Earnings management, audit adjustments, and the financing of corporate acquisitions: Evidence from China

Transcription

1 ACCOUNTING WORKSHOP Earnings management, audit adjustments, and the financing of corporate acquisitions: Evidence from China By Clive Lennox* Leventhal School of Accounting, University of Southern California Zi-Tian Wang School of Accountancy, Shanghai University of Finance and Economics, China Xi Wu School of Accountancy, Central University of Finance and Economics, China Thursday, June 2 nd, :20 2:50 p.m. Room C06 *Speaker Paper Available in Room 447

2 Earnings management, audit adjustments, and the financing of corporate acquisitions: Evidence from China Clive Lennox Leventhal School of Accounting, University of Southern California, USA Zi-Tian Wang School of Accountancy, Shanghai University of Finance and Economics, China Xi Wu School of Accountancy, Central University of Finance and Economics, China May 2016 Acknowledgements: This study has benefited from the comments of Udi Hoitash, Arnie Wright, and workshop participants at the University of Bath, Michigan State University, and Northeastern University. Xi Wu thanks the Inspection Bureau of the Chinese Ministry of Finance for providing data support.

3 Earnings management, audit adjustments, and the financing of corporate acquisitions: Evidence from China ABSTRACT Acquirers are motivated to overstate earnings prior to stock-financed acquisitions because this can increase the acquirer s stock price, making it less costly to purchase a target using the acquiring company s stock. We hypothesize that audits help to correct this tendency to overstate earnings prior to stock-financed acquisitions. We test this using a difference-in-differences research design, which compares audit adjustments for stock-financed and cash-financed acquirers in the periods before versus after the acquisitions. Consistent with auditors correcting earnings overstatements prior to stock-financed acquisitions, we find that auditors require larger downward adjustments to earnings in the year before stock-financed acquisitions.

4 1. Introduction When a company pays for an acquisition using its own stock, the cost of the acquisition is lower when the acquirer s stock price is higher. This means that acquirers have incentives to overstate earnings when they plan to finance acquisitions using stock rather than cash. 1 Auditors are responsible for testing whether the acquirers financial statements are fairly presented and auditors should require downward adjustments to earnings when they find that earnings are overstated. We argue that stock-financed acquirers are motivated to overstate earnings and audits help to detect and correct these earnings overstatements. Accordingly, we hypothesize that auditors help to prevent low quality financial reporting prior to stock-financed acquisitions by detecting managers earnings overstatements and requiring corrections through downward adjustments to earnings. However, the results may not support this hypothesis for several reasons. First, the managers of stock-financed acquirers may choose to report conservatively rather than overstate earnings because target companies and potential litigants are likely to carefully scrutinize the acquirer s financial statements when the acquirer uses its own equity to pay for the acquisition (Ball and Shivakumar, 2008). 2 Second, when managers are motivated to overstate earnings they are likely to try and hide the misstatements from their auditors. 1 Although stock financing increases an acquirer s incentives to overstate earnings, stock financing can also help to deter earnings overstatements by the target company. For example, Hansen (1987) argues that any ex-post losses arising from overpayment cause the combined company s stock price to decline, and these losses are shared between the target s and acquirer s shareholders. Therefore, stockfinanced acquirers are able to protect themselves from overpaying for a target, by paying for the acquisition using equity rather than cash. Prior research is consistent with target companies being motivated to overstate earnings prior to the acquisition announcement date (Anilowski et al., 2009; Marquardt and Zur, 2010). However, research also suggests that target companies understate earnings during the period after the announcement date and before the completion date in order to provide a boost to the acquirer s earnings and cash flows during the post-acquisition period (Chen et al., 2016b). In our sample, very few of the target companies are publicly traded so we do not examine the audit adjustments of target companies. Instead, we focus on the audit adjustments of acquiring companies. 2 Consistent with this argument, Ball and Shivakumar (2008) find that U.K. companies report conservatively prior to Initial Public Offerings (IPOs). We examine stock-financed acquisitions rather than IPOs because the audit adjustment data in China are only available after a company becomes publicly traded. We do not examine Seasoned Equity Offerings (SEOs) because in China equity is often sold to corporate insiders and existing major shareholders. This means that Chinese companies do not have strong incentives to overstate earnings prior to SEOs. 1

5 Therefore, the benefits of an audit may be greater for correcting accidental misstatements rather than deliberate misstatements (Botosan et al., 2016). Furthermore, an acquiring company is unlikely to inform its auditor about an impending stock-financed acquisition until the company is ready to make a public announcement about the acquisition. One reason is the need for secrecy prior to the M&A announcement date. Another reason is that, if an acquirer informed its auditor about an impending stock-financed acquisition, the auditor would likely scrutinize the accounts more carefully and charge a higher audit fee. Finally, our analysis of audit adjustments is based on data from China, where the legal enforcement and investor protection are relatively weak as compared with more developed markets (Allen et al., 2005; Chen et al., 2013; Wong, 2014; Chen et al., 2016a) and the threat of legal liability toward auditors is relatively low as well (Chan et al., 2006; Chen et al., 2011). This means that Chinese auditors may not have strong incentives to detect and correct earnings overstatements. 3, 4 We test our hypothesis using a difference-in-differences research design. The treatment sample comprises stock-financed acquirers (STOCK = 1), while the control sample comprises cash-financed acquirers (STOCK = 0). We code the fiscal year-end immediately before the acquisition announcement as the pre-period (BEFORE = 1) and the fiscal year-end immediately after the acquisition completion (or termination) date as the post-period (BEFORE = 0). The treatment variable is the interaction term, STOCK BEFORE. Consistent 3 We undertake our study in China because the data on audit adjustments are generally unavailable in other countries. In China, it is mandatory for every audit firm to report to the Ministry of Finance (MOF) the pre-audit and audited values of earnings and total assets for their publicly traded clients. The pre-audit data are not publicly available but the MOF has provided these data to us and other researchers for the purpose of academic study. Audit firms report the pre-audit values of earnings and total assets, but they are not required to report to the MOF other line items from the pre-audit financial statements across our whole sample period, such as the pre-audit values of operating cash flows. 4 Although auditors in China face a relatively low threat of civil litigation, they can suffer severe punishments from the regulatory agencies (Chen et al., 2011) and Chinese auditors increasingly have market-based incentives to develop and maintain reputations for high quality auditing (Chen et al., 2010; He et al., 2016). 2

