) (−2.4682) SIZE −0.0332*** −0.0204*** t-1 (−4.0042) (−3.5152) LEV −0.0792* −0.0483 t-1 (−1.6473) (−1.4100) MB 0.0097*** 0.0072*** t-1 (3.5542) (3.8090) ROA −0.0716 −0.0832 t-1 (−0.4407) (−0.6683) DTURN −0.1522*** −0.1065*** t-1 (−6.0979) (−5.9748) RET 0.0830*** 0.0562*** t-1 (6.5044) (5.9416) SIGMA −0.8026** −0.9928*** t-1 (−2.0564) (−3.5363) ACCM 0.0550 −0.0270 t-1 (0.4502) (−0.2970) NCSKEW 0.0443*** t-1 (3.5632) DUVOL 0.0422*** t-1 (3.5210) Interpret 0.3806** 0.2341* (2.1340) (1.8603) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 7153 7153 Adjusted R 0.0410 0.0340 This table reports the impact of institutional investors on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix. t-Statistics are reported in par- entheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. variable. Table 11 presents the results. Consistent with our prediction, the coeﬃcient of interaction term is signiﬁcantly negative, suggesting that the institutional investors could mitigate the impact of performance commitment on crash risk. 6.3. Performance commitment, information asymmetry and stock price crash risk We further test the impact of information asymmetry on the relation between perfor- mance commitment in M&As and acquiring ﬁrms’ future crash risk. We proxy the informa- tion asymmetry in two ways: ownership of controlling shareholder and the divergence between the control and cash ﬂow rights of controlling shareholder. The results are presented in Tables 12 and 13. The results show that the positive relation between CHINA JOURNAL OF ACCOUNTING STUDIES 335 Table 12. PC contracts, the divergence between the control and cash ﬂow rights of controlling shareholder and acquiring ﬁrms’ stock price crash risk in related-party M&A deals High divergence Low divergence (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t COMMIT 0.0774** 0.0594** 0.0174 0.0207 t-1 (2.0501) (2.3692) (0.5861) (0.9589) SIZE −0.0329** −0.0226** −0.0444*** −0.0364*** t-1 (−2.2362) (−2.3207) (−3.9182) (−4.4288) LEV 0.0244 −0.0203 0.0568 0.0783 t-1 (0.2922) (−0.3528) (0.8405) (1.5943) MB 0.00410 0.00360 0.0097*** 0.0060** t-1 (0.8823) (1.0982) (2.8853) (2.4441) ROA −0.130 −0.0872 0.259 0.0507 t-1 (−0.4659) (−0.4415) (1.2296) (0.3307) DTURN −0.0279 −0.1057*** −0.0490 −0.0416 t-1 (−0.5196) (−3.3129) (−1.2406) (−1.4512) RET 0.0447 0.0572*** 0.0749*** 0.0626*** t-1 (1.3731) (3.4226) (2.9982) (3.4530) SIGMA 0.670 −1.0034** 0.622 −0.223 t-1 (0.8265) (−1.9849) (0.9533) (−0.4712) ACCM 0.102 0.153 0.194 0.0542 t-1 (0.4324) (0.8948) (1.0401) (0.4011) NCSKEW 0.0513** 0.0588*** t-1 (2.3127) (3.3705) DUVOL 0.0227 0.0621*** t-1 (1.0478) (3.5761) Interpret 0.317 0.257 0.5936** 0.5770*** (0.9692) (1.1988) (2.3464) (3.1402) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 2220 2220 3344 3344 Adjusted R 0.0390 0.0180 0.0720 0.0720 This table reports the impact of divergence between the control and cash ﬂow rights of controlling shareholder on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. performance commitment and acquiring ﬁrms’ future is dominated by the subsample of M&A transactions with higher information asymmetry. 6.4. Propensity Score Matching (PSM) method We also use the Propensity Score Matching (PSM) method to eliminate the endogeneity problem. Table 4 shows that there is a signiﬁcant diﬀerence in the mean of the control variables for the two groups categorised by the presence or absence of performance commitment in M&As. It could be argued that the positive eﬀect of performance commit- ment on future crash risk may be due to the diﬀerence in corporate fundamental factors. To eliminate this endogeneity problem, we also use the closest PSM method to re-test my main hypotheses. Table 14 reports the results using the PSM method. The results are quite similar to the main test reported above. So, our main results are unlikely to be driven by ﬁrm characteristics. 336 J. LI ET AL. Table 13. PC contracts, the ownership of controlling shareholder and acquiring ﬁrms’ stock price crash risk in related-party M&A deals. High ownership of controlling shareholder Low ownership of controlling shareholder (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t COMMIT 0.1068*** 0.0942*** 0.0277 0.00780 t-1 (3.2760) (4.0594) (0.8745) (0.3348) SIZE −0.0207* −0.0199** −0.0440*** −0.0372*** t-1 (−1.6680) (−2.2467) (−3.3032) (−3.8123) LEV 0.0629 0.0758 0.000400 −0.00150 t-1 (0.7827) (1.3227) (0.0054) (−0.0297) MB 0.0152*** 0.0071** 0.00500 0.0042* t-1 (3.2527) (2.1240) (1.4711) (1.6766) ROA 0.370 0.3130* −0.167 −0.255 t-1 (1.4227) (1.6887) (−0.7596) (−1.5825) DTURN −0.1214*** −0.0753** −0.0771* −0.0857*** t-1 (−2.9459) (−2.5670) (−1.8031) (−2.7356) RET 0.0868*** 0.0654*** 0.0677** 0.0512*** t-1 (3.9316) (4.1559) (2.5064) (2.5844) SIGMA −1.4041** −1.4586*** 0.991 0.0338 t-1 (−2.0968) (−3.0584) (1.4495) (0.0675) ACCM 0.182 −0.00910 0.111 0.164 t-1 (0.8570) (−0.0604) (0.5413) (1.0930) NCSKEW 0.0276 0.0585*** t-1 (1.4294) (3.0280) DUVOL 0.0225 0.0594*** t-1 (1.1721) (3.1164) Interpret 0.0310 0.186 0.5009* 0.4857** (0.1128) (0.9529) (1.7256) (2.2803) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 2713 2713 2851 2851 Adjusted R 0.0270 0.0260 0.0550 0.0610 This table reports the impact of the ownership of controlling shareholder on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. 6.5. Firm ﬁxed eﬀects regressions Since the empirical literature on forecasting crash risk is relatively new, it is possible that our analysis omits from the regressions some crash determinants that correlated with other included variables. To mitigate potential problems that can arise form correlated omitted variables, we follow Petersen (2009), we re-estimate regressions in Table 5 after controlling for ﬁrm-ﬁxed eﬀects and using the standard errors corrected for year cluster- ing. Table 15 presents the results of this exercise. As shown in Table 15, the relation between performance commitment in M&As and future crash risk remains highly sig- niﬁcant with an expected positive sign regardless of crash measures, suggesting that our results reported in Table 5 are unlikely to be driven by omitted correlated time-invariant variables. CHINA JOURNAL OF ACCOUNTING STUDIES 337 Table 14. The results of Propensity Score Matching (PSM) regressions.. First stage: COMMIT Second-stage: Crash risk Column(1): NCSKEW Column(2):DUVOL t t Variables Coeﬃcient Z-stat Coeﬃcient t-stat Coeﬃcient t-stat COMMIT 0.0734*** 3.2198 0.0415*** 2.6949 t-1 SIZE −0.1481*** −3.14 −0.0473*** −4.7896 −0.0382*** −5.8434 t-1 LEV −0.9556 −3.82 0.0163 0.2913 0.0294 0.8002 t-1 MB −0.0120 −1.00 0.0042 1.3160 0.00220 1.0979 t-1 ROA −0.5682 −0.70 0.0658 0.3925 −0.0522 −0.4395 t-1 DTURN −0.1649 −1.39 −0.0245 −0.7229 −0.0202 −0.8903 t-1 RET 0.2124*** 3.04 0.1147*** 5.5597 0.0863*** 5.7393 t-1 SIGMA 0.6489 0.33 −0.115 −0.2193 −0.503 −1.3077 t-1 ACCM 0.1241 0.17 0.124 0.8648 0.0297 0.2911 t-1 NCSKEW 0.0646*** 4.6866 t-1 DUVOL 0.0715*** 5.5018 t-1 MAJOR 1.8124*** 18.51 t-1 RELATED 0.1562 1.53 t-1 Interpret 0.6752 0.66 0.6699*** 3.0621 0.5857*** 4.0827 Year No Yes Yes Industry Yes Yes Yes N 8137 6372 6372 Pseudo R2 0.1451 Adj. R 0.0600 0.0620 This table reports the results of propensity score matching (PSM) regressions. The ﬁrst stage of the procedure involves a logit analysis. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in second-stage regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. 6.6. The impact of fulﬁlment rate of performance commitment We further test whether the fulﬁlment rate of performance commitment inﬂuent future crash risk of acquiring ﬁrms. We keep the ﬁrm years with both performance forecast and realised performance and construct two variables (i.e. FULFILL1 and FULFILL2) to measure the fulﬁlment ratio of performance commitment. Table 16 presents the results. As the results shown, the commitment fulﬁlment rate does not have a signiﬁcant impact on acquiring ﬁrms’ future crash risk. These results further support our conclusion that PC regulation per se is problematic. 6.7. Controlling more acquisition characteristics In this section, we further control some important acquisition characteristics which may inﬂuent the acquiring ﬁrms’ future crash risk. Directional private placement is common- used in corporate ﬁnancing for acquisitions. Previous literature ﬁnds the acquisitions with directional private placement ﬁnancing have severer information asymmetries. We further control for the ﬁnancing way of acquirer ﬁrms. We construct two variables. DPPt-1 is a dummy variable which equals 1 if the acquirers obtain the ﬁnancing by directional private placement, and 0 otherwise. CDPP t-1 is a dummy variable which equals 1 if the acquirers obtain the ﬁnancing by directional private placement and controlling shareholder is involved in the directional private placement, and 0 otherwise. As shown in Table 17,the main resultsremain after controlling these two variables. 338 J. LI ET AL. Table 15. Firm ﬁxed eﬀects regressions. (1) (2) NCSKEW DUVOL t t COMMIT 0.0796** 0.0543** t-1 (3.1072) (2.6811) SIZE 0.0008 0.0132 t-1 (0.0145) (0.3327) LEV −0.0483 −0.0513 t-1 (−0.6385) (−1.0631) MB 0.0086** 0.0068* t-1 (2.3569) (2.2613) ROA −0.2445 −0.2013 t-1 (−1.6958) (−1.7275) DTURN −0.1788*** −0.1366*** t-1 (−3.7770) (−3.7816) RET 0.0965*** 0.0692** t-1 (3.7303) (3.1671) SIGMA −0.460 −0.7299 t-1 (−0.4050) (−0.8987) ACCM 0.1525* 0.0373 t-1 (1.9300) (0.4318) NCSKEW −0.1516** t-1 (−2.9174) DUVOL −0.1515** t-1 (−3.0612) Constant −0.3934 −0.5176 (−0.3429) (−0.5964) Firm FE Yes Yes Cluster by year Yes Yes N 8137 8137 Adj. R 0.0900 0.0900 This table reports the results of ﬁrm ﬁxed eﬀects regressions. We re- estimate regressions in Table 5 after controlling for ﬁrm-ﬁxed eﬀects and using the standard errors corrected for year clustering. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by year. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. 6.8. Multi-period future stock price crash risk In our OLS regressions thus far, we examine the predictive ability of our performance commitment variable with respect to future crash risk. In measuring crash risk, we consider future crash occurrences in the 1-year-ahead forecast window. In this section, we re-examine the impact of performance commitment on future crash risk by using t3- and 4-year-ahead crash risk. As seen in Table 18,the relation between performance commitment in M&As and future three- and 4-year-ahead crash risk remains positive. 7. Conclusion We analyse the impact of performance commitment in M&As on future stock price crash risk of acquiring ﬁrms. The PC regulation in M&As is an innovative policy made by China authorities. PC regulation is expected to work as a mechanism to alleviate information asymmetry, thereby reduce adverse selection and overvaluation risk in M&A. However, CHINA JOURNAL OF ACCOUNTING STUDIES 339 Table 16. The fulﬁlment rate of PCs and acquiring ﬁrms’ stock price crash risk. (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t FULFILL1 0.0256 0.00900 t-1 (0.4728) (0.2253) FULFILL2 −0.00250 −0.00100 t-1 (−0.1657) (−0.1095) SIZE −0.0323 −0.00530 −0.0356 −0.0107 t-1 (−1.0707) (−0.2424) (−1.1939) (−0.4888) LEV −0.00410 −0.0398 0.0260 −0.00790 t-1 (−0.0280) (−0.3627) (0.1786) (−0.0720) MB 0.000700 0.00290 −0.000100 0.00210 t-1 (0.1456) (0.7279) (−0.0134) (0.5220) ROA −0.146 −0.263 −0.105 −0.202 t-1 (−0.2661) (−0.5387) (−0.1887) (−0.4086) DTURN −0.0725 −0.0563 −0.0524 −0.0421 t-1 (−1.2685) (−1.2296) (−0.9405) (−0.9392) RET 0.1326*** 0.0993*** 0.1280*** 0.0979*** t-1 (3.7394) (3.6815) (3.6011) (3.6232) SIGMA 0.0979 −0.967 0.0477 −1.058 t-1 (0.1113) (−1.4912) (0.0544) (−1.6437) ACCM −0.6571** −0.5367** −0.6746** −0.5312** t-1 (−1.9839) (−2.1436) (−2.0315) (−2.1126) NCSKEW 0.0158 0.0208 t-1 (0.5307) (0.6922) DUVOL 0.00170 0.00790 t-1 (0.0578) (0.2532) Interpret 0.0150 −0.370 0.109 −0.241 (0.0244) (−0.8373) (0.1801) (−0.5475) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes N 1000 1000 1000 1000 Adj. R 0.0330 0.0290 0.0310 0.0270 The fulﬁlment rate of PCs and acquiring ﬁrms’ stock price crash risk. This table estimates the impact of fulﬁlment ratio of performance commitment in M&A transactions and acquiring ﬁrms’ future stock price crash risk. All other variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors corrected for clustering by ﬁrm. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. due to the theoretical and institutional defects related to current performance commit- ment regulation, this regulation tends to provide the parties with private information an opportunity to overstate the values of inferior target assets. We ﬁnd that the performance commitment increases the acquiring ﬁrms’ future stock price crash risk. Furthermore, we ﬁnd the relation is more pronounced in related-party M&As. These ﬁndings are consistent with the hypothesis that controlling shareholders expropriate from minority shareholders of acquirers. We provide two main explanations for this ﬁnding. The ﬁrst explanation is based on signalling theory. The performance commitment in M&A is a costless commitment rather than a costly signal. The default cost is not high enough to constrain the promiserstomakeanopportunistic commitment. Secondly, with the respect to institutional environment, in M&A transactions, major shareholders of listed compa- nies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the agency conﬂicts between them and minority share- holders. Therefore, performance commitment does not work as a mechanism to 340 J. LI ET AL. Table 17. The results after controlling more charac- teristics of acquisitions. (1) (2) NCSKEWt DUVOLt COMMIT 0.0578** 0.0367** t-1 (2.4428) (2.0986) DPP −0.0312 −0.0179 t-1 (−1.1710) (−0.9324) CDPP 0.0127 0.00100 t-1 (0.4318) (0.0474) SIZE −0.0387*** −0.0320*** t-1 (−4.7234) (−5.6398) LEV −0.00760 0.0270 t-1 (−0.1651) (0.8291) MB 0.0062** 0.0033* t-1 (2.3331) (1.8416) ROA 0.0497 0.00190 t-1 (0.3513) (0.0170) DTURN −0.0361 −0.0243 t-1 (−1.3681) (−1.2738) RET 0.0818*** 0.0604*** t-1 (4.9732) (5.0268) SIGMA 0.591 −0.0954 t-1 (1.4202) (−0.3140) ACCM 0.0549 −0.0209 t-1 (0.4423) (−0.2302) NCSKEW 0.0664*** t-1 (5.5775) DUVOL 0.0682*** t-1 (5.9679) Interpret 0.4994*** 0.5064*** (2.7666) (4.0513) Year Yes Yes Industry Yes Yes N 8137 8137 Adj. R 0.0570 0.0580 This table reports the results after controlling more character- istics of acquisitions. DPP is a dummy variable which equals t-1 1 if the acquirers obtain the ﬁnancing by directional private is a dummy variable placement, and 0 for otherwise. CDPP t-1 which equals 1 if the acquirers obtain the ﬁnancing by direc- tional private placement and controlling shareholder is involved in directional private placement, and 0 for otherwise. All other variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors corrected for clustering by ﬁrm. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. reduce information asymmetry, but drive the parties with private information to hide the true value of the target assets; the stock price will suddenly drop when the accumulated hidden bad news release to the market. We argue that the current performance commitment regulation may not be suitable for the China market in which type II agency conﬂict is dominated. Our ﬁndings suggest that policy design and implementation need to ﬁt with China institutional environment. Our results shed some light on how to revise the PC regulation in M&As. CHINA JOURNAL OF ACCOUNTING STUDIES 341 Table 18. Multi-period future stock price crash risk. (1) (2) (3) (4) NCSKEW NCSKEW DUVOL DUVOL t+3 t+4 t+3 t+4 COMMIT 0.0683 0.1623* 0.0723** 0.0811 t-1 (1.4192) (1.8404) (2.1161) (1.2964) SIZE −0.0704*** −0.0732*** −0.0517*** −0.0569*** t-1 (−6.7869) (−6.3081) (−7.0138) (−6.9209) LEV 0.1891*** 0.1177* 0.1313*** 0.0926* t-1 (3.0527) (1.6689) (2.9860) (1.8528) MB 0.00280 0.00450 0.00130 0.00110 t-1 (0.7477) (1.0824) (0.4952) (0.3540) ROA 0.7397*** 0.4387* 0.4134*** 0.203 t-1 (3.5828) (1.9467) (2.8221) (1.2734) DTURN −0.000900 −0.0132 0.0147 −0.0242 t-1 (−0.0219) (−0.2834) (0.4953) (−0.7327) RET −0.0353 −0.0295 0.00680 −0.0155 t-1 (−1.4220) (−1.0499) (0.3846) (−0.7760) SIGMA 0.598 1.6508* −0.100 0.702 t-1 (0.7387) (1.7266) (−0.1752) (1.0395) ACCM −0.0502 0.0255 −0.0619 −0.0567 t-1 (−0.2851) (0.1286) (−0.4953) (−0.4028) NCSKEW 0.0459*** 0.0474** t-1 (2.5926) (2.2433) DUVOL 0.0390** 0.0377* t-1 (2.2730) (1.8868) Interpret 0.9799*** 1.0584*** 0.8214*** 0.9288*** (4.2149) (4.0495) (4.9799) (5.0166) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes N 3865 2937 3865 2937 Adj. R 0.0600 0.0730 0.0620 0.0730 This table tests the impact of PC contracts and acquiring ﬁrms’ multi-period future stock price crash risk. All other variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors corrected for clustering by ﬁrm. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. Disclosure statement No potential conﬂict of interest was reported by the authors. Funding This work was supported by the National Natural Science Foundation of China [71702196, 71772181, 71702036]; Fundamental Research Funds for the Central Universities [17wkpy16]. References Allen, F., Qian, J., & Qian, M. (2005). Law, ﬁnance, and economic growth in China. Journal of Financial Economics, 77(1), 57–116. Ang, J.S., Cole, R.A., & Lin, J.W. (2000). Agency costs and ownership structure. The Journal of Finance, 55(1), 81–106. Cadman, B., Carrizosa, R., & Faurel, L. (2014). Economic determinants and information environment eﬀects of earnouts: New insights from SFAS 141(r). Journal of Accounting Research, 52(1), 37–74. Cain, M.D., Denis, D.J., & Denis, D.K. (2011). Earnouts: A study of ﬁnancial contracting in acquisition agreements. Journal of Accounting and Economics, 51(1–2), 151–170. Chen, J., Hong, H., & Stein, J.C. (2001). Forecasting crashes: Trading volume, past returns, and conditional skewness in stock prices. Journal of Financial Economics, 61(3), 345–381. 342 J. LI ET AL. Cheynel, E. (2013). A theory of voluntary disclosure and cost of capital. Review of Accounting Studies, 18(4), 987–1020. Chu, J., & Fang, J. (2016). Margin-trading, short-selling and the deterioration of crash risk. Economic Research Journal, 5, 143–158 (in Chinese). Datar, S., Frankel, R., & Wolfson, M. (2001). Earnouts: The eﬀects of adverse selection and agency costs on acquisition techniques. Journal of Law Economics and Organization, 17(1), 201–238. Deng, J., Zeng, Y., & He, J. (2011). The root and the eﬀects of relative merger and acquisition. Chinese Journal of Management, 8, 1238–1246 (in Chinese). Francis, J, Nanda, D, & Olsson, P. (2008). Voluntary disclosure, earnings quality, and cost of capital. Journal Of Accounting Research, 46(1), 53–99. doi:10.1111/j.1475-679X.2008.00267.x Hou, Q., Jin, Q., Yang, R., Yuan, H., & Zhang, G. (2015). Performance commitments of controlling shareholders and earnings management. Contemporary Accounting Research, 32(3), 1099–1127. Hutton, A.P., Marcus, A.J., & Tehranian, H. (2009). Opaque ﬁnancial reports, R2, and crash risk. Journal of Financial Economics, 94(1), 67–86. Ji, H., & Ma, L. (2016). Regulatory soft constraints, false compensation promises and investor protection. Communication of Finance and Accounting, 15, 73–78 (in Chinese). Ji, H., Wei, M., & Liu, J. (2010). Asset injection, securities market regulation and performance. Accounting Research,2,47–56 (in Chinese). Jiang, G., Lee, C.M.C., & Yue, H. (2010). Tunneling through intercorporate loans: The China experience. Journal of Financial Economics, 98(1), 1–20. Jin, L., & Myers, S.C. (2006). R2 around the world: New theory and new tests. Journal of Financial Economics, 79(2), 257–292. Kim, J.-B., Li, Y., & Zhang, L. (2011a). CFOs versus CEOs: Equity incentives and crashes. Journal of Financial Economics, 101(3), 713–730. Kim, J.-B., Li, Y., & Zhang, L. (2011b). Corporate tax avoidance and stock price crash risk: Firm level analysis. Journal of Financial Economics, 100(3), 639–662. Li, Z., Yu, Q., & Wang, X. (2005). Tunneling, propping and M& A: Evidence from Chinese listed companies. Economic Research Journal, 2005(1), 95–105. (in Chinese). Liu, F., Zhong, R., & Jin, T. (2007). The transfer and “looting” of listed companies’ mastery under loose legal control. Management World, 12, 106–116 (in Chinese). Lv, C., & Han, H. (2014). VAM, synergy and distribution of gains from M&A. Audit & Economy Research, 6, 3–13 (in Chinese). Pan, A., Qiu, J., & Yang, Y. (2017). Research on the incentive eﬀect of valuation adjustment mechan- ism in M&As: Evidence from listed companies on SEM and GEM board in China. Accounting Research,3,46–52 (in Chinese). Petersen, M.A. (2009). Estimating standard errors in ﬁnance panel data sets: comparing approaches. The Review Of Financial Studies, 22(1), 435–480. doi:10.1093/rfs/hhn053 Ragozzino, R., & Reuer, J.J. (2009). Contingent earnouts in acquisitions of privately-held targets. Journal of Management, 35(4), 857–879. Rey, P., & Salanié, B. (1996). On the value of commitment with asymmetric information. Econometrica, 64(6), 1395–1414. Spence, M. (1973). Job market signaling. Quarterly Journal of Economics, 87(3), 355–374. Tang, Q., & Han, H. (2018). Related party M&As and ﬁrm value: The governance eﬀect of accounting conservatism. Nankai Business Review,3, 25–36 (in Chinese). Wei, M., Cai, G., & Cheng, M. (2016). A comprehensive analyzing framework of ownership character- istics: Based on the Chinese phenomena and theory. Accounting Research,5,26–33 (in Chinese). Xie, J., & Zhang, Q. (2016). Accounting treatments for proﬁt compensation commitment in listed companies’ holding mergers: Case studies of ﬁve China listed companies. Accounting Research,6, 15–20 (in Chinese). Xu, N., Jiang, X., Yi, Z., & Xu, X. (2012). Conﬂicts of interest, analyst optimism and stock price crash risk. Economic Research Journal, 7, 127–140 (in Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 343 Xu, N., Yu, S., & Yi, Z. (2013). Institutional investor herding and stock price crash risk. Management World,7, 31–43 (in Chinese). Xu, N., Zhang, H., & Wu, S. (2008). Do additional commitments have signaling eﬀect? Evidence from the non-tradable share reform in China. Management World, 3, 142–151 (in Chinese). 344 J. LI ET AL. Appendix 1. Variable deﬁnitions Variables Deﬁnitions NCSKEW Negative skewness of ﬁrm-speciﬁc weekly returns over the ﬁscal year period. See Equation (2) for details. DUVOL The natural logarithm of the ratio of the standard deviation of down weeks to the standard deviation of up weeks. See Equation (3) for details. COMMIT A dummy variable with a value of 1 if the merged ﬁrm is in the post-performance commitment period, and 0 otherwise. DTURN The average monthly share turnover over the current ﬁscal year period minus the average monthly share turnover over the previous ﬁscal year period. RET The mean of ﬁrm-speciﬁc weekly returns over the ﬁscal year period, times 100. SIGMA The standard deviation of ﬁrm-speciﬁc weekly returns over the ﬁscal year period. ROA The net income divided by lagged total assets SIZE The log of total assets LEV The total liability divided by total assets MB The market value of equity divided by the book value of equity ACCM The absolute value of discretionary accruals, where discretionary accruals are estimated from modiﬁed Jones model. RELATED A dummy variable equals 1 if the acquisition is a related-party transaction, and 0 for otherwise. MAJOR A dummy variable equals 1 if the acquisition is a signiﬁcant asset restructuring, and 0 for otherwise. V_COMMIT A dummy variable equals 1 if the PC contract is made voluntarily, and 0 for otherwise. M_COMMIT A dummy variable equals 1 if the PC contract is made mandatorily, and 0 for otherwise. FULFILL1 A dummy variable equals 1 if the acquirees meet the performance forecast, and 0 for otherwise. FULFILL2 The realised performance of target divided by the performance forecast UNCERTAINTY The absolute value of the diﬀerence between assessment value of target assets and book value of target asset, divided by book value of target assets TOP_AI A dummy variable equals 1 if the assessment institution is 1 of top 10 institutions, and 0 for otherwise TOP_FA A dummy variable equals 1 if the ﬁnancial advisor is 1 of top 10 institutions, and 0 for otherwise. INST The total ownership of institutional investors. DPP A dummy variable equals 1 if the acquiers obtain ﬁnancing through directional private placement, and 0 for otherwise. CDPP A dummy variable equals 1 if the controlling shareholder is involved in directional private placement, and 0 for otherwise. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png China Journal of Accounting Studies Taylor & Francis http://www.deepdyve.com/lp/taylor-francis/performance-commitment-in-m-as-and-stock-price-crash-risk-26AEznTbp4