6 with the hypothesis, we find that auditors require larger downward adjustments to earnings in the year before stock-financed acquisitions. This suggests that stock-financed acquirers attempt to overstate earnings and auditors help to detect and correct these overstatements by requiring earnings to be adjusted downwards. In a supplementary analysis, we examine whether auditors are able to resolve all of the accounting issues before the auditor issues an opinion on the acquirer s financial statements. If an acquirer refuses to make a necessary adjustment, the auditor can respond by disclosing the problem in the audit report. We therefore examine whether auditors reports disclose unresolved problems at the end of the audit. We expect that such problems are more likely to be disclosed prior to stock-financed acquisitions because managers of stock-financed acquirers have incentives to overstate earnings. We test this using the same difference-in-differences research design that we use in our tests of audit adjustments. We find that auditors reports are more likely to disclose accounting problems in the period prior to stock-financed acquisitions. This suggests that not all of the attempts to overstate earnings are corrected through audit adjustments. Thus, the quality of the audited financial statements remains relatively low prior to stock-financed acquisitions even though audit adjustments are helpful for correcting earnings overstatements. Our study contributes to two distinct literatures. Prior studies in the auditing literature typically examine the cross-sectional associations between various proxies for earnings management and audit characteristics. For example, Becker et al. (1998) examine the association between accruals and Big N auditors. 5 A short-coming of prior studies is that the audit characteristics examined in prior studies are typically endogenous (e.g., Big N) which makes it difficult to draw inferences. Another short-coming is that the cross-sectional association tests do not reveal what auditors actually do to curb earnings management. Our 5 Lennox et al. (2016) focus on the consequences of audit adjustments whereas our study focuses on the determinants of audit adjustments. 3

7 study uses information on audit adjustments to provide more direct evidence about how exactly auditors curb earnings management. Moreover, we employ a difference-indifferences research design to strengthen our causal inferences. In addition, our study examines a setting namely stock-financed acquisitions in which managers are strongly motivated to overstate earnings. 6 Our study also contributes to the earnings management literature. Prior studies have used accruals variables to test whether managers overstate earnings before stock-financed acquisitions, but the extant findings are rather mixed (Erickson and Wang, 1999; Louis, 2004; Heron and Lie, 2002; Pungaliya and Vijh, 2009). 7 The mixed results could be attributable to well-documented biases and noise in accruals. The biases are particularly problematic around significant corporate events such as acquisitions (Hribar and Collins, 2002). The biases are potentially even more problematic when companies pay for acquisitions using stock rather than cash because stock-financed acquirers tend to have stronger growth prospects than cash-financed acquirers, which means that stock-financed acquirers tend to have larger signed accruals even if they are not engaged in earnings management. Pungaliya and Vijh (2009) argue that this could explain why some studies have found abnormally large accruals prior to stock-financed acquisition. 8 Moreover, accruals are notoriously noisy measures of earnings management (Dechow et al., 1995; Subramanyam, 1996; Dechow and Dichev, 2002; Tucker and Zarowin, 2006; Dechow et al., 2010). This could explain why some 6 There is some experimental evidence on auditors decisions to waive proposed audit adjustments. For example, Cohen et al. (2011) find that auditors are more likely to waive adjustments when the CEO has more influence over the independence of the audit committee. Our study is different because we focus on non-waived rather than waived audit adjustments and we examine the setting of stockfinanced acquisitions. 7 Erickson and Wang (1999) and Louis (2004) report significant evidence of earnings management. In contrast, Heron and Lie (2002) and Pungaliya and Vijh (2009) find insignificant results using larger samples and alternative research designs. 8 The problem of bias is greater when researchers calculate working capital accruals from the balance sheet (Hribar and Collins, 2002). This type of bias is also affected by the method of financing because companies have less need to convert working capital into cash when acquisitions are financed using stock rather than cash (Collins et al., 2012). 4