Loading next page...

References (35)

Chenchuramaiah Bathala (2006)
R2 Around the World: New Theory and New Tests
Cfa Digest, 36
Qingchuan Hou, Qinglu Jin, Rong Yang, Hongqi Yuan, Guochang Zhang (2015)
Performance Commitments of Controlling Shareholders and Earnings Management
Contemporary Accounting Research, 32
Matthew Cain, David Denis, Diane Denis (2010)
Earnouts: A Study of Financial Contracting in Acquisition Agreements
Corporate Finance: Governance
(1996)
On the value of commitment with asymmetric information
, 64
Guohua Jiang, Charles Lee, Heng Yue (2010)
Tunneling through intercorporate loans: The China experience☆
Journal of Financial Economics, 98
(2018)
Related party M&As and firm value: The governance effect of accounting conservatism
Franklin Allen, Jun Qian, Meijun Qian (2004)
Law, Finance, and Economic Growth in China
Law & Economics
Joseph Chen, Harrison Hong, Jeremy Stein (1999)
Forecasting Crashes: Trading Volume, Past Returns and Conditional Skewness in Stock Prices
Capital Markets: Market Efficiency
(2008)
Estimating Standard Errors in Finance Panel Data Sets: Comparing Approaches
H. Jia (2011)
The Root and the Effects of Relative Merger and Acquisition
Chinese Journal of Management
Amy Hutton, A. Marcus, H. Tehranian (2009)
Opaque financial reports, R2, and crash risk
Journal of Financial Economics, 94
(2016)
A comprehensive analyzing framework of ownership characteristics: Based on the Chinese phenomena and theory
(2007)
The transfer and “looting” of listed companies’mastery under loose legal control
Brian Cadman, Richard Carrizosa, Lucile Faurel (2013)
Economic Determinants and Information Environment Effects of Earnouts: New Insights from SFAS 141(R)
Financial Accounting eJournal
(2012)
Conflicts of interest, analyst optimism and stock price crash risk
Jeong‐Bon Kim, Yinghua Li, Liandong Zhang (2011)
CFOs versus CEOs: Equity Incentives and Crashes
SPGMI: Compustat Fundamentals (Topic)
J. Francis, D. Nanda, Per Olsson (2008)
Voluntary Disclosure, Earnings Quality, and Cost of Capital
Journal of Accounting Research, 46
Jeong‐Bon Kim, Yinghua Li, Liandong Zhang (2010)
Corporate Tax Avoidance and Stock Price Crash Risk: Firm-Level Analysis
Microeconomics: General Equilibrium & Disequilibrium Models of Financial Markets eJournal
S. Datar, Richard Frankel, Mark Wolfson (2001)
Earnouts: The Effects of Adverse Selection and Agency Costs on Acquisition Techniques
Journal of Law Economics & Organization, 17
(2017)
Research on the incentive effect of valuation adjustment mechanism in M&As: Evidence from listed companies on SEM and GEM board in China
(2016)
Margin-trading, short-selling and the deterioration of crash risk
(2013)
Institutional investor herding and stock price crash risk
(2016)
Regulatory soft constraints, false compensation promises and investor protection
James Ang, Rebel Cole, James Lin (2000)
Agency Costs and Ownership Structure
Journal of Finance, 55
(2005)
Tunneling, propping and M& A: Evidence from Chinese listed companies
E. Cheynel (2013)
A theory of voluntary disclosure and cost of capital
Review of Accounting Studies, 18
(2010)
Asset injection, securities market regulation and performance
R. Ragozzino, J. Reuer (2009)
Contingent Earnouts in Acquisitions of Privately Held Targets
Journal of Management, 35
(2016)
Accounting treatments for profit compensation commitment in listed companies’ holding mergers: Case studies of five China listed companies
P. Rey, B. Salanié (1996)
Long-Term, Short-term and Renegotiation: On the Value of Commitment with Asymmetric Information
M. Spence (1973)
Job Market Signaling
Quarterly Journal of Economics, 87
(2016)
Regulatory soft constraints, false compensation promises and investor protection. Communication of Finance and Accounting
Luo Changjian (2014)
VAM,Synergy and Distribution of Gains from M&A
Journal of Audit & Economics
(2006)
R2 around the world: New theory and new tests
(2008)
Do additional commitments have signaling effect? Evidence from the non-tradable share reform in China

Publisher: Taylor & Francis
Copyright: © 2019 Accounting Society of China
ISSN: 2169-7221
eISSN: 2169-7213
DOI: 10.1080/21697213.2019.1695948
Publisher site: See Article on Publisher Site