8 accruals studies fail to find significant evidence of earnings management prior to stockfinanced acquisitions (Heron and Lie, 2002; Pungaliya and Vijh, 2009). Another explanation for the insignificant results is that managers attempts to overstate earnings are detected and corrected by auditors before the audited financial statements are released to investors. The confounding factors that create bias and noise in accruals affect both the preaudit and audited earnings of a given company in the same fiscal year. These confounds are eliminated in our analysis of audit adjustments because an audit adjustment captures the difference between pre-audit and audited earnings (see Section 2.3 for further details). Our audit adjustments results are consistent with managers attempting to overstate earnings prior to stock-financed acquisitions and with auditors correcting these overstatements before the audited financial statements are issued to investors. Section 2 discusses the prior literature on earnings management prior to stockfinanced acquisitions and the role of auditors in curbing earnings management. It then develops our hypothesis for audit adjustments. Section 3 describes the difference-indifferences research design and presents the sample and descriptive statistics. Section 4 reports the main results for audit adjustments. Section 5 reports the results of supplementary analyses examining audit reporting decisions, propensity score matching, and audit fees. Section 6 concludes. 2. Prior research and hypothesis development 2.1 Earnings management prior to stock-financed acquisitions Prior studies hypothesize that companies are motivated to manage earnings upward prior to stock-financed acquisitions. This earnings management hypothesis has been tested using accruals variables. In a sample of 55 stock-financed acquisitions, Erickson and Wang (1999) 5

9 find that signed discretionary accruals are larger in the period before stock-financed acquisitions compared with the period afterwards. Louis (2004) examines working capital accruals for a sample of 236 stock-financed acquirers. Similar to Erickson and Wang (1999), Louis (2004) finds that working capital accruals are larger prior to stock-financed acquisitions compared with afterwards. 9, 10 Unfortunately, accruals are susceptible to producing false positives in tests of earnings management. The risk of false positives is particularly high when researchers examine changes in accruals around significant corporate transactions such as acquisitions and stock offerings (Hribar and Collins, 2002; Dechow et al., 2010; Ball, 2013). 11 Moreover, companies tend to pay for acquisitions using stock rather than cash when they expect that they will need to retain cash to finance future growth (Collins et al., 2012). Thus, stockfinanced acquirers typically have better growth prospects than cash-financed acquirers. Moreover, high-growth companies tend to have larger signed accruals (Fairfield et al., 2003). Therefore, stock-financed acquirers are likely to exhibit larger signed accruals due to their strong growth prospects even if they are not managing earnings upwards (Pungaliya and Vijh, 2009). 12 In addition, when an acquirer chooses to finance the acquisition using its own stock rather than cash, the acquirer has less need to generate cash from its non-cash working 9 Erickson and Wang (1999) and Louis (2004) also examine samples of cash-financed acquisitions but they do not use the cash-financed acquisitions to conduct difference-in-differences tests. 10 Gong et al. (2008) report that abnormal accruals are significantly greater than zero prior to 395 stockfinanced acquisitions (see their Table 3). However, Gong et al. (2008) do not examine abnormal accruals during the post-acquisition period and they do not compare the abnormal accruals of stockfinanced acquirers to the abnormal accruals of cash-financed acquirers. 11 Ball (2013, p. 851) comments that: In study after study, an explanation of the reported results that does not involve manipulation is at least as plausible as one that does. 12 Louis (2004) finds that stock-financed acquirers have significantly lower book-to-market ratios compared with cash-financed acquirers, which suggests that stock-financed acquirers have better growth opportunities. However, Louis (2004) does not control for the book-to-market ratio in his examination of abnormal accruals (see his Table 4). Pungaliya and Vijh (2009) argue that controlling for growth opportunities overturns the conclusion that companies manage earnings upwards prior to stock-financed acquisitions. Consistent with these studies, we find that stock-financed acquirers have better growth opportunities than cash-financed acquirers and we control for this in our research design. 6

10 capital (e.g., by selling inventory or collecting accounts receivable). Therefore, a stockfinanced acquirer is less likely to reduce its non-cash working capital and is more likely to increase its non-cash working capital prior to the acquisition date. This could also explain why some studies find abnormally large working capital accruals prior to stock-financed acquisitions. There are two accruals studies that find insignificant results using larger samples than those used by Erickson and Wang (1999) and Louis (2004). Heron and Lie (2002) examine 427 stock-financed acquisitions, 342 cash-financed acquisitions, and 90 acquisitions that are financed using both stock and cash. Across the three groups, Heron and Lie (2002) find no significant association between abnormal accruals and the method of acquisition financing. 13 Pungaliya and Vijh (2009) examine 895 stock-financed acquisitions and 1,719 cash-financed acquisitions. After controlling for the effects of growth on discretionary accruals, they find no evidence of upward earnings management prior to stock-financed acquisitions. They suggest that the significant results of Erickson and Wang (1999) and Louis (2004) might be attributable to the strong growth prospects of stock-financed acquirers rather than upward earnings management. On the other hand, the insignificant results in Heron and Lie (2002) and Pungaliya and Vijh (2009) do not necessarily mean that companies do not attempt to manage earnings upwards prior to stock-financed acquisitions. First, auditors can help to ensure that any overstatements of the pre-audit earnings are detected and corrected during the audit. These audit corrections mean that the audited earnings may not be overstated even if the pre-audit earnings were overstated. We test this by directly examining audit adjustments to reported earnings. Second, abnormal accruals are rather noisy measures of earnings management 13 Heron and Lie (2002) acknowledge the limitation that their earnings management tests rely on annual data. To the extent that companies overstate their (unaudited) quarterly earnings rather than their annual earnings, the tests of Heron and Lie (2002) may lack sufficient power to detect earnings management 7