Abstract

CHINA JOURNAL OF ACCOUNTING STUDIES 2019, VOL. 7, NO. 3, 317–344 https://doi.org/10.1080/21697213.2019.1695948 ARTICLE a b c Jingjing LI , Yingwen GUO and Minghai WEI a b School of Economics and Management, Harbin Institute of Technology, Shenzhen, China; School of Business, Nanjing University, Nanjing, China; Center for Accounting, Finance and Institutions, Sun Yat-sen University, Guangzhou, China ABSTRACT KEYWORDS Performance commitment This paper investigates the impact of performance commitment in M&A; stock price crash in M&A transactions on acquiring ﬁrms’ future crash risk. We ﬁnd risk; agency conﬂict; a positive relation between performance commitment and acquir- signalling theory; costless ing ﬁrms’ future crash risk, and this result is more pronounced in signal related-party M&A deals. These ﬁndings are consistent with our prediction that performance commitment regulation may bring about negative consequences by providing the parties with private information an opportunity to overstate the values of inferior target assets; the stock price will suddenly drop when the accumulated hidden bad news release to the market. We further systematically discuss the shortcomings in the theories and regulation related to performance commitment in M&As. We document that the current performance commitment contract is designed as a costless pro- mise with low default cost, resulting in the information insiders make use of the regulation loopholes to aggressively expropriate from less-informed minority shareholders. 1. Introduction Performance commitment regulation (PC hereafter) in mergers and acquisitions (M&As) is an innovative policy made by China authorities. The original intention of the regulation is to constrain the opportunistic behaviours of those with information advantages and improve information transparency. Ideally, performance commitment is expected to alleviate information asymmetry, thereby reducing moral hazard and adverse selection problems in M&As (Cadman, Carrizosa, & Faurel, 2014; Cain, Denis, &Denis, 2011;Lv&Han, 2014; Ragozzino & Reuer, 2009). Moreover, the PC contract is expected to increase the re-negotiation eﬃciency and reduce contract frictions caused by information incompleteness or uncertainty. However, the current performance commitment regulation in M&As has been criticised by academicians, media and investors. They claim that PC regulation provides acquirees with an opportunity to overstate the values of inferior target assets. The less-sophisticated small investors in Chinese capital market cannot distinguish ‘opportunistic’ performance CONTACT Minghai WEI mnswmh@mail.sysu.edu.cn Center for Accounting, Finance and Institutions, Sun Yat-sen University, Guangzhou, China Paper accepted by Kangtao Ye. This article has been republished with minor changes. These changes do not impact the academic content of the article. © 2019 Accounting Society of China 318 J. LI ET AL. commitments from reliable performance commitments, which may result in a stock price crash in the future when the accumulated bad news is ﬁnally released. Most prior studies on performance commitment contracts focus on the short-term eﬀects of performance commitment. Lv and Han (2014) found a positive cumulative abnormal return (CAR) around the announcement day of performance commitment in M&As and concluded that the PC contract works as an eﬀective signalling mechanism in M&A deals. Pan, Qiu, and Yang (2017) investigated the eﬀects of performance commitments on the ﬁnancial performances of target ﬁrms in the ﬁrst year after acquisition. Given most of the existing studies examine the short- term eﬀect of performance commitment, it is diﬃcult to distinguish the investor irrationality from the signalling eﬀect of performance commitment. In this paper,weexaminetheimpactof performance commitment on future stock price crash risk of acquiring ﬁrms, using the sample from 2008 to 2017. We ﬁnd that the performance commitment increases the acquiring ﬁrms’ future crash risk. We explore the theoretical and institutional reasons why the performance commitment regulation causes negative consequences in the Chinese capital mar- ket. The ﬁrst explanation is based on signalling theory (Spence, 1973). Signalling theory implies that the eﬀective signal should be costly. However, the default cost of performance commitment in M&As is nothighenoughto constrain thepromi- sers’ opportunistic behaviours. In other words, the performance commitment in M&As may become a costless commitment or cheap talk. Secondly, with respect to an institutional environment, in M&A transactions, major shareholders of listed companies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the agency conﬂicts between them and minority shareholders. Therefore, performance commitment does not work as a signalling mechanism to reduce information asymmetry, but drive the parties with private information to hide the true value of the target assets; the stock price will suddenly drop when the accumulated hidden bad news release to the market. We contribute to the literature in three aspects. First, to our best knowledge, this is the ﬁrst paper to systematically discuss the shortcomings in the theories and regulation related to performance commitment in M&As. Secondly, there is a growing research examining the determinants of ﬁrms’ stock price crash risk. This paper extends previous studies by providing the evidence that PC in M&As could increase the acquiring ﬁrms’ future crash risk. Third, policymakers and regulators may be interested in our ﬁndings. We argue that the current performance commitment regulation may not be suitable for the China market in which type II agency conﬂict is dominated. Our ﬁndings suggest that policy design and implementation need to ﬁt with China institutional environment. Our results shed some light on how to revise the PC regulation in M&As. 2. Institutional backgrounds In 2008, China Securities Regulatory Commission (CSRC) issued ‘The Administration Measures for Signiﬁcant Asset Restructuring of Listed Companies’ (hereafter The CHINA JOURNAL OF ACCOUNTING STUDIES 319 Administration Measures) to alleviate the information asymmetry and valuation risk in M&As. For the M&A transactions are valuated based on discounted future- expected cash ﬂows, CSRC explicitly stipulated that the acquirees should sign performance commitment contracts with acquirers so that acquirers can reliably evaluate the values of targets based on targets’ future-expected performances. In addition, the target shareholders need to sign a compensation agreement with the acquirer. The target shareholders need to pay back the cash or stocks if the realised performance of the target misses their promised performance goal. During the pre-acquisition period, to get a high premium in acquisitions, target shareholders have strong incentives to camouﬂage their real situations, leading to valuation risk. The authorities expect the performance commitment regulation to work as a mechanism to reduce valuation risk and protect minority shareholders from low-quality M&A deals. The ‘Q&A of Listed Companies’ Mergers and Acquisitions’ (p.128) compiled by the Shenzhen Stock Exchange clearly states that ‘. . . high valuations need to be based on a matched high-performance commitment . . .’. Target shareholders need to make a tradeoﬀ between the compensation risk and valuation gain. However, the current performance commitment regulation in M&A has considerate theoretical ‘loopholes’, which may drive the target shareholders to camouﬂage their real situations at the expense of triggering the compensation clause. We analyse the ‘loopholes’ of PC regulation in four aspects. First, in terms of legal nature, performance commitment contract in M&As is a kind of anonymous contract without name prescribed in Contract Law. As a nameless contract, it does not have compulsory enforcement force, and thus it cannot provide a strong protection to the acquirer shareholders. Second, based on signalling theory, we argue that performance commitment contract in M&As is a costless commitment, rather than costly signal. According to current regulation, when the acquiree fails to realise the performance goal, they are only required to compensate for the diﬀerence between the actual realised earnings and the promised earnings, rather the diﬀerence in valuations. The default cost is too low to constrain acquirees’ incentive to overstate their performance forecast. Therefore, the design of performance commitment regulation ignores costly signal- ling theory which indicates that eﬀective signals must be costly (Spence, 1973). The theoretical loopholes of current performance commitment regulation may result in opportunistic behaviours of the parties with information advantages. Third, with respect to agency conﬂict, agency conﬂicts are intensiﬁed by the concentrated ownership structure in China market (Allen, Qian, & Qian, 2005;Ang, Cole, & Lin, 2000; Jiang,Lee,&Yue, 2010). The major shareholders of listed compa- nies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the aggravated agency conﬂicts between them and min- ority shareholders. Fourth, the existing regulatory policies are insuﬃcient, thereby giving the parties with private information a space for opportunistic behaviour. There is a limited restriction on the behaviours of acquiring ﬁrms’ original (controlling) shareholders 320 J. LI ET AL. and targets’ original shareholders in the post-commitment period. Therefore, these information insiders are capable to obtain excess beneﬁtbymakinguse of those regulatory defects. Earnout is one of the debt-like claims proposed by the investors to address adverse selection issues in western countries, such as the UK and US. Previous literature suggests that the earnout contract can reduce adverse selection and moral hazard issues in M&As (Cain et al., 2011), and earnout contracts perform better in M&A deals with severer information asymmetry (Datar, Frankel, & Wolfson, 2001; Ragozzino & Reuer, 2009). PC contract is a kind of debt-like contract which is similar to earnout. However, PC contracts diﬀer from earnouts in term of payment and default cost. With respect to payment, in earnout, the acquirer will sign up a contract with the target which divides the payment into two parts, the ﬁrst part is an upfront payment while the second is the key contingent component based on the target’s future perfor- mance. But in PCs, acquirers are required to immediately pay for the acquisitions when acquisition contracts are signed; target shareholders need to pay back the cash or stocks if their future performances fail to meet the expectation. With respect to default cost, earnout contract links the default cost to the valuation of target, while performance commitment contract links the default cost to the diﬀerence between realised proﬁt and promised proﬁt. The limited default cost of PC contracts gives the parities with private information a motivation to release opportunistic commitment. 3. Related literature and hypotheses In this section, we further clarify the theoretical framework of the PC regulation in M&As and provides two competing hypotheses: eﬃciency prediction and agency conﬂict prediction. On one hand, based on signalling theory, performance commit- ment regulation requires the acquirees to disclose their performance forecasts for target assets and make a compensation agreement in case they fail to meet the performance. One can expect that it may work as a signalling mechanism, reduce the information asymmetry and valuation risk. This is an eﬃciency prediction. On the other hand, performance commitment in M&As may further worsen the type II agency conﬂict. Making use of the theoretical and implementation loopholes of PC regulation, controlling shareholders of listed companies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the agency conﬂicts between them and minority shareholders. That is, agency conﬂict prediction of PC in M&As. 3.1. Eﬃciency prediction Rey and Salanié (1996) theoretically analysed the value of commitment in the contract under the circumstance of high asymmetric information. They ﬁnd commitment can reduce the adverse selection problems. Xu, Zhang, and Wu (2008) found the additional CHINA JOURNAL OF ACCOUNTING STUDIES 321 commitments in the share-split reform have signalling eﬀects. Using share-split reform setting, Hou, Jin, Yang, Yuan, and Zhang (2015) found performance commitments can help controlling shareholders to reduce the share compensation that they have to pay. Ideally, performance commitment contracts are expected to reduce adverse selection problems caused by information asymmetry. Moreover, the performance commitment contract may reduce post-merger re-negotiation cost. Meanwhile, the compensation agreement seems to align the interest of acquirees with the acquirer, and thus facilitate alleviating the moral hazard problem and motivating targets to meet their performance forecast. 