11 (e.g., Dechow et al., 1995). Therefore, the insignificant results of Heron and Lie (2002) and Pungaliya and Vijh (2009) could be attributable to low power tests (i.e., false negatives) rather than an absence of earnings management. We examine this explanation by testing whether pre-audit accruals have sufficient power to detect the earnings overstatements that are corrected by auditors. 2.2 The effects of auditing on earnings management Ours is not the first study to examine the effects of auditing on earnings management. Indeed, there is a large auditing literature that correlates proxies for earnings management with various audit characteristics, such as audit firm size (Becker et al., 1998; Francis et al., 1999; Khurana and Raman, 2004; Lennox and Pittman, 2010; Chen et al., 2011), audit office size (Francis and Yu, 2009; Francis et al., 2013), non-audit fees (Frankel et al., 2002; Ashbaugh et al., 2003; Chung and Kallapur, 2003; Ferguson et al., 2004), auditor tenure (Johnson et al., 2002; Myers et al., 2003; Chen et al., 2008; Chi et al., 2009; Davis et al., 2009), auditor industry expertise (Balsam et al., 2003; Gul et al., 2009; Reichelt and Wang 2010), and audit market concentration (Boone et al., 2012; Francis et al., 2012; Newton et al., 2013). Although many studies have tried to assess the impact of auditing on earnings management, they have been unable to provide direct evidence on how exactly auditors curb earnings management. We address this limitation by examining the audit adjustments that are booked to earnings. Another limitation is that many auditing studies rely on accruals variables to identify earnings management. This is potentially problematic because accruals contain significant biases and noise, particularly around significant transactions such as acquisitions and equity offerings (Hribar and Collins, 2002; Dechow et al., 2010; Ball, 2013). The biases increase the odds that a researcher will conclude that earnings management exists when in fact it is absent (i.e., a false positive). On the other hand, noise increases the odds 8

12 that a researcher will fail to find significant evidence of earnings management when in fact it exists (i.e., a false negative). 14 Moreover, simple cross-sectional analyses of the correlations between earnings management and audit characteristics are prone to generating false positives because audit characteristics are typically endogenous. For example, Lawrence et al. (2011) argue that the negative correlation between earnings management and audit firm size is attributable to client characteristics rather than superior audits by the Big N audit firms. They present evidence that the Big N quality differential disappears when Big N clients are matched to non-big N clients based on their observable characteristics. 15 Likewise, Minutti-Meza (2013) reports that the negative correlations between earnings management and auditor industry expertise become insignificant when clients are matched using propensity scores or simply with client size. We address these limitations in several ways. First, instead of conducting crosssectional comparisons, we employ a difference-in-differences research design that controls for time-invariant differences between the treatment group (stock-financed acquirers) and the control group (cash-financed acquirers). Second, we strengthen our inferences by focusing on a specific setting, namely stock-financed acquisitions, where managers are motivated to overstate earnings. Third, we use information on audit adjustments to identify the mechanism though which auditors help to curb earnings management. Fourth, 14 Ball (2013, p. 850) argues that false positives are common in studies that use accruals variables to measure earnings management: A powerful cocktail of authors strong priors, strong ethical and moral views, limited knowledge of the determinants of accruals in the absence of manipulation, and willingness to ignore correlated omitted variables in order to report a result, seems to have fostered a research culture that tolerates grossly inadequate research designs and publishes blatantly false positives. In my opinion it has, anyway. Of course earnings management goes on. Agency costs are positive. People have been tried and convicted. I personally have testified in several high profile cases and some of the malfeasance that took place in them was disgusting. But whether the typical research paper in this dismally unscientific literature has come even remotely close to reliably documenting earnings management is another matter entirely. 15 Contrary to Lawrence et al. (2011), DeFond et al. (2016) find that client characteristics do not fully explain the quality differential between Big N and non-big N audits. 9

13 examining audit adjustments allows us to avoid the non-accounting factors that create bias and noise in accruals. This fourth point is further explained in Section Audit adjustments and the confounding factors that afflict accruals An audit adjustment captures the difference between the pre-audit earnings reported by management to the auditor (E PRE) and the audited earnings reported to investors at the end of the audit (E AUD). The pre-audit earnings are influenced by the accounting choices of management (A MGMT), while the audited earnings are influenced by the accounting choices of both management and the audit firm (A MGMT+AUD). The pre-audit earnings and the audited earnings correspond to the same fiscal yearend. Therefore, they are equally affected by confounding factors - such as the company s growth prospects - that are not discretionary accounting choices (NA). This is important in our setting because stock-financed acquirers tend to have relatively strong growth prospects and companies with strong growth prospects tend to have larger accruals even if they are not engaged in earnings management (Pungaliya and Vijh, 2009). Our analysis of audit adjustments allows us to avoid these confounding factors. To illustrate this, we express the equations for pre-audit earnings and audited earnings as follows: E PRE = A MGMT + NA (1) E AUD = A MGMT+AUD + NA (2) where E PRE = pre-audit earnings, E AUD = audited earnings, A MGMT = management s discretionary accounting choices over pre-audit earnings; A MGMT+AUD = management s and auditor s discretionary accounting choices over audited earnings; NA = any factors that affect pre-audit and audited earnings but are not discretionary accounting choices (e.g., a company s growth prospects). 10

14 An audit adjustment to earnings is simply the difference between audited earnings and pre-audit earnings. Subtracting (1) from (2) gives the audit adjustment to earnings: E AUD - E PRE = A MGMT+AUD - A MGMT (3) Eq. (3) shows that the audit adjustment (E AUD - E PRE) depends on the accounting choices of management and the audit firm (i.e., A MGMT+AUD and A MGMT) but the adjustment variable is not affected by the NA factors because NA is differenced out when we subtract (1) from (2). This is important because NA are the non-accounting factors, such as a company s growth prospects, that can potentially confound the analyses of accruals. Therefore, examining the determinants of audit adjustments allows us to avoid such confounds. 2.4 Hypothesis development When an acquisition is financed using the acquirer s stock, the shareholders of the target company receive a specified number of the acquirer s shares in return for their shares in the target. The exchange ratio is usually agreed by the acquirer and target before any public announcement of the acquisition (Erickson and Wang, 1999). The ratio is determined by the relative stock prices of the acquirer and target (or the target s appraised price if the target is a private company). Consequently, a stock-financed acquirer has an incentive to overstate its earnings in order to boost its stock price in the period before the M&A announcement date. 16 Auditors are responsible for testing whether the financial statements are fairly presented. If an auditor finds that the acquirer s pre-audit earnings are overstated, the auditor should propose a downward adjustment to earnings. The acquirer can accept the proposed adjustment in which case earnings are adjusted downwards, or the acquirer can refuse to make the adjustment. If the acquirer refuses to make the adjustment, the auditor 16 Studies in China find significantly positive earnings response coefficients, signifying that stock prices increase when Chinese companies report higher earnings (e.g., Gul et al., 2003). 11

15 can disclose the accounting problem in the audit report. In practice, an auditor s threat to disclose the problem in the audit report is often (but not always) sufficient to persuade the client to book the proposed adjustment. We expect companies to overstate earnings prior to stock-financed acquisitions. We also expect auditors to detect and correct these earnings overstatements. We therefore hypothesize that earnings are adjusted downwards when audits take place in the period prior to the announcement of stock-financed acquisitions. H1 Auditors require larger downward adjustments to earnings prior to stock-financed acquisitions. There are several reasons why the results may not support our hypothesis. First, Ball and Shivakumar (2008) argue that managers are motivated to report conservatively prior to equity offerings, which is opposite to the traditional view that managers overstate earnings in order to boost the stock price (Teoh et al., 1998a, 1998b). Ball and Shivakumar (2008) argue that equity issuers report conservatively because the IPO prospectuses are closely scrutinized by auditors and other interested parties (e.g., potential litigants, regulators, the press, and investors). In our setting, however, auditors may not know about the future stock-financed acquisition until the point at which the M&A is publicly announced. Therefore, the financial statements of stock-financed acquirers are unlikely to receive extra scrutiny from the auditor in the period prior to the M&A announcement. 17 Second, managers who deliberately overstate earnings may try to hide the misstatements from their auditors (Botosan et al., 2016). This would make it harder for 17 The stock-financed acquirers have different characteristics than the cash-financed acquirers. Therefore, auditors may be able to anticipate the method of financing even if the acquirers do not inform their auditors. In Section 5.3, we use propensity score matching to match each stock-financed acquirer to a cash-financed acquirer and the resulting matched samples are very similar in terms of their observable characteristics. As reported in Section 5.3., our inferences for H1 are unchanged using the matched sample. This suggests that our results are not affected by the ability of auditors to anticipate the acquirer s method of financing. 12

16 auditors to detect and correct earnings overstatements prior to stock-financed acquisitions. Relatedly, an acquirer may choose not to inform its auditor about a future acquisition until the acquirer is ready to make a public announcement. One reason is that the acquirer would want as few people as possible to know about its private negotiations with the target company. Another reason is that the acquirer may not want the auditor to closely scrutinize its financial statements because close scrutiny could result in more downward adjustments to earnings and higher audit fees. 18 Finally, auditors in China face a relatively low threat of being sued (Chan et al., 2006; Chen et al., 2011). This means that auditors in China have less incentive to detect and correct earnings overstatements. 3. Research design, sample, and descriptive statistics 3.1 Difference-in-differences research design There is a downward adjustment to earnings when the audited earnings are less than the pre-audit earnings; i.e., E AUD, it < E PRE, it. Under H1, auditors require larger downward adjustments prior to stock-financed acquisitions. We test H1 by estimating the following model that explains the absolute magnitudes of downward audit adjustments to earnings: ADJ_DN it = α 0 + α 1 STOCK i + α 2 BEFORE t + α 3 STOCK i BEFORE t + CONTROLS + u (4) The dependent variable in eq. (4) is constructed using a company s pre-audit earnings (E PRE,it) and its audited earnings (E AUD,it). In particular, ADJ_DN it equals ( E AUD,it E PRE,it )/ E PRE,it when earnings are adjusted downwards (E AUD,it < E PRE,it), and zero when earnings are not adjusted downwards (E AUD,it E PRE,it). Therefore, ADJ_DN it takes positive values 18 Based on our own audit experiences, an acquirer would usually not tell its auditor about a future acquisition until the acquirer is ready to make a public announcement. Consistent with auditors being unaware of impending acquisitions, we find in Section 5.4 that auditors do not negotiate higher audit fees before the public announcement of a stock-financed acquisition. 13

17 when earnings are adjusted downwards, and zero values otherwise. We estimate eq. (4) using tobit regression because ADJ_DN it is truncated at zero. As an alternative to using the absolute value of pre-audit earnings as the denominator, we also scale by the absolute value of pre-audit total assets. The alternative dependent variable ( ADJ_DN1 it ) equals ( E AUD,it E PRE,it ) / TA PRE,it when earnings are adjusted downwards (E AUD,it < E PRE,it), and zero when earnings are not adjusted downwards (E AUD,it E PRE,it), where TA PRE,it is pre-audit total assets. Once again, ADJ_DN1 it takes positive values when earnings are adjusted downwards, and zero values otherwise. The STOCK i variable equals one if the acquisition of company i is financed using stock, and zero if it is financed using cash. 19 The BEFORE t variable equals one in the most recent fiscal year immediately before the M&A announcement date, and zero in the first fiscal year immediately after the M&A completion (or termination) date. The timing for BEFORE t is illustrated in Figure 1. For example, when the M&A is first announced on July 1, 2010, the BEFORE t variable equals one for the previous fiscal year-end (i.e., Dec 31, 2009). 20 When the M&A is completed or terminated on July 1, 2011, the BEFORE t variable equals zero for the following fiscal year-end (i.e., Dec 31, 2011). Our treatment variable is the interaction, STOCK i BEFORE t. Under H1, auditors require larger downward adjustments to earnings prior to stock-financed acquisitions. Therefore, we predict a positive coefficient on STOCK i BEFORE t in eq. (4) (i.e., α 3 > 0). [INSERT FIGURE 1 HERE] Similar to eq. (4), we also examine the upward adjustments to earnings: ADJ_UP it = β 0 + β 1 STOCK i + β 2 BEFORE t + β 3 STOCK i BEFORE t + CONTROLS + u (5) 19 All the acquisitions in our sample are financed using either cash or stock but not a combination of both stock and cash. 20 In China, every company is required to have a December 31 fiscal year-end. 14

18 The ADJ_UP it variable equals ( E AUD,it E PRE,it )/ E PRE,it when earnings are adjusted upwards (E AUD,it > E PRE,it), and zero when earnings are not adjusted upwards (E AUD,it E PRE,it). We also estimate an alternative specification where upward adjustments are scaled by the absolute value of pre-audit assets rather than the absolute value of pre-audit earnings. The ADJ_UP1 it variable equals ( E AUD,it E PRE,it )/ TA PRE,it when earnings are adjusted upwards, and zero when earnings are not adjusted upwards. We do not have a hypothesis for upward adjustments so we do not make a prediction about the coefficient on STOCK i BEFORE t in eq. (5). 3.2 Sample The data on audit adjustments starts in We require one year of data prior to the M&A announcement date when coding the pre-period (BEFORE t = 1), so our sample comprises M&A deals announced on or after January 1, The sample ends with deals announced in 2013 because we require data for the year after the M&A completion or termination date (BEFORE t = 0) and 2014 is the last year that audit adjustment data are available to us. Panel A of Table 1 shows how the sample is constructed. We begin with 2,466 M&A announcements where data are available from the CSMAR database in the year before the M&A announcement and the year after the M&A completion or termination date. 21 This yields an initial sample of 4,932 company-year observations, with two observations for each M&A deal (= 2 2,466). We lose 262 observations (131 deals) where audit adjustment data 21 We manually checked each M&A record in the CSMAR database and traced the outcome of each verified M&A deal by the end of our sample period. There are 41 terminated stock-financed acquisitions and 118 terminated cash-financed acquisitions in our sample. Our results are qualitatively unchanged if we drop the terminated deals from the sample. In untabulated tests, we find that stockfinanced acquisitions are more likely to be terminated than cash-financed acquisitions. This is consistent with stock-financed acquisitions having a higher likelihood of termination due to the risk of the acquirer s equity being overvalued. Unfortunately, the number of terminated stock-financed acquisitions is too small for us to test whether stock-financed acquisitions are more likely to be terminated when there is a downward audit adjustment to earnings. 15

19 are missing in the Ministry of Finance (MOF) database. We also drop 148 observations (74 deals) where there are inconsistencies between the CSMAR and MOF databases in the values of audited earnings (E AUD,it). 22 [INSERT TABLE 1 HERE] Panel B of Table 1 reports the number of deals by announcement year. There are 273 deals in the first year (2007) and 285 in the final year (2013). The most deals are announced in 2008 (378), while 2007 has the fewest (273). In total, there are 2,035 cash-financed acquisitions and 226 stock-financed acquisitions. Panel C shows there are 4,070 observations relating to the cash-financed acquisitions (STOCK i = 0) and 452 observations relating to the stockfinanced acquisitions (STOCK i = 1). 3.3 Descriptive statistics Table 2 presents descriptive statistics for all the variables used in our analyses. The definitions of each variable are provided in Appendix A. [INSERT TABLE 2 HERE] As a percentage of the absolute value of pre-audit earnings, we find that the mean signed audit adjustment (ADJ it) is negative (-3.4%) while the median adjustment is zero (0.0%). These statistics reflect that 47.81% of the audits in our sample are subject to downward earnings adjustments, 31.11% have no adjustment to earnings, and 21.07% are subject to upward earnings adjustments. Therefore, downward adjustments occur more than twice as often as upward adjustments which is consistent with prior research (Kinney and 22 Lennox et al. (2016) find that the inconsistencies are partly explained by the MOF personnel entering data from the parent company accounts rather than the group accounts. After taking into account the rounding differences between the CSMAR and MOF databases, we define the two databases as being inconsistent with each other when the reported difference in audited earnings is at least ±1%. 16

20 Martin, 1994; Lennox et al., 2016). Moreover, downward adjustments are typically larger than upward adjustments. For example, Table 2 shows that the tenth and ninetieth percentiles of signed adjustments (ADJ it) are -16.2% and 2.5% respectively. Not surprisingly, the adjustment magnitudes are smaller when we scale adjustments using the absolute value of pre-audit total assets (ADJ1 it). Table 2 presents similar descriptive statistics for the absolute magnitudes of downward adjustments ( ADJ_DN it and ADJ_DN1 it ) and upward adjustments ( ADJ_UP it and ADJ_UP1 it ). Table 2 also presents descriptive statistics for the accruals variables. We construct two alternative measures: 1) performance-matched total accruals (PMA), and 2) discretionary accruals (DA) estimated using the modified Jones model. 23 We expect that managers to overstate their pre-audit earnings prior to stock-financed acquisitions, so we focus on the absolute magnitudes of income-increasing pre-audit accruals (i.e., PMA_UP PRE,it and DA_UP PRE,it ). However, for the sake of completeness, we also examine the absolute magnitudes of income-decreasing pre-audit accruals ( PMA_DN PRE,it and DA_DN PRE,it ), the absolute magnitudes of income-increasing audited accruals ( PMA_UP AUD,it and DA_UP AUD,it ), and the absolute magnitudes of income-decreasing audited accruals ( PMA_DN AUD,it and DA_DN AUD,it ). Finally, Table 2 presents descriptive statistics for the control variables. We control for company size (SIZE it), the market-to-book ratio (MB it), and leverage (LEV it). We include an indicator for state-owned enterprises (SOE it) because government ownership is common in China (Wang et al., 2008). We control for corporate governance characteristics using the proportion of independent directors on the board (IN_DIR it) and the number of board members (BD_SIZE it). We control for a company s performance using annual buy-and-hold stock returns (BHRET it). We control for liquidity using the ratio of cash to total assets 23 In untabulated tests, we also examine total accruals and performance-matched discretionary accruals. Our inferences are unchanged using these alternative measures of earnings management. 17

21 (CASH it) and we also control for the company s age (AGE it). Prior research finds that the Big 10 audit firms in China supply higher quality audits (DeFond et al., 2000; Chen et al., 2001), so we also include an indicator for the Big 10 audit firms (BIG10 it). Finally, we include an indicator variable for audit firm changes (AUDCH it). 4. Main Results 4.1 Univariate results Table 3 reports the univariate difference-in-differences tests for H1, which predicts that auditors require larger downward adjustments to earnings prior to stock-financed acquisitions. The downward adjustments are scaled by the absolute value of pre-audit earnings or the absolute value of pre-audit assets ( ADJ_DN it and ADJ_DN1 it ). We find that the difference-in-differences tests are highly significant for both of the downward adjustment variables (t-stats. = 4.28, 4.24). Therefore, consistent with H1, the downward audit adjustments to earnings are significantly larger prior to the public announcement of stock-financed acquisitions. Panel B reports the univariate results for the upward adjustments to earnings ( ADJ_UP it and ADJ_UP1 it ). The difference-in-differences tests for upward adjustments are insignificant, signifying that auditors do not require larger upward adjustments prior to stock-financed acquisitions. [INSERT TABLE 3 HERE] 4.2 Multivariate results Table 4 reports the results from tobit regressions, where the dependent variables capture the downward adjustments ( ADJ_DN it and ADJ_DN1 it ) and upward adjustments ( ADJ_UP it and ADJ_UP1 it ). Cols. (1) and (2) show that the coefficients on the treatment variable, STOCK i BEFORE t, are significantly positive in the models of downward 18

22 adjustments (t-stats. = 3.039, 2.779). Therefore, consistent with H1, auditors require significantly larger downward adjustments to earnings before the public announcement of stock-financed acquisitions. [INSERT TABLE 4 HERE] Cols. (3) and (4) show that the coefficients on STOCK i BEFORE t are negative in the models of upward adjustments. This is consistent with auditors requiring smaller upward adjustments prior to the public announcement of stock-financed acquisitions. However, in contrast to the significant results for downward adjustments, the coefficients on STOCK i BEFORE t are insignificant in the models of upward adjustments (t-stats. = , ). Results for the control variables show that the downward adjustments tend to be significantly smaller when companies are larger (SIZE it) and have stronger stock return performance (BHRET it). Downward adjustments are positively related to leverage (LEV it), suggesting that companies with high leverage are more likely to overstate pre-audit earnings. Consistent with Cohen et al. (2011), we also find that auditors require larger downward adjustments when boards have a higher proportion of independent directors (IN_DIR it). The other control variables are insignificant in explaining downward adjustments. Cols. (3) and (4) find that upward adjustments are significantly smaller when companies are younger (AGE it), have lower market-to-book ratios (MB it), larger boards (BD_SIZE it), and the audit firm is newly appointed to the engagement (AUDCH it). 5. Supplementary analyses 5.1 Auditors reporting choices Although an auditor can request an adjustment, it is the company s own management that ultimately decides whether to book the adjustment. This reflects the principle that a company is responsible for preparing the financial statements, while the auditor is 19

23 responsible for issuing an opinion on the company s financial statements. When a company refuses to make an adjustment, the auditor can respond by disclosing the problem in the audit report. Therefore, to the extent that accounting problems are not fully resolved through audit adjustments, we would expect to find more problems being disclosed in audit reports prior to stock-financed acquisitions. Panel A of Table 5 shows the different types of audit opinions in our sample. There are 4,368 (96.59%) clean opinions and 154 unclean opinions (3.41%). Of the 154 unclean opinions, there are 33 unqualified opinions that are modified due to accounting issues, 27 opinions that are qualified due to accounting issues, 2 opinion disclaimers that mention accounting issues, 86 unclean opinions that are modified due to fundamental uncertainties relating to going-concern, and 6 unclean opinions that are modified due to fundamental uncertainties relating to impending lawsuits. [INSERT TABLE 5 HERE] The fundamental uncertainty opinions do not directly reference accounting issues. Nevertheless, these types of opinion could indicate a lack of accounting conservatism because auditors issue going-concern opinions to warn investors that the book values of assets (reported under the going-concern assumption) are substantially higher than their liquidation values. That is, auditors are more likely to issue going-concern modifications when companies refuse to write down the book values of assets to their liquidation values. Thus, going-concern modifications can signal a lack of accounting conservatism. On the other hand, a going-concern modification can also reflect the auditor s assessment that a company is financially distressed. Therefore, it is not necessarily the case that the goingconcern modifications signal low financial reporting quality. 20

24 With this in mind, we create two audit opinion variables: UNCLEAN it and OPINION it. The UNCLEAN it variable equals zero for the 4,368 clean opinions, and one for the 154 unclean opinions. The OPINION it variable equals zero for the 4,368 clean opinions; one for the 62 (= ) unclean opinions that directly reference accounting problems; and two for the 92 (= 86+6) unclean opinions that disclose fundamental uncertainties relating to going-concern and impending lawsuits. 24 Panel B of Table 5 shows that 16.8% of audit opinions are unclean in the year prior to stock-financed acquisitions, whereas only 7.1% are unclean in the year after stock-financed acquisitions. In the sample of cash-financed acquisitions, 2.3% of audit opinions are unclean in the pre-acquisition period and 2.7% are unclean in the post-acquisition period. The difference-in-differences is 10.1% (= (16.8% - 7.1%) (2.3% - 2.7%)) and is statistically significant at the 1% level (z-stat. = 3.05). Therefore, auditors are significantly more likely to issue unclean opinions prior to stock-financed acquisitions. Panel C of Table 5 reports results for the difference-in-differences regressions, where the dependent variables are UNCLEAN it and OPINION it. UNCLEAN it = STOCK i + 2 BEFORE t + 3 STOCK i BEFORE t + CONTROLS + u (6) OPINION it = γ 0 + γ 1 STOCK i + γ 2 BEFORE t + γ 3 STOCK i BEFORE t + CONTROLS + u (7) Eq. (6) is estimated using ordinary logit because the dependent variable is binary (UNCLEAN it = 0, 1); Eq. (7) is estimated using multinomial logit because the dependent variable takes three values (OPINION it = 0, 1, 2). Col. 1 of Table 5 presents equation (6) for the unclean opinions; Col. 2 presents the equation for accounting-related problems (i.e., where OPINION it = 1 in eq. (7)); Col. (3) presents the equation for the uncertainty-related 24 In our sample, some of the audit reports that disclose accounting-related problems also disclose fundamental uncertainties about going-concern. We assign these companies to the OPINION it = 1 group because we are primarily interested in the audit reports that disclose accounting-related problems. 21

25 opinions (i.e., where OPINION it = 2 in eq. (7)). We employ the same set of control variables as in Table 4 except that we add a control for the return on assets (ROA) because auditors are more likely to issue going-concern opinions when companies are less profitable. 25 Col. (1) finds a significant positive coefficient on STOCK i BEFORE t in the regression for unclean opinions (z-stat. = 2.110). Therefore, consistent with the univariate difference-indifferences test reported in Panel B of Table 5, we find that auditors are more likely to issue unclean opinions prior to stock-financed acquisitions. Col. (2) also finds a significant positive coefficient on STOCK i BEFORE t in the regression for the accounting-related unclean opinions (z-stat. = 2.249). Therefore, auditors are more likely to disclose accounting problems prior to stock-financed acquisitions. In contrast, Col. (3) finds an insignificant positive coefficient on STOCK i BEFORE t in the regression for the uncertainty-related unclean opinions (z-stat. = 0.636). Together, these results imply that the significant results in Col. (1) are mainly driven by the accounting-related unclean opinions rather than the uncertaintyrelated unclean opinions. Overall, these results suggest that auditors are more likely to disclose accounting problems before companies engage in stock-financed acquisitions. This also suggests that not all of the accounting problems are resolved through audit adjustments; i.e., there are unresolved accounting problems in the audited financial statements even after the audit adjustments have been booked. However, we are cautious in our conclusions as there are only 154 unclean opinions and there are only 62 opinions that directly reference accounting problems in the audited financial statements. 5.2 Accruals prior to stock-financed acquisitions 25 We do not include ROA in the models of audit adjustments because ROA is mechanically correlated with audit adjustments and both variables are a function of audited earnings. 22