3.2. Agency conﬂict prediction In mergers and acquisitions, information asymmetry leads to lots of problems. Moreover, agency conﬂicts are intensiﬁed by the concentrated ownership structure (Allen et al., 2005; Ang et al., 2000; Jiang et al., 2010). For the M&As in the Chinese capital market, the large shareholders of listed companies have information advantages, and the minority shareholders have limited access to information. Large shareholders have incentive and ability to expropriate from minority shareholders via related-party M&As (Ji, Wei, & Liu, 2010; Liu, Zhong, & Jin, 2007;Wei,Cai,&Cheng, 2016). For the M&As with performance commitment, the type II agency conﬂict may be further aggravated. PC contract may make the type II agency conﬂict in M&A transactions changed from the traditional conﬂict between large shareholder and minority share- holders to the aggravated conﬂict between ‘Large shareholder+ Commitment parties’ and minority shareholders. Under ideal conditions, performance promises in M&As could reduce information asymmetries by pre-disclosing the target assets performance forecast and clearly establishing a commitment. However, due to the theoretical ﬂaws and institutional environment, performance commitment may not serve as a signal mechanism; instead, the parties with private information may use it as a tool for concealing bad news and engaging in opportunistic behaviour. 3.3. Research hypothesis development Lv and Han (2014) found a positive cumulative abnormal return (CAR) around the announcement day of performance commitment in M&As and concluded that the per- formance commitment contract works as an eﬀective signal mechanism in M&A deals. However, their result could be explained in an alternative way. Due to the immature institutional environment and dominance of small individual investors in the Chinese stock market, the positive CARs eﬀect could be explained by investor irrationality. To address the unresolved research question that whether performance commitment works as a signalling mechanism in M&A deals, we study on the impact of performance commitments on future stock price crash risk of acquiring ﬁrms, rather than short-term eﬀect. The previous studies ﬁnd that the managers have the incentive to withhold bad news, and bad news hoarding can lead to stock price crashes (Chu & Fang, 2016; 322 J. LI ET AL. Hutton, Marcus, & Tehranian, 2009; Jin & Myers, 2006;Kim,Li, &Zhang, 2011a, 2011b; Xu, Jiang, Yi, & Xu, 2012;Xu,Yu,&Yi, 2013). Ideally, M&A PC regulation requires the acquirees to disclose target asset performance forecast for certain post-merger years and clearly make a commitment, and a compensation agreement to compensate acquirers’ loss in the case that the acquirees cannot fulﬁl the commitment. This regulation may reduce acquirees’ incentive to hide bad news related to target quality, and thus reduce the stock price crash risk of acquiring ﬁrms. However, based on the above analysis of the theoretical and implementation defects of the current PC regulation, acquirees may make use of performance commitment regula- tion to conceal the true value of the inferior target asset, and in turn increase the future stock price crash risk of acquiring ﬁrms. Above analyses lead to the following competing hypotheses: H1a: Performance commitments in M&As reduce the acquiring ﬁrms’ future stock price crash risk. H1b: Performance commitments in M&As increase the acquiring ﬁrms’ future stock price crash risk. Many Chinese ﬁrms have a high divergence between the control and cash ﬂow rights of controlling shareholders (Jiang et al., 2010). Previous literature suggests that control- ling shareholders tend to expropriate from minority shareholders through related-party M&As in China (Deng, Zeng, & He, 2011; Li, Yu, & Wang, 2005; Tang & Han, 2018). Li et al. (2005) ﬁnd a negative market reaction to related-party M&As. Ji and Ma (2016) used a case analysis to indicate that controlling shareholder tends to manipulate performance com- mitment contracts in related-party M&A transactions and minority shareholders suﬀer the losses eventually. Compared with non-related-party M&As, the performance commitments signed in related-party M&As are more likely to be manipulated by the parties with private information (e.g. controlling shareholders). When the negative news related to target performance accumulate to a certain extent and release to the market, stock price crashes. This leads to the following hypothesis: H2: Performance commitments in related-party M&As have a stronger impact on stock price crash risk of acquiring ﬁrms. Theprimary strand of voluntarydisclosureliteraturedocuments that managers disclose their private information because rational market participants would other- wise interpret non-disclosure as unfavourable news and consequently discount the value of the ﬁrm’s assets. Voluntary disclosure mitigates the adverse selection problem in capital markets by reducing information asymmetry between the ﬁrm and investors. Previous literature ﬁnds that voluntary disclosure could reduce ﬁrms’ cost of capital and increase their stock liquidity (Cheynel, 2013;Francis,Nanda,& Olsson, 2008). We are interested to investigate whether voluntarily signed perfor- mance commitments have a better performance than mandatory commitments. Based on prior analyses, voluntary commitment may not be a kind of high-quality CHINA JOURNAL OF ACCOUNTING STUDIES 323 voluntary disclosure, but an opportunistic tool used by the parties with information advantages. Given that, voluntarily commitment may not be a better signal of high- quality acquisition than mandatory commitment. Above analyses lead us to develop the following hypothesis. H3: The impact of voluntary performance commitments on acquiring ﬁrms’ future stock price crash risk is not diﬀerent from mandatory ones. 4. Sample and research design 4.1. Sample and data source We get the mergers and acquisitions (M&As) data from CSMAR database. The performance commitment data and ﬁrm ﬁnancial data are obtained from WIND database. We only keep the M&A events in which acquiring ﬁrms are listed non- ﬁnancial ﬁrms. We also drop the M&A transactions which are less than 5 million yuan. Our sample period starts from the year of 2008 and ends in 2017. Our ﬁnal sample includes 8137 ﬁrm-year observations. Table 1 shows the distribution of performance commitments in M&As by year and industry. Panel A of Table1 shows the distribution of performance commit- ments in M&As by year. The performance commitment regulation is issued in 2008. In our sample, there are rare performance commitments in the beginning 3 years of our sample, the number of performance commitment in M&As is growing since the year of 2011. Overall, 35.69% of M&A deals during our sample period contain performance commitments. Part B of Table 1 describes the distribution of performance commitments in M&As by industry. The top three industries with the most amount of M&A transactions are (1) Manufacturing industry, (2) Information transmission, software and information technology services industry, and (3) Wholesale and retail industry. The top three industries with the highest proportion of performance commitments in M&A deals are (1) Information transmission, software and information technology services indus- try, (2) Culture, sports and entertainment industry, and (3) Leasing and business services industry. The proportion of M&A transactions with performance commit- ments in these three industries was 56.02%, 50.67% and 48.57%, respectively. The above-mentioned statistics show that the presence of M&A performance commit- ments varies among industries, and thus we control the industry-ﬁxed eﬀects in the subsequent empirical regressions. 4.2. Measuring ﬁrm-speciﬁc crash risk Following prior work (e.g., Chen, Hong, & Stein, 2001; Kim et al., 2011a), we use two measures of crash risk. The ﬁrst measure is negative conditional return skewness (NCSKEW) measure. Speciﬁcally, we compute NCSKEW for each ﬁrm i in ﬁscal year t as follows: 324 J. LI ET AL. Table 1. Sample distribution. Year Total acquisitions Acquisitions with PCs Percentage (PCs) Panel A: the distribution of acquisition events by year 2008 130 0 0.00% 2009 128 1 0.78% 2010 106 0 0.00% 2011 157 10 6.40% 2012 234 16 6.84% 2013 323 99 30.65% 2014 465 238 51.18% 2015 641 339 52.89% 2016 558 236 42.30% 2017 462 204 44.16% Total 3203 1143 35.69% Panel B:the distribution of acquisitions events according to the acquiring ﬁrms’ industry A 38 8 21.05 B 85 18 21.17 C 1811 669 36.94 D 147 29 19.73 E 75 26 34.67 F 180 41 22.78 G 87 16 18.39 H 14 1 7.14 I 332 186 56.02 K 159 25 15.72 L 70 34 48.57 M 38 18 47.37 N 49 19 38.78 P5 2 40 Q 16 7 43.75 R 75 38 50.67 S 22 6 27.27 Total 3203 1143 35.69 We show the numbers and proportions of acquisitions with PCs in Table 1. Panel A represents the annual distribution of acquisitions announced by Chinese listed companies between 2008 and 2017. It can be found that during the sample period, acquisitions with PC contracts are increasing year by year. Panel B shows that the distribution of acquisition announced by Chinese listed companies between 2008 and 2017 according to the acquiring ﬁrms’ industry. We follow the guidance on the industry category of listed companies issued by CSRC, where A = agriculture, B = mining, C = manufacturing, D = electricity, gas, and water, E = building and construction, F = wholesale and retail business, G = transportation, warehousing and post, H = accommodation and catering, I = information transportation, software and information technology service, K = real estate, L = leasing and business services, M = scientiﬁc research, N = water conservancy and public service, P = education, Q = health and community service, R = culture, sports and entertain- ment, S = the others. We ﬁrst account for the general market eﬀect on crash risk by estimating ﬁrm-speciﬁc weekly returns, denoted as W, as the natural logarithm of one plus the residual return from the expanded market model regression for each ﬁrm and year: r ¼ α þ β r þ β r þ β r þ β r þ β r þ ε (1) i;t m;t2 m;t1 m;t m;tþ1 m;tþ2 i;t 1;i 2;i 3;i 4;i 5;i where r is the return on stock i in week t and r is the value-weighted A-share market i;t m;t return in week t. The ﬁrm-speciﬁc weekly returns for ﬁrm i in week t are represented by W ¼ ln 1 þ ε , where ε is the residual in Equation (1). We use the negative coeﬃcient i;t i;t i;t of skewness, NCSKEW, to measure crash risk: CHINA JOURNAL OF ACCOUNTING STUDIES 325 hi 3 P nðn 1Þ w i;t NCSKEW ¼ (2) i;t ðÞ n 1ðÞ n 2 w i;t where n is the number of observations of ﬁrm-speciﬁc weekly returns of ﬁrm i - during year t. Our second measure of crash risk is the down-to-up volatility measure (DUVOL) of the crash likelihood, which captures asymmetric volatilities between negative and positive ﬁrm-speciﬁc weekly returns. For each ﬁrm i in ﬁscal year t, DUVOL is computed as the natural logarithm of the ratio of the standard deviation of down weeks to the standard deviation of up weeks: hi 8 9 < = ðÞ n 1 w down i;t hi DUVOL ¼ ln (3) i;t : ; ðÞ n 1 w up i:t where n andn are, respectively, the number of up weeks and the number of down weeks. u d 4.3. Research design To test our main hypothesis, we estimate the following two regressions that link our measures of crash risk of acquiring ﬁrms in year t to performance commitment in year t-1 and to a set of control variables in year t-1: NCSKEW ¼ α þ β COMMIT þ ControlVariables þ ε (4) i;t i; t1 t1 i; t DUVOL ¼ α þ β COMMIT þ ControlVariables þ ε (5) i; t i; t1 t1 i; t where NCSKEW is the negative skewness of ﬁrm-speciﬁc weekly returns calculated i,t basedonEquation(2); DUVOL is a down-to-up volatility measure calculated based i,t on Equation (3).COMMIT is adummy variable with avalue of 1ifthe merged ﬁrm i,t-1 is in the post-performance commitment period, and 0 otherwise. Both Equations (5) and (6) are estimated using ordinary least squares (OLS) regressions. The set of control variables includes DTURN , NCSKEW (DUVOL ), SIGMA , i,t-1 i,t-1 i,t-1 i,t-1 RET , SIZE ,MB , LEV , ROA , and ACCM , which are taken from Chen i,t-1 i,t-1 i,t-1 i,t-1 i,t-1 i,t-1 et al. (2001) and Hutton et al. (2009). The variable DTURN is the detrended average i, t-1 monthly stock turnover in year t-1. The variable SIGMA is the standard deviation of i, t-1 ﬁrm-speciﬁc weekly returns over the ﬁscal year period t-1. The variable RET is the i, t-1 arithmetic average of ﬁrm-speciﬁc weekly returns in the ﬁscal year period t-1. The SIZE is deﬁned as the log of total assets in year t-1. The variable MB is the market i, t-1 i, t-1 value of equity divided by the book value of equity in year t-1. The variable LEV is i, t-1 the total liability divided by total assets. The variable ROA is deﬁned as net income i, t-1 divided by lagged total assets. The variable ACCM is the measure of accrual i, t-1 manipulation, which is measured by the absolute discretionary accruals. We use the modiﬁed Jones model to estimate the discretionary accruals. 326 J. LI ET AL. To test H2, we split our sample into two subsamples. One is related-party M&A transactions group. The other one is non-related-party M&As group. We test the impact of performance commitment on acquiring ﬁrms’ stock price crash risk in each subsample. To test H3, we construct two variables. V_COMMIT is deﬁned as 1 if the performance commitment is voluntarily made by deal parties, and 0 otherwise. M_COMMIT is deﬁned as 1 if the performance commitment is mandatory, and 0 otherwise. We drop the COMMIT variable from Equations (4) and (5) and augment Equations (4) and (5) with the variables V_COMMIT and M_COMMIT as follows: NCSKEW ¼ α þ β V COMMIT þ β M COMMIT þ ControlVariables þ ε (6) i; t i; t1 i; t1 t1 i; t 1 2 DUVOL ¼ α þ β V COMMIT þ β M COMMIT þ ControlVariables þ ε (7) i; t i; t1 i; t1 t1 i; t 1 2 5. Empirical results 5.1. Descriptive statistics Table 2 presents descriptive statistics for all the variables used in the regression analyses, based on the sample of ﬁrm years with non-missing control variables. As seen in Table 2, the mean value of NCSKEW is −0.284, the mean value of DUVOL is −0.191. This result is similar to that of Chu et al. (2016). The mean value of COMMIT is 0.24, suggesting 24% of the ﬁrm-years are in the post-commitment period. Table 3 presents the correlations for all the variables used in the regression analyses. Table 3 shows that the two crash risk measures (i.e. NCSKEW and DUVOL) are highly correlated with a ratio of 0.871. More importantly, both measures of future crash risk are positively correlated with COMMIT, which is consistent with our predictions that performance commitments in M&As increase the acquiring ﬁrms’ future crash risk. Table 2. Descriptive statistics. Variables N Mean Std Min Max NCSKEW 8137 −0.284 0.653 −2.194 1.496 DUVOL 8137 −0.192 0.471 −1.319 1.050 COMMIT 8137 0.240 0.426 0 1 t-1 SIZE 8137 22.364 1.237 18.932 25.672 t-1 LEV 8137 0.491 0.213 0.049 1.360 t-1 MB 8137 4.159 3.986 −1.963 27.718 t-1 ROA 8137 0.033 0.056 −0.309 0.205 t-1 DTURN 8137 −0.033 0.389 −1.718 0.943 t-1 RET 8137 0.340 0.767 −0.713 3.331 t-1 SIGMA 8137 0.056 0.024 0.018 0.192 t-1 ACCM 8137 0.058 0.060 0.001 0.329 t-1 NCSKEW 8137 −0.315 0.636 −2.1942 1.495 t-1 DUVOL 8137 −0.221 0.462 −1.319 1.050 t-1 RELATED 8137 0.683 0.465 0 1 MAJOR 8137 0.315 0.464 0 1 This table reports summary statistics of the main variables used in this study. We provide the detailed deﬁnitions of all the variables in Appendix 1. CHINA JOURNAL OF ACCOUNTING STUDIES 327 Table 3. Correlations. AB C D E F G H I G K NCSKEW A1 DUVOL B 0.871*** 1 COMMIT C 0.088*** 0.079*** 1 t-1 SIZE D −0.122*** −0.114*** −0.115*** 1 t-1 LEV E −0.059*** −0.050*** −0.197*** 0.436*** 1 t-1 MB F 0.111*** 0.098*** 0.154*** −0.452*** −0.033*** 1 t-1 ROA G 0.002 −0.005 0.060*** 0.029** −0.356*** 0.007 1 t-1 DTURN H −0.082*** −0.092*** 0.004 0.058*** 0.053*** 0.077*** −0.074*** 1 t-1 RET I 0.045*** 0.022* 0.187*** −0.148*** −0.081*** 0.409*** 0.057*** 0.410*** 1 t-1 SIGMA G 0.036*** 0.009 0.291*** −0.180*** −0.080*** 0.419*** −0.044*** 0.362*** 0.607*** 1 t-1 ACCM K 0.022* 0.013 0.003 −0.060*** 0.108*** 0.104*** −0.093*** −0.005 0.020* 0.069*** 1 t-1 This table presents the correlations of the main variables used in this study. The superscripts ***, **, and *indicate statistical signiﬁcance at the 1%, 5%, and 10% levels, respectively. Deﬁnitions of the variables are given in Appendix 1. 328 J. LI ET AL. Table 4. Univariate analysis. (1) (2) COMMIT =1 COMMIT = 0 The diﬀerence t-1 t-1 Variables (N = 1949) (N = 6188) (1)-(2) NCSKEW −0.199 −0.311 0.111*** DUVOL −0.132 −0.211 0.078*** SIZE 22.033 22.298 −0.265*** t-1 LEV 0.415 0.515 −0.100*** t-1 MB 5.229 3.822 1.407*** t-1 ROA 0.038 0.032 0.006*** t-1 DTURN −0.043 −0.030 −0.013 t-1 RET 0.421 0.315 0.106*** t-1 SIGMA 0.067 0.053 0.014*** t-1 ACCM 0.057 0.058 −0.002 t-1 RELATED 1.029 0.980 0.048** t-1 MAJOR 0.856 0.209 0.647*** t-1 NCSKEW −0.271 −0.329 0.059*** t-1 DUVOL −0.195 −0.230 0.035*** t-1 This table provides the univariate analysis. The superscripts ***, **, and * indicate statistical signiﬁcance at the 1%, 5%, and 10% levels, respectively. Deﬁnitions of the variables are given in Appendix 1. Table 5. The impact of PCs on acquiring ﬁrms’ stock price crash risk (H1). (1) (2) NCSKEW DUVOL t t COMMIT 0.0411** 0.0269* t-1 (2.1129) (1.8827) SIZE −0.0389*** −0.0321*** t-1 (−4.7432) (−5.6766) LEV −0.00650 0.0276 t-1 (−0.1402) (0.8489) MB 0.0063** 0.0034* t-1 (2.3811) (1.8866) ROA 0.0439 −0.00170 t-1 (0.3109) (−0.0157) DTURN −0.0352 −0.0239 t-1 (−1.3443) (−1.2631) RET 0.0817*** 0.0603*** t-1 (4.9663) (5.0167) SIGMA 0.534 −0.131 t-1 (1.2972) (−0.4372) ACCM 0.0533 −0.0224 t-1 (0.4296) (−0.2472) NCSKEW 0.0661*** t-1 (5.5629) DUVOL 0.0677*** t-1 (5.9395) Intercept 0.5047*** 0.5114*** (2.8006) (4.1057) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0570 0.0580 This table presents the results of the impact of PCs on acquiring ﬁrms’ stock price crash risk. The sample contains ﬁrm-year observations from 2008 to 2017 with non-missing values for all the control variables. The t-values are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate statistical signiﬁcance at 10%, 5%, and 1% levels, respectively. Variables are deﬁned in Appendix 1. CHINA JOURNAL OF ACCOUNTING STUDIES 329 Table 6. Related-party M&As, PC contracts and acquiring ﬁrms’ stock price crash risk (H2). Related-party M&As Non-related-party M&As (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t COMMIT 0.0461* 0.0312* 0.0251 0.0153 t-1 (1.9560) (1.7863) (0.7199) (0.6104) SIZE −0.0394*** −0.0329*** −0.0388** −0.0303*** t-1 (−3.9947) (−4.8131) (−2.3765) (−2.7762) LEV 0.0295 0.0389 −0.0886 0.00130 t-1 (0.5226) (0.9506) (−1.0150) (0.0236) MB 0.0084*** 0.0051** 0.000200 −0.00170 t-1 (2.7086) (2.3689) (0.0374) (−0.5745) ROA 0.0702 −0.0126 0.0180 0.0655 t-1 (0.4043) (−0.0940) (0.0750) (0.3369) DTURN −0.0363 −0.0310 −0.0334 −0.0133 t-1 (−1.0193) (−1.2144) (−0.9141) (−0.4880) RET 0.0604*** 0.0493*** 0.1215*** 0.0779*** t-1 (3.0820) (3.4548) (4.0203) (3.5982) SIGMA 0.704 −0.0813 0.236 −0.134 t-1 (1.4515) (−0.2301) (0.3039) (−0.2393) ACCM 0.138 0.0811 −0.161 −0.255 t-1 (0.9157) (0.7458) (−0.7272) (−1.4971) NCSKEW 0.0565*** 0.0830*** t-1 (3.9303) (3.9181) DUVOL 0.0453*** 0.1006*** t-1 (3.0635) (4.8084) Interpret 0.4867** 0.444 0.5117*** 0.343 (2.2299) (1.2361) (3.3763) (1.4459) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 5564 5564 2573 2573 Adjusted R 0.0610 0.0620 0.0480 0.0550 This table presents the impact of related-party M&As on the relation between PC contracts and acquiring ﬁrms’ stock price crash risk. The sample contains ﬁrm-year observations from 2008 to 2017 with non-missing values for all the control variables. We split the sample into related-party M&As group and non-related-party M&As group. The t-values are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate statistical signiﬁcance at 10%, 5%, and 1% levels, respectively. Variables are deﬁned in Appendix 1. Table 4 provides the univariable analyses. The results are generally consistent with our previous prediction. As seen in Table 4, the mean values of NCSKEW and DUVOL for the two groups in which COMMIT equals 0 or 1 are signiﬁcantly diﬀerent at 1% level. 5.2. Multivariate test of H1 Table 5 presents the multivariate regression analyses for testing H1. As shown, acquiring ﬁrms future crash risk in year t measured by both NCSKEW and DUVOL are positively related to COMMIT in year t-1. These ﬁndings support hypothesis H1b, indicating that performance commitment increases the acquiring ﬁrms’ future crash risk. 5.3. Test of H2 If the positive relation between performance commitment in M&As and acquiring ﬁrms’ future crash risk is due to performance commitment facilitating opportunistic 330 J. LI ET AL. Table 7. Voluntarily signed PCs and acquiring ﬁrms’ stock price crash risk (H3). (1) (2) NCSKEW DUVOL t t V_Commit 0.0866*** 0.0638*** t-1 (4.0522) (4.1257) M_Commit 0.0884*** 0.0755*** t-1 (3.2655) (3.7642) SIZE −0.0306*** −0.0218*** t-1 (−3.9846) (−3.9863) LEV −0.0429 −0.0183 t-1 (−0.9336) (−0.5509) MB 0.0095*** 0.0066*** t-1 (3.7546) (3.7711) ROA −0.0289 −0.0926 t-1 (−0.2043) (−0.8341) DTURN −0.1529*** −0.1093*** t-1 (−6.6893) (−6.6528) RET 0.0917*** 0.0632*** t-1 (7.7174) (7.1459) SIGMA −0.7922** −1.0018*** t-1 (−2.1792) (−3.8333) ACCM 0.101 −0.00270 t-1 (0.8116) (−0.0292) NCSKEW 0.0483*** t-1 (4.0799) DUVOL 0.0496*** t-1 (4.3202) Interpret 0.3486** 0.2867** (2.0718) (2.3866) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0310 0.0280 Coef. of V_Commit = Coef. M_Commit. F-value 0.00 0.27 This table reports the impact of voluntarily signed PCs in M&A transac- tions and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and ***indicate signiﬁcance at10%, 5%, and 1% levels, respectively. behaviours of information insiders, such ashiding the true valueoftargetassets and provide opportunistic high-performance forecasts, one can expect that the strength of the relation to be stronger for ﬁrms involved in related-party M&As, as hypothesised in H2. We report the result for related-party M&As in column 1 and 2 of Table 6.As reported in column 1and 2of Table 6, when future crash risk is measured by NCSKEW and DUVOL,respectively, thecoeﬃcients of COMMIT are highly signiﬁcant with an expected sign. However, as shown in column 3 and 4 of Table 6,the coeﬃcients of COMMIT are not signiﬁcant in the subsample of non-related party M&As. These results suggest that the positive impact of performance CHINA JOURNAL OF ACCOUNTING STUDIES 331 Table 8. PC contracts, target quality uncertainty and acquiring ﬁrms’ stock price crash risk. (1) (2) NCSKEW DUVOL t t COMMIT 0.0480** 0.0335** t-1 (2.1340) (2.0222) UNCERTAINTY −0.000400 −0.000200 t-1 (−0.3729) (−0.3855) COMMIT* UNCERTAINTY −0.0005 −0.000800 t-1 (−0.3104) (−0.6594) SIZE −0.0396*** −0.0312*** t-1 (−4.3605) (−4.8866) LEV −0.0245 −0.00360 t-1 (−0.4801) (−0.0992) MB 0.0051* 0.0035* t-1 (1.7306) (1.7117) ROA −0.0276 −0.0546 t-1 (−0.1748) (−0.4439) DTURN −0.0546* −0.0359* t-1 (−1.8993) (−1.7145) RET 0.0922*** 0.0655*** t-1 (4.9844) (4.8188) SIGMA 0.657 0.00270 t-1 (1.4582) (0.0081) ACCM 0.0263 −0.0635 t-1 (0.1847) (−0.6107) NCSKEW 0.0621*** t-1 (4.8127) DUVOL 0.0643*** t-1 (5.1048) Interpret 0.3872* 0.3517** (1.9528) (2.5016) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 6655 6655 Adjusted R 0.0590 0.0620 This table reports the impact of target quality uncertainty on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. commitment on acquiring ﬁrms’ future crash risk is dominated by related-party M&As transactions. The above results are consistent with H2. 5.4. Test of H3 Table 7 presents the impact of voluntary performance commitment on acquiring ﬁrms’ future crash risk. We ﬁnd the coeﬃcient of V_COMMIT is not signiﬁcantly diﬀerent from the coeﬃcient of M_COMMIT, no matter measuring stock price crash risk by NCSKEW or DUVOL. This result supports our previous prediction. In short, the results presented in prior tables, taken together, are consistent with the agency theory explanation for performance commitment regulation in M&As. 332 J. LI ET AL. Table 9. PC contracts, assessment institution reputation and acquiring ﬁrms’ stock price crash risk.at10%, (1) (2) NCSKEW DUVOL t t COMMIT 0.0773*** 0.0605*** t-1 (4.0093) (4.3390) TOP_AI 0.0204 −0.00750 t-1 (0.5211) (−0.2653) COMMIT*TOP_AI 0.0175 0.0323 t-1 (0.3253) (0.8277) SIZE −0.0322*** −0.0229*** t-1 (−4.4552) (−4.3786) LEV −0.0344 −0.0145 t-1 (−0.8378) (−0.4896) MB 0.0093*** 0.0065*** t-1 (4.1888) (4.0554) ROA −0.0274 −0.0931 t-1 (−0.1944) (−0.9126) DTURN −0.1501*** −0.1070*** t-1 (−6.8215) (−6.7231) RET 0.0905*** 0.0622*** t-1 (7.3482) (6.9792) SIGMA −0.8020** −1.0089*** t-1 (−2.1478) (−3.7345) ACCM 0.0817 −0.0194 t-1 (0.6799) (−0.2228) NCSKEW 0.0489*** t-1 (4.3153) DUVOL 0.0502*** t-1 (4.4473) Interpret 0.4092*** 0.3289*** (2.6113) (2.8991) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0300 0.0270 This table reports the impact of assessment institution (AI) reputation on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. 6. Additional tests and robustness checks 6.1. Does the target quality uncertainty matter? We measure the target quality uncertainty in two ways. First, we measure the target quality uncertainty by the diﬀerence between assessment valuation and book valua- tion of target assets. Secondly, we measure the target quality uncertainty by the quality of ﬁnancial intermediaries in M&As (i.e. ﬁnancial advisor and assessment institution). We augment Equations (4) and (5) with target quality uncertainty variables and their interactions with performance commitment variable. Tables 8–10 present results for these analyses. The results show that the coeﬃcients of interaction terms are CHINA JOURNAL OF ACCOUNTING STUDIES 333 Table 10. PC contracts, ﬁnancial advisor reputation and acquiring ﬁrms’ stock price crash risk. (1) (2) NCSKEW DUVOL t t COMMIT 0.0780*** 0.0617*** t-1 (4.2102) (4.6043) TOP_FA −0.0057 −0.0130 t-1 (−0.0737) (−0.2334) COMMIT*TOP_FA 0.0568 0.0478 t-1 (0.6320) (0.7339) SIZE −0.0322*** −0.0230*** t-1 (−4.4581) (−4.3994) LEV −0.0346 −0.0143 t-1 (−0.8426) (−0.4812) MB 0.0093*** 0.0065*** t-1 (4.1921) (4.0452) ROA −0.0291 −0.0931 t-1 (−0.2065) (−0.9130) DTURN −0.1496*** −0.1069*** t-1 (−6.7999) (−6.7132) RET 0.0903*** 0.0622*** t-1 (7.3346) (6.9734) SIGMA −0.8063** −1.0100*** t-1 (−2.1600) (−3.7402) ACCM 0.0804 −0.0202 t-1 (0.6689) (−0.2323) NCSKEW 0.0492*** t-1 (4.3448) DUVOL 0.0502*** t-1 (4.4546) Interpret 0.4108*** 0.3308*** (2.6229) (2.9181) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 8137 8137 Adjusted R 0.0300 0.0270 This table reports the impact of ﬁnancial advisor (FA) reputation on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. insigniﬁcant, suggesting performance commitment could increase the acquiring ﬁrms’ future crash risk regardless of the target quality uncertainty and quality of external monitoring. These results further support our conjecture that PC regulation per se is problematic. 6.2. Do institutional investors matter? Institutional investors are more sophistical investors who have more ability to access to private information. The institutional investors may be capable to ‘see through’ the problematic performance commitment regulation. We augment Equations (4) and (5) with institutional ownership variable and its interaction with performance commitment 334 J. LI ET AL. Table 11. PC contracts, institutional ownership and acquiring ﬁrms’ stock price crash risk. (1) (2) NCSKEW DUVOL t t COMMIT 0.1113*** 0.0999*** t-1 (3.5397) (4.4552) INST 0.0155*** 0.0097*** t-1 (7.6286) (6.5258) COMMIT* INST −0.0078* −0.0078** t-1 (−1.7414) (−2.4682) SIZE −0.0332*** −0.0204*** t-1 (−4.0042) (−3.5152) LEV −0.0792* −0.0483 t-1 (−1.6473) (−1.4100) MB 0.0097*** 0.0072*** t-1 (3.5542) (3.8090) ROA −0.0716 −0.0832 t-1 (−0.4407) (−0.6683) DTURN −0.1522*** −0.1065*** t-1 (−6.0979) (−5.9748) RET 0.0830*** 0.0562*** t-1 (6.5044) (5.9416) SIGMA −0.8026** −0.9928*** t-1 (−2.0564) (−3.5363) ACCM 0.0550 −0.0270 t-1 (0.4502) (−0.2970) NCSKEW 0.0443*** t-1 (3.5632) DUVOL 0.0422*** t-1 (3.5210) Interpret 0.3806** 0.2341* (2.1340) (1.8603) Year ﬁxed eﬀects Yes Yes Industry ﬁxed eﬀects Yes Yes No. of observations 7153 7153 Adjusted R 0.0410 0.0340 This table reports the impact of institutional investors on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix. t-Statistics are reported in par- entheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. variable. Table 11 presents the results. Consistent with our prediction, the coeﬃcient of interaction term is signiﬁcantly negative, suggesting that the institutional investors could mitigate the impact of performance commitment on crash risk. 6.3. Performance commitment, information asymmetry and stock price crash risk We further test the impact of information asymmetry on the relation between perfor- mance commitment in M&As and acquiring ﬁrms’ future crash risk. We proxy the informa- tion asymmetry in two ways: ownership of controlling shareholder and the divergence between the control and cash ﬂow rights of controlling shareholder. The results are presented in Tables 12 and 13. The results show that the positive relation between CHINA JOURNAL OF ACCOUNTING STUDIES 335 Table 12. PC contracts, the divergence between the control and cash ﬂow rights of controlling shareholder and acquiring ﬁrms’ stock price crash risk in related-party M&A deals High divergence Low divergence (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t COMMIT 0.0774** 0.0594** 0.0174 0.0207 t-1 (2.0501) (2.3692) (0.5861) (0.9589) SIZE −0.0329** −0.0226** −0.0444*** −0.0364*** t-1 (−2.2362) (−2.3207) (−3.9182) (−4.4288) LEV 0.0244 −0.0203 0.0568 0.0783 t-1 (0.2922) (−0.3528) (0.8405) (1.5943) MB 0.00410 0.00360 0.0097*** 0.0060** t-1 (0.8823) (1.0982) (2.8853) (2.4441) ROA −0.130 −0.0872 0.259 0.0507 t-1 (−0.4659) (−0.4415) (1.2296) (0.3307) DTURN −0.0279 −0.1057*** −0.0490 −0.0416 t-1 (−0.5196) (−3.3129) (−1.2406) (−1.4512) RET 0.0447 0.0572*** 0.0749*** 0.0626*** t-1 (1.3731) (3.4226) (2.9982) (3.4530) SIGMA 0.670 −1.0034** 0.622 −0.223 t-1 (0.8265) (−1.9849) (0.9533) (−0.4712) ACCM 0.102 0.153 0.194 0.0542 t-1 (0.4324) (0.8948) (1.0401) (0.4011) NCSKEW 0.0513** 0.0588*** t-1 (2.3127) (3.3705) DUVOL 0.0227 0.0621*** t-1 (1.0478) (3.5761) Interpret 0.317 0.257 0.5936** 0.5770*** (0.9692) (1.1988) (2.3464) (3.1402) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 2220 2220 3344 3344 Adjusted R 0.0390 0.0180 0.0720 0.0720 This table reports the impact of divergence between the control and cash ﬂow rights of controlling shareholder on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. performance commitment and acquiring ﬁrms’ future is dominated by the subsample of M&A transactions with higher information asymmetry. 6.4. Propensity Score Matching (PSM) method We also use the Propensity Score Matching (PSM) method to eliminate the endogeneity problem. Table 4 shows that there is a signiﬁcant diﬀerence in the mean of the control variables for the two groups categorised by the presence or absence of performance commitment in M&As. It could be argued that the positive eﬀect of performance commit- ment on future crash risk may be due to the diﬀerence in corporate fundamental factors. To eliminate this endogeneity problem, we also use the closest PSM method to re-test my main hypotheses. Table 14 reports the results using the PSM method. The results are quite similar to the main test reported above. So, our main results are unlikely to be driven by ﬁrm characteristics. 336 J. LI ET AL. Table 13. PC contracts, the ownership of controlling shareholder and acquiring ﬁrms’ stock price crash risk in related-party M&A deals. High ownership of controlling shareholder Low ownership of controlling shareholder (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t COMMIT 0.1068*** 0.0942*** 0.0277 0.00780 t-1 (3.2760) (4.0594) (0.8745) (0.3348) SIZE −0.0207* −0.0199** −0.0440*** −0.0372*** t-1 (−1.6680) (−2.2467) (−3.3032) (−3.8123) LEV 0.0629 0.0758 0.000400 −0.00150 t-1 (0.7827) (1.3227) (0.0054) (−0.0297) MB 0.0152*** 0.0071** 0.00500 0.0042* t-1 (3.2527) (2.1240) (1.4711) (1.6766) ROA 0.370 0.3130* −0.167 −0.255 t-1 (1.4227) (1.6887) (−0.7596) (−1.5825) DTURN −0.1214*** −0.0753** −0.0771* −0.0857*** t-1 (−2.9459) (−2.5670) (−1.8031) (−2.7356) RET 0.0868*** 0.0654*** 0.0677** 0.0512*** t-1 (3.9316) (4.1559) (2.5064) (2.5844) SIGMA −1.4041** −1.4586*** 0.991 0.0338 t-1 (−2.0968) (−3.0584) (1.4495) (0.0675) ACCM 0.182 −0.00910 0.111 0.164 t-1 (0.8570) (−0.0604) (0.5413) (1.0930) NCSKEW 0.0276 0.0585*** t-1 (1.4294) (3.0280) DUVOL 0.0225 0.0594*** t-1 (1.1721) (3.1164) Interpret 0.0310 0.186 0.5009* 0.4857** (0.1128) (0.9529) (1.7256) (2.2803) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes No. of observations 2713 2713 2851 2851 Adjusted R 0.0270 0.0260 0.0550 0.0610 This table reports the impact of the ownership of controlling shareholder on the relation between PC contracts and acquiring ﬁrms’ future stock price crash risk. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in all regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. 6.5. Firm ﬁxed eﬀects regressions Since the empirical literature on forecasting crash risk is relatively new, it is possible that our analysis omits from the regressions some crash determinants that correlated with other included variables. To mitigate potential problems that can arise form correlated omitted variables, we follow Petersen (2009), we re-estimate regressions in Table 5 after controlling for ﬁrm-ﬁxed eﬀects and using the standard errors corrected for year cluster- ing. Table 15 presents the results of this exercise. As shown in Table 15, the relation between performance commitment in M&As and future crash risk remains highly sig- niﬁcant with an expected positive sign regardless of crash measures, suggesting that our results reported in Table 5 are unlikely to be driven by omitted correlated time-invariant variables. CHINA JOURNAL OF ACCOUNTING STUDIES 337 Table 14. The results of Propensity Score Matching (PSM) regressions.. First stage: COMMIT Second-stage: Crash risk Column(1): NCSKEW Column(2):DUVOL t t Variables Coeﬃcient Z-stat Coeﬃcient t-stat Coeﬃcient t-stat COMMIT 0.0734*** 3.2198 0.0415*** 2.6949 t-1 SIZE −0.1481*** −3.14 −0.0473*** −4.7896 −0.0382*** −5.8434 t-1 LEV −0.9556 −3.82 0.0163 0.2913 0.0294 0.8002 t-1 MB −0.0120 −1.00 0.0042 1.3160 0.00220 1.0979 t-1 ROA −0.5682 −0.70 0.0658 0.3925 −0.0522 −0.4395 t-1 DTURN −0.1649 −1.39 −0.0245 −0.7229 −0.0202 −0.8903 t-1 RET 0.2124*** 3.04 0.1147*** 5.5597 0.0863*** 5.7393 t-1 SIGMA 0.6489 0.33 −0.115 −0.2193 −0.503 −1.3077 t-1 ACCM 0.1241 0.17 0.124 0.8648 0.0297 0.2911 t-1 NCSKEW 0.0646*** 4.6866 t-1 DUVOL 0.0715*** 5.5018 t-1 MAJOR 1.8124*** 18.51 t-1 RELATED 0.1562 1.53 t-1 Interpret 0.6752 0.66 0.6699*** 3.0621 0.5857*** 4.0827 Year No Yes Yes Industry Yes Yes Yes N 8137 6372 6372 Pseudo R2 0.1451 Adj. R 0.0600 0.0620 This table reports the results of propensity score matching (PSM) regressions. The ﬁrst stage of the procedure involves a logit analysis. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by ﬁrm. Year and industry ﬁxed eﬀects are included in second-stage regressions. Here *, **, and *** indicate signiﬁcance at10%, 5%, and 1% levels, respectively. 6.6. The impact of fulﬁlment rate of performance commitment We further test whether the fulﬁlment rate of performance commitment inﬂuent future crash risk of acquiring ﬁrms. We keep the ﬁrm years with both performance forecast and realised performance and construct two variables (i.e. FULFILL1 and FULFILL2) to measure the fulﬁlment ratio of performance commitment. Table 16 presents the results. As the results shown, the commitment fulﬁlment rate does not have a signiﬁcant impact on acquiring ﬁrms’ future crash risk. These results further support our conclusion that PC regulation per se is problematic. 6.7. Controlling more acquisition characteristics In this section, we further control some important acquisition characteristics which may inﬂuent the acquiring ﬁrms’ future crash risk. Directional private placement is common- used in corporate ﬁnancing for acquisitions. Previous literature ﬁnds the acquisitions with directional private placement ﬁnancing have severer information asymmetries. We further control for the ﬁnancing way of acquirer ﬁrms. We construct two variables. DPPt-1 is a dummy variable which equals 1 if the acquirers obtain the ﬁnancing by directional private placement, and 0 otherwise. CDPP t-1 is a dummy variable which equals 1 if the acquirers obtain the ﬁnancing by directional private placement and controlling shareholder is involved in the directional private placement, and 0 otherwise. As shown in Table 17,the main resultsremain after controlling these two variables. 338 J. LI ET AL. Table 15. Firm ﬁxed eﬀects regressions. (1) (2) NCSKEW DUVOL t t COMMIT 0.0796** 0.0543** t-1 (3.1072) (2.6811) SIZE 0.0008 0.0132 t-1 (0.0145) (0.3327) LEV −0.0483 −0.0513 t-1 (−0.6385) (−1.0631) MB 0.0086** 0.0068* t-1 (2.3569) (2.2613) ROA −0.2445 −0.2013 t-1 (−1.6958) (−1.7275) DTURN −0.1788*** −0.1366*** t-1 (−3.7770) (−3.7816) RET 0.0965*** 0.0692** t-1 (3.7303) (3.1671) SIGMA −0.460 −0.7299 t-1 (−0.4050) (−0.8987) ACCM 0.1525* 0.0373 t-1 (1.9300) (0.4318) NCSKEW −0.1516** t-1 (−2.9174) DUVOL −0.1515** t-1 (−3.0612) Constant −0.3934 −0.5176 (−0.3429) (−0.5964) Firm FE Yes Yes Cluster by year Yes Yes N 8137 8137 Adj. R 0.0900 0.0900 This table reports the results of ﬁrm ﬁxed eﬀects regressions. We re- estimate regressions in Table 5 after controlling for ﬁrm-ﬁxed eﬀects and using the standard errors corrected for year clustering. All variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors clustered by year. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. 6.8. Multi-period future stock price crash risk In our OLS regressions thus far, we examine the predictive ability of our performance commitment variable with respect to future crash risk. In measuring crash risk, we consider future crash occurrences in the 1-year-ahead forecast window. In this section, we re-examine the impact of performance commitment on future crash risk by using t3- and 4-year-ahead crash risk. As seen in Table 18,the relation between performance commitment in M&As and future three- and 4-year-ahead crash risk remains positive. 7. Conclusion We analyse the impact of performance commitment in M&As on future stock price crash risk of acquiring ﬁrms. The PC regulation in M&As is an innovative policy made by China authorities. PC regulation is expected to work as a mechanism to alleviate information asymmetry, thereby reduce adverse selection and overvaluation risk in M&A. However, CHINA JOURNAL OF ACCOUNTING STUDIES 339 Table 16. The fulﬁlment rate of PCs and acquiring ﬁrms’ stock price crash risk. (1) (2) (3) (4) NCSKEW DUVOL NCSKEW DUVOL t t t t FULFILL1 0.0256 0.00900 t-1 (0.4728) (0.2253) FULFILL2 −0.00250 −0.00100 t-1 (−0.1657) (−0.1095) SIZE −0.0323 −0.00530 −0.0356 −0.0107 t-1 (−1.0707) (−0.2424) (−1.1939) (−0.4888) LEV −0.00410 −0.0398 0.0260 −0.00790 t-1 (−0.0280) (−0.3627) (0.1786) (−0.0720) MB 0.000700 0.00290 −0.000100 0.00210 t-1 (0.1456) (0.7279) (−0.0134) (0.5220) ROA −0.146 −0.263 −0.105 −0.202 t-1 (−0.2661) (−0.5387) (−0.1887) (−0.4086) DTURN −0.0725 −0.0563 −0.0524 −0.0421 t-1 (−1.2685) (−1.2296) (−0.9405) (−0.9392) RET 0.1326*** 0.0993*** 0.1280*** 0.0979*** t-1 (3.7394) (3.6815) (3.6011) (3.6232) SIGMA 0.0979 −0.967 0.0477 −1.058 t-1 (0.1113) (−1.4912) (0.0544) (−1.6437) ACCM −0.6571** −0.5367** −0.6746** −0.5312** t-1 (−1.9839) (−2.1436) (−2.0315) (−2.1126) NCSKEW 0.0158 0.0208 t-1 (0.5307) (0.6922) DUVOL 0.00170 0.00790 t-1 (0.0578) (0.2532) Interpret 0.0150 −0.370 0.109 −0.241 (0.0244) (−0.8373) (0.1801) (−0.5475) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes N 1000 1000 1000 1000 Adj. R 0.0330 0.0290 0.0310 0.0270 The fulﬁlment rate of PCs and acquiring ﬁrms’ stock price crash risk. This table estimates the impact of fulﬁlment ratio of performance commitment in M&A transactions and acquiring ﬁrms’ future stock price crash risk. All other variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors corrected for clustering by ﬁrm. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. due to the theoretical and institutional defects related to current performance commit- ment regulation, this regulation tends to provide the parties with private information an opportunity to overstate the values of inferior target assets. We ﬁnd that the performance commitment increases the acquiring ﬁrms’ future stock price crash risk. Furthermore, we ﬁnd the relation is more pronounced in related-party M&As. These ﬁndings are consistent with the hypothesis that controlling shareholders expropriate from minority shareholders of acquirers. We provide two main explanations for this ﬁnding. The ﬁrst explanation is based on signalling theory. The performance commitment in M&A is a costless commitment rather than a costly signal. The default cost is not high enough to constrain the promiserstomakeanopportunistic commitment. Secondly, with the respect to institutional environment, in M&A transactions, major shareholders of listed compa- nies may collude with the commitment parties to make ‘opportunistic’ performance commitment, resulting in the agency conﬂicts between them and minority share- holders. Therefore, performance commitment does not work as a mechanism to 340 J. LI ET AL. Table 17. The results after controlling more charac- teristics of acquisitions. (1) (2) NCSKEWt DUVOLt COMMIT 0.0578** 0.0367** t-1 (2.4428) (2.0986) DPP −0.0312 −0.0179 t-1 (−1.1710) (−0.9324) CDPP 0.0127 0.00100 t-1 (0.4318) (0.0474) SIZE −0.0387*** −0.0320*** t-1 (−4.7234) (−5.6398) LEV −0.00760 0.0270 t-1 (−0.1651) (0.8291) MB 0.0062** 0.0033* t-1 (2.3331) (1.8416) ROA 0.0497 0.00190 t-1 (0.3513) (0.0170) DTURN −0.0361 −0.0243 t-1 (−1.3681) (−1.2738) RET 0.0818*** 0.0604*** t-1 (4.9732) (5.0268) SIGMA 0.591 −0.0954 t-1 (1.4202) (−0.3140) ACCM 0.0549 −0.0209 t-1 (0.4423) (−0.2302) NCSKEW 0.0664*** t-1 (5.5775) DUVOL 0.0682*** t-1 (5.9679) Interpret 0.4994*** 0.5064*** (2.7666) (4.0513) Year Yes Yes Industry Yes Yes N 8137 8137 Adj. R 0.0570 0.0580 This table reports the results after controlling more character- istics of acquisitions. DPP is a dummy variable which equals t-1 1 if the acquirers obtain the ﬁnancing by directional private is a dummy variable placement, and 0 for otherwise. CDPP t-1 which equals 1 if the acquirers obtain the ﬁnancing by direc- tional private placement and controlling shareholder is involved in directional private placement, and 0 for otherwise. All other variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors corrected for clustering by ﬁrm. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. reduce information asymmetry, but drive the parties with private information to hide the true value of the target assets; the stock price will suddenly drop when the accumulated hidden bad news release to the market. We argue that the current performance commitment regulation may not be suitable for the China market in which type II agency conﬂict is dominated. Our ﬁndings suggest that policy design and implementation need to ﬁt with China institutional environment. Our results shed some light on how to revise the PC regulation in M&As. CHINA JOURNAL OF ACCOUNTING STUDIES 341 Table 18. Multi-period future stock price crash risk. (1) (2) (3) (4) NCSKEW NCSKEW DUVOL DUVOL t+3 t+4 t+3 t+4 COMMIT 0.0683 0.1623* 0.0723** 0.0811 t-1 (1.4192) (1.8404) (2.1161) (1.2964) SIZE −0.0704*** −0.0732*** −0.0517*** −0.0569*** t-1 (−6.7869) (−6.3081) (−7.0138) (−6.9209) LEV 0.1891*** 0.1177* 0.1313*** 0.0926* t-1 (3.0527) (1.6689) (2.9860) (1.8528) MB 0.00280 0.00450 0.00130 0.00110 t-1 (0.7477) (1.0824) (0.4952) (0.3540) ROA 0.7397*** 0.4387* 0.4134*** 0.203 t-1 (3.5828) (1.9467) (2.8221) (1.2734) DTURN −0.000900 −0.0132 0.0147 −0.0242 t-1 (−0.0219) (−0.2834) (0.4953) (−0.7327) RET −0.0353 −0.0295 0.00680 −0.0155 t-1 (−1.4220) (−1.0499) (0.3846) (−0.7760) SIGMA 0.598 1.6508* −0.100 0.702 t-1 (0.7387) (1.7266) (−0.1752) (1.0395) ACCM −0.0502 0.0255 −0.0619 −0.0567 t-1 (−0.2851) (0.1286) (−0.4953) (−0.4028) NCSKEW 0.0459*** 0.0474** t-1 (2.5926) (2.2433) DUVOL 0.0390** 0.0377* t-1 (2.2730) (1.8868) Interpret 0.9799*** 1.0584*** 0.8214*** 0.9288*** (4.2149) (4.0495) (4.9799) (5.0166) Year ﬁxed eﬀects Yes Yes Yes Yes Industry ﬁxed eﬀects Yes Yes Yes Yes N 3865 2937 3865 2937 Adj. R 0.0600 0.0730 0.0620 0.0730 This table tests the impact of PC contracts and acquiring ﬁrms’ multi-period future stock price crash risk. All other variables are deﬁned in Appendix 1. t-Statistics are reported in parentheses and are based on standard errors corrected for clustering by ﬁrm. Here *, **, and *** indicate signiﬁcance at 10%, 5%, and 1% levels, respectively. Disclosure statement No potential conﬂict of interest was reported by the authors. Funding This work was supported by the National Natural Science Foundation of China [71702196, 71772181, 71702036]; Fundamental Research Funds for the Central Universities [17wkpy16]. References Allen, F., Qian, J., & Qian, M. (2005). Law, ﬁnance, and economic growth in China. Journal of Financial Economics, 77(1), 57–116. Ang, J.S., Cole, R.A., & Lin, J.W. (2000). Agency costs and ownership structure. The Journal of Finance, 55(1), 81–106. Cadman, B., Carrizosa, R., & Faurel, L. (2014). Economic determinants and information environment eﬀects of earnouts: New insights from SFAS 141(r). Journal of Accounting Research, 52(1), 37–74. Cain, M.D., Denis, D.J., & Denis, D.K. (2011). Earnouts: A study of ﬁnancial contracting in acquisition agreements. Journal of Accounting and Economics, 51(1–2), 151–170. Chen, J., Hong, H., & Stein, J.C. (2001). Forecasting crashes: Trading volume, past returns, and conditional skewness in stock prices. Journal of Financial Economics, 61(3), 345–381. 342 J. LI ET AL. Cheynel, E. (2013). A theory of voluntary disclosure and cost of capital. Review of Accounting Studies, 18(4), 987–1020. Chu, J., & Fang, J. (2016). Margin-trading, short-selling and the deterioration of crash risk. Economic Research Journal, 5, 143–158 (in Chinese). Datar, S., Frankel, R., & Wolfson, M. (2001). Earnouts: The eﬀects of adverse selection and agency costs on acquisition techniques. Journal of Law Economics and Organization, 17(1), 201–238. Deng, J., Zeng, Y., & He, J. (2011). The root and the eﬀects of relative merger and acquisition. Chinese Journal of Management, 8, 1238–1246 (in Chinese). Francis, J, Nanda, D, & Olsson, P. (2008). Voluntary disclosure, earnings quality, and cost of capital. Journal Of Accounting Research, 46(1), 53–99. doi:10.1111/j.1475-679X.2008.00267.x Hou, Q., Jin, Q., Yang, R., Yuan, H., & Zhang, G. (2015). Performance commitments of controlling shareholders and earnings management. Contemporary Accounting Research, 32(3), 1099–1127. Hutton, A.P., Marcus, A.J., & Tehranian, H. (2009). Opaque ﬁnancial reports, R2, and crash risk. Journal of Financial Economics, 94(1), 67–86. Ji, H., & Ma, L. (2016). Regulatory soft constraints, false compensation promises and investor protection. Communication of Finance and Accounting, 15, 73–78 (in Chinese). Ji, H., Wei, M., & Liu, J. (2010). Asset injection, securities market regulation and performance. Accounting Research,2,47–56 (in Chinese). Jiang, G., Lee, C.M.C., & Yue, H. (2010). Tunneling through intercorporate loans: The China experience. Journal of Financial Economics, 98(1), 1–20. Jin, L., & Myers, S.C. (2006). R2 around the world: New theory and new tests. Journal of Financial Economics, 79(2), 257–292. Kim, J.-B., Li, Y., & Zhang, L. (2011a). CFOs versus CEOs: Equity incentives and crashes. Journal of Financial Economics, 101(3), 713–730. Kim, J.-B., Li, Y., & Zhang, L. (2011b). Corporate tax avoidance and stock price crash risk: Firm level analysis. Journal of Financial Economics, 100(3), 639–662. Li, Z., Yu, Q., & Wang, X. (2005). Tunneling, propping and M& A: Evidence from Chinese listed companies. Economic Research Journal, 2005(1), 95–105. (in Chinese). Liu, F., Zhong, R., & Jin, T. (2007). The transfer and “looting” of listed companies’ mastery under loose legal control. Management World, 12, 106–116 (in Chinese). Lv, C., & Han, H. (2014). VAM, synergy and distribution of gains from M&A. Audit & Economy Research, 6, 3–13 (in Chinese). Pan, A., Qiu, J., & Yang, Y. (2017). Research on the incentive eﬀect of valuation adjustment mechan- ism in M&As: Evidence from listed companies on SEM and GEM board in China. Accounting Research,3,46–52 (in Chinese). Petersen, M.A. (2009). Estimating standard errors in ﬁnance panel data sets: comparing approaches. The Review Of Financial Studies, 22(1), 435–480. doi:10.1093/rfs/hhn053 Ragozzino, R., & Reuer, J.J. (2009). Contingent earnouts in acquisitions of privately-held targets. Journal of Management, 35(4), 857–879. Rey, P., & Salanié, B. (1996). On the value of commitment with asymmetric information. Econometrica, 64(6), 1395–1414. Spence, M. (1973). Job market signaling. Quarterly Journal of Economics, 87(3), 355–374. Tang, Q., & Han, H. (2018). Related party M&As and ﬁrm value: The governance eﬀect of accounting conservatism. Nankai Business Review,3, 25–36 (in Chinese). Wei, M., Cai, G., & Cheng, M. (2016). A comprehensive analyzing framework of ownership character- istics: Based on the Chinese phenomena and theory. Accounting Research,5,26–33 (in Chinese). Xie, J., & Zhang, Q. (2016). Accounting treatments for proﬁt compensation commitment in listed companies’ holding mergers: Case studies of ﬁve China listed companies. Accounting Research,6, 15–20 (in Chinese). Xu, N., Jiang, X., Yi, Z., & Xu, X. (2012). Conﬂicts of interest, analyst optimism and stock price crash risk. Economic Research Journal, 7, 127–140 (in Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 343 Xu, N., Yu, S., & Yi, Z. (2013). Institutional investor herding and stock price crash risk. Management World,7, 31–43 (in Chinese). Xu, N., Zhang, H., & Wu, S. (2008). Do additional commitments have signaling eﬀect? Evidence from the non-tradable share reform in China. Management World, 3, 142–151 (in Chinese). 344 J. LI ET AL. Appendix 1. Variable deﬁnitions Variables Deﬁnitions NCSKEW Negative skewness of ﬁrm-speciﬁc weekly returns over the ﬁscal year period. See Equation (2) for details. DUVOL The natural logarithm of the ratio of the standard deviation of down weeks to the standard deviation of up weeks. See Equation (3) for details. COMMIT A dummy variable with a value of 1 if the merged ﬁrm is in the post-performance commitment period, and 0 otherwise. DTURN The average monthly share turnover over the current ﬁscal year period minus the average monthly share turnover over the previous ﬁscal year period. RET The mean of ﬁrm-speciﬁc weekly returns over the ﬁscal year period, times 100. SIGMA The standard deviation of ﬁrm-speciﬁc weekly returns over the ﬁscal year period. ROA The net income divided by lagged total assets SIZE The log of total assets LEV The total liability divided by total assets MB The market value of equity divided by the book value of equity ACCM The absolute value of discretionary accruals, where discretionary accruals are estimated from modiﬁed Jones model. RELATED A dummy variable equals 1 if the acquisition is a related-party transaction, and 0 for otherwise. MAJOR A dummy variable equals 1 if the acquisition is a signiﬁcant asset restructuring, and 0 for otherwise. V_COMMIT A dummy variable equals 1 if the PC contract is made voluntarily, and 0 for otherwise. M_COMMIT A dummy variable equals 1 if the PC contract is made mandatorily, and 0 for otherwise. FULFILL1 A dummy variable equals 1 if the acquirees meet the performance forecast, and 0 for otherwise. FULFILL2 The realised performance of target divided by the performance forecast UNCERTAINTY The absolute value of the diﬀerence between assessment value of target assets and book value of target asset, divided by book value of target assets TOP_AI A dummy variable equals 1 if the assessment institution is 1 of top 10 institutions, and 0 for otherwise TOP_FA A dummy variable equals 1 if the ﬁnancial advisor is 1 of top 10 institutions, and 0 for otherwise. INST The total ownership of institutional investors. DPP A dummy variable equals 1 if the acquiers obtain ﬁnancing through directional private placement, and 0 for otherwise. CDPP A dummy variable equals 1 if the controlling shareholder is involved in directional private placement, and 0 for otherwise.

Journal

China Journal of Accounting Studies – Taylor & Francis

Published: Jul 3, 2019

Keywords: Performance commitment in M&A; stock price crash risk; agency conflict; signalling theory; costless signal

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

Learn More →

Performance commitment in M&As and stock price crash risk

Performance commitment in M&As and stock price crash risk

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

Learn More →

Performance commitment in M&As and stock price crash risk

Performance commitment in M&As and stock price crash risk

References (35)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies