Market for corporate control and demand for auditing: evidence from international M&A laws

Abstract

We investigate whether and how the market for corporate control affects the demand for audit service in a cross-country setting. In so doing, we exploit the staggered enactments of merger and acquisition (M&A) laws as an exogenous shock that substantially increases takeover pressure. We find that firms are more likely to choose Big 4 auditors in the period after the enactment of M&A laws, suggesting that the takeover pressure heightened by the passage of M&A laws increases the demand for audit verification and assurance by high-quality auditors. We also find that the enactment of M&A laws leads to greater demand for Big 4 auditors through two channels: managerial commitment to curtailing agency problems and the enhancement of board monitoring. We further show that improved auditor quality facilitates creditors’ and investors’ reliance on accounting information, as reflected in greater use of accounting-based debt covenants and enhanced earnings informativeness, respectively, in the post-enactment period. Overall our results suggest that auditors play a key role in strengthening corporate governance after the enactment of M&A laws.

1 Introduction

The market for corporate control disciplines entrenched managers and limits agency problems by increasing the likelihood of their forced turnovers when they do not serve the interests of shareholders (Manne 1965; Fama and Jensen 1983; Jensen and Ruback 1983). An integral part of this governance mechanism is the performance evaluation by shareholders and directors, which relies largely on audited financial statements (DeAngelo 1988). In addition, potential acquirers use financial statements as a key source of information to identify underperforming targets and make takeover bids (Raman et al. 2013). Despite the importance of financial reporting and auditing in the corporate control mechanism, researchers have a limited understanding of the relation between the corporate control market and external auditing. To fill this void in the literature, we investigate whether and, if so, how the market for corporate control affects the demand for audit verification and assurance services.

We argue that external takeover pressure increases the demand for high-quality auditors for two distinct but interrelated reasons. First, an active takeover market enhances managerial discipline and prompts managers to serve the interests of shareholders to mitigate the threats of losing their jobs and reputational capital. Managers facing takeover threats feel great pressure to better use corporate resources and pursue value-adding investments to fend off takeover attempts. In this circumstance, managers often increase financial leverage to constrain themselves from diverting free cash flows to inefficient capital investments for personal gain (Jensen 1988; Berger et al. 1997; Servaes and Tamayo 2014; Khurana and Wang 2019). Likewise, managers can use the appointment of high-quality auditors to commit themselves to curtailing agency problems by enhancing financial reporting quality and facilitating shareholders’ evaluation of managerial performance (managerial commitment channel). Because this commitment is costly, it can serve as a credible signal that managers voluntarily constrain managerial opportunism in financial reporting (Jensen and Meckling 1976; Datar et al. 1991; Hirshleifer and Thakor 1998). This would, in turn, increase firm value and hence decrease the likelihood of being targeted by unwanted takeover attempts (Khurana and Wang 2019; Balachandran et al. 2020).

Second, takeover threats incentivize board members to appoint high-quality auditors to better monitor incumbent managers (board monitoring channel). In the presence of takeover threats, the board and its audit committee are expected to more closely scrutinize managerial reporting opportunism to protect their positions (Mikkelson and Partch 1997; Coles and Hoi 2003; Lel and Miller 2015), maintain their reputational capital in the labor market for corporate directors (Fama and Jensen 1983; Gilson 1990), and avoid potential legal liability (Lennox and Pittman 2010). Accordingly, the board is likely to demand higher-quality auditors for effective monitoring when faced with takeover pressure (Beasley and Petroni 2001; Carcello et al. 2002; Lee et al. 2004; Chen and Zhou 2007).

To test the impact of takeover pressure on auditor choice, we exploit the staggered adoption of merger and acquisition (M&A) laws that have been enacted across countries around the world. During the period from 1991 to 2004, many countries passed M&A laws in a staggered manner to strengthen their markets for corporate control by simplifying the legislation governing M&A transactions and removing unnecessary and time-consuming approval procedures, thus enhancing the efficacy of the corporate control market (Lel and Miller 2015).¹ These M&A laws were generally intended to reduce barriers to transactions, encourage information dissemination, and increase minority investor protection (Nenova 2006; Lel and Miller 2015; Glendening et al. 2016). Exploiting country-level regulatory shocks is appealing from an empirical test standpoint because it alleviates concerns about potential endogeneity. Specifically, the staggered enactments of M&A laws across countries allow us to conduct a quasi-natural experiment in which we apply a difference-in-differences (DiD) analysis to examine the causal effect of the market for corporate control on the demand for high-quality audits.

Using a sample of firms from 32 countries for the period 1986–2009, we find that firms in treatment countries with M&A law enactments are more likely to appoint Big 4 auditors in the post-law period than in the pre-law period, compared to those in control countries with no enactment.² The probability of appointing Big 4 auditors is 5.6 percentage points higher in the post-law period than in the pre-law period, suggesting that the enactment of M&A laws leads to a significant increase in the demand for high-quality audits.³ To ensure the validity of our analysis, we check whether our results are driven by the violation of the parallel trends assumption in the pre-law period and the treatment effect heterogeneity. We find that the effect of the M&A law enactment on auditor choice is insignificant in the pre-law period (consistent with the parallel trends assumption) and becomes significant in the year of the law’s passage and thereafter. We further find that our baseline results remain robust when we apply event-study DiD estimations that control for potential biases associated with the treatment effect heterogeneity in staggered DiD estimations (Baker et al. 2022). Overall our results provide strong and reliable evidence in support of the causal effect of M&A law passage on the demand for Big 4 auditors.

Next our cross-sectional analyses reveal that firms are more likely to appoint Big 4 auditors in countries that experience more intense takeover activity. This suggests that takeover threats play a greater disciplinary role in increasing the demand for Big 4 auditors when the passage of M&A laws fosters takeovers. We also perform channel analyses to see whether the impact of M&A laws on auditor selection runs through our two predicted channels. First, we find that the enactment of M&A laws leads to a greater increase in the appointment of Big 4 auditors for firms with higher agency costs, proxied by those with multiple segments and those experiencing a larger increase in leverage ratio or a greater decline in capital investment in response to takeover threats.⁴ The finding supports the existence of a managerial commitment channel in which managers facing greater agency problems are more likely to voluntarily commit to enhancing financial reporting quality via the appointment of Big 4 auditors. Second, we find that the increased demand for Big 4 auditors is primarily driven by firms in countries with higher values on the anti-director index and the anti-self-dealing index, where board members are exposed to higher legal liabilities and thus have stronger incentives to increase their oversight (Lennox and Pittman 2010). Our results are also concentrated among firms in countries with higher auditor legal liability, where Big 4 auditors appointed by the boards are motivated to produce higher-quality audits to protect their reputational capital and mitigate litigation risk than are non-Big 4 auditors (Khurana and Raman 2004; Choi et al. 2008). Both findings are consistent with the existence of a board monitoring channel.

We further investigate how takeover pressure influences creditors’ and investors’ demands for audit verification and assurance in the context of debt contracting and equity valuation, respectively. When firms seek to borrow, lenders demand audit assurance for covenant compliance in debt contracting (Ball and Shivakumar 2008; Baylis et al. 2017). Lenders are likely to rely more on accounting-based covenants in debt contracting if borrowers appoint higher-quality auditors in response to takeover threats. Consistent with this expectation, we find that the usage of accounting-based covenants, relative to non-accounting-based covenants, increases significantly following the passage of M&A laws. We also find that the price informativeness of accounting earnings increases significantly from the pre- to post-law period, consistent with investors relying more on accounting information in their performance evaluation in the post-law period. This evidence suggests that the demand from both credit and equity investors for high-quality audits increases in the presence of heightened takeover pressure.

Finally, we explore additional consequences of takeover pressure on audit quality and audit fees. We find that the enactment of M&A laws increases audit quality in that the likelihood of erroneously issuing clean opinions decreases significantly in the post-law period. This finding suggests that takeover market governance improves auditor governance of clients’ financial reporting. Moreover, we find that audit fees in treatment countries increase less than those in control countries during our sample period surrounding the passage of M&A laws. This finding is consistent with the view that the fee-decreasing effect of improvements in the quality of (pre-audited) financial statements (supply-side effect) dominates the fee-increasing effect arising from the increased demand for audit verifications (demand-side effect). Overall our results suggest that the well-functioning corporate control market facilitates auditor governance and thus improves audit quality without substantial increases in audit fees.

This study contributes to the literature in several ways. First, it presents an important link between the market for corporate control and external auditing in the context of corporate governance mechanisms. The role of takeover pressure as a primary mechanism of external governance is well studied in the finance literature. Surprisingly, however, this literature has paid little attention to the effects of the corporate control market on the demand for and provision of audit services, even though performance evaluation based on audited financial reports is a critical part of this external governance mechanism. This study advances understanding of the relation between the market for corporate control and auditor choice by showing that the two governance mechanisms are complementary. Our results provide new insights into the interplay between these two distinct but interrelated governance mechanisms.

Second, as noted by Carcello et al. (2011), studies on the association between external governance and auditing have largely ignored the endogenous nature of governance characteristics and provided, at best, mixed results.⁵ Specifically, the association between the two is often subject to potential endogeneity, in particular, because governance characteristics and audit variables are jointly influenced by common firm-specific factors, such as firm size and operating performance (Chaney et al. 2004; Carcello et al. 2011; DeFond and Zhang 2014). To the best of our knowledge, this study is among the few, if not the first, to provide large-sample, systematic evidence that the efficacy of the corporate control market has a significant impact on auditor choice, with potential endogeneity being accounted for.

Third, our study contributes to the international accounting and auditing literature, especially the literature on the consequences of M&A laws. We show that heightened takeover pressure under the M&A laws incentivizes the management and the board to increase the demand for both high-quality assurance services and credible financial reporting. We also find that the improved reporting quality in the post-law period enables credit and equity investors to rely more on financial statement information in debt contracting and equity valuation, respectively. The above findings support the view that the passage of M&A laws boosts the demand for external audit verification and assurance. Our cross-country evidence also provides valuable insights into whether and how high-quality auditing could be achieved globally by establishing the well-functioning market for corporate control (DeFond and Francis 2005).

This paper proceeds as follows. Section 2 presents the literature review and develops the hypothesis. Section 3 describes the research design and the data used. Section 4 reports the primary results. Section 5 presents cross-sectional tests and channel analyses. Section 6 reports the results for the external demand for audit services. Section 7 details further analyses of audit quality and audit fees. The final section concludes.

2 Literature review and hypothesis development

2.1 The market for corporate control as an external governance mechanism

The market for corporate control involves transactions for shareholder control over a company and includes all mergers, acquisitions, and reorganizations (Bittlingmayer 1998). In a perfect world, corporate assets could be channeled toward their best possible use. M&A, as an external governance mechanism, facilitate this process by reallocating control over companies, mitigating agency risks, and disciplining or replacing inefficient management (Manne 1965; Jensen 1988; Scharfstein 1988).⁶ Consistent with this view, studies show that management turnover increases following completed takeover transactions, and occurs more often following acquisitions of firms that previously underperformed, relative to their industry benchmark (e.g., Grossman and Hart 1980; Scharfstein 1988; Martin and McConnell 1991; Mikkelson and Partch 1997; Kini et al. 2004). More recently, Lel and Miller (2015) show that, after the passage of M&A laws, underperforming firms are more likely to replace their managers and improve performance to avoid being takeover targets, and consequently management turnover becomes more sensitive to operating performance. They show that management turnover is more prevalent among poorly performing firms, especially in the presence of active corporate control markets.⁷

Although audited financial reports are a primary source of information for performance evaluation, the literature on the corporate control market has paid little attention to the role of financial reporting and auditing in this external governance mechanism. Some recent accounting studies show that a target’s accounting quality affects the acquirer’s takeover decisions (Raman et al. 2013), deal structure and completion (Marquardt and Zur 2015), and announcement returns (McNichols and Stubben 2015). Yet none of these studies examines the relationship between the market for corporate control and external auditing.

External auditors play a governance role by ameliorating managerial agency problems that are often unchallenged internally by the board or individual directors and externally by corporate control markets (Fan and Wong 2005; Choi and Wong 2007).⁸ As reputable information intermediaries, external auditors verify reported accounting numbers and assure the credibility of financial statements. The audit verification and assurance not only enhance the informativeness and credibility of financial reports but also foster monitoring effectiveness and contracting efficiency. Our study examines whether and, if so, how the market for corporate control, as an external disciplinary force, affects the role of auditor governance.

2.2 The effect of the market for corporate control on auditor choice

To investigate how the market for corporate control impacts the demand for high-quality auditors, we consider two plausible channels through which the passage of M&A laws affects auditor choice.

2.2.1 Managerial commitment channel

Takeover pressure disciplines incumbent managers by prompting them to act in the interest of shareholders (Manne 1965; Fama and Jensen 1983; Safieddine and Titman 1999). As takeover threats damp managerial agency problems, entrenched managers under takeover pressure have less incentive to obfuscate financial accounting information. Consistent with this view, Healy and Palepu (2001) find that managers who face the risk of job loss due to poor performance in a takeover market tend to improve reporting quality to reduce the likelihood of undervaluation of the firm and explain away poor earnings performance. Baber et al. (2015) further show that firms with stronger takeover defenses (i.e., weaker takeover pressure) are associated with higher incidences of restatements, suggesting the positive effect of takeover threats on external reporting quality.

Studies find that an active takeover market induces managers to commit to mitigating agency conflicts by increasing leverage and curtailing nonvalue-adding capital expenditures (Denis and Denis 1993; Berger et al. 1997; Safieddine and Titman 1999; Servaes and Tamayo 2014). Likewise, the appointment of high-quality auditors serves as a commitment mechanism to mitigate agency problems (Jensen and Meckling 1976). Specifically, managers have incentives to appoint high-quality auditors to convey a costly signal that they will voluntarily constrain managerial opportunism in financial reporting. Viewed in this manner, the passage of the M&A laws is likely to increase managers’ demand for high-quality auditors to credibly commit that they will refrain from opportunistically managing reported earnings and will provide timely, relevant, and reliable information to investors and other stakeholders.

Takeover pressure can further increase a manager’s demand for a high-quality auditor through its impact on board oversight. In their analytical model, Hirshleifer and Thakor (1998) show that it is in the management’s best interest to provide the board with more accurate information because the board would dismiss the manager more aggressively when performance measures are noisy. Consistent with this prediction, Khurana and Wang (2019) and Balachandran et al. (2020) find that the enactment of M&A laws, which strengthen board oversight on managerial performance, leads to a significant increase in accounting conservatism and earnings quality.⁹ Therefore we expect that, to mitigate the likelihood of their firms being takeover targets, managers under takeover pressure are more likely to appoint high-quality auditors.

2.2.2 Board monitoring channel

The external labor market for board members values their ability to monitor managers in an environment of high takeover threats (Coles and Hoi 2003). Specifically, board members whose firms are targeted in M&A often lose not only their seats in the targeted firms but also their seats in other firms (Lel and Miller 2015). Accordingly, corporate boards are more likely to replace poorly performing managers when takeover markets are more active (Hadlock and Lumer 1997; Mikkelson and Partch 1997; Lel and Miller 2015).

Financial reporting quality plays a crucial role in the monitoring process because the boards evaluate managerial performance by referencing audited financial statements.¹⁰ When faced with takeover threats, the boards and audit committees are more likely to appoint high-quality auditors to improve financial reporting quality. Stated differently, the demand for independent audit verification and assurance increases when the corporate control market is active; more effective audit verification allows vigilant boards to acquire credible information from managers in a timelier manner. Consistent with this perspective, studies show that stronger board oversight is associated with lower levels of opportunistic earnings management and accounting irregularities (Bedard and Johnstone 2004; Carcello et al. 2006), a higher likelihood of appointing industry specialist auditors (Beasley and Petroni 2001), and higher-quality successor auditors after auditor resignations (Lee et al. 2004). Therefore we expect that the external takeover threat encourages corporate boards and audit committees to appoint high-quality auditors to effectively monitor and discipline managers’ financial reporting.

2.2.3 Hypothesis

A large body of research shows that Big 4 auditors have large financial and reputational capital at stake in the event of audit failure; therefore they are likely to provide higher-quality audits than non-Big 4 auditors (DeAngelo 1981; Francis and Wilson 1988; DeFond and Zhang 2014; Jiang et al. 2019). Big 4 auditors are also perceived to have international reputations and to be more independent in response to heightened takeover threats (Khurana and Raman 2004). Our channels discussed in the preceding subsections suggest that, in the presence of external takeover threats, the demand for high-quality audits increases, inducing managers and boards to appoint high-quality auditors. We thus predict that firms are more likely to appoint Big 4 auditors following the passage of M&A laws. To provide systematic evidence on this under-researched issue, we propose and test the hypothesis below, stated in alternative form.

• Hypothesis: All else equal, the passage of M&A laws increases the likelihood of appointing a Big 4 auditor.

3 Research design

3.1 Empirical model

To examine the effect of the market for corporate control on auditor selection, we conduct a quasi-natural experiment by exploiting the staggered enactments of M&A laws around the world. Following Lel and Miller (2015), we view the passage of M&A laws as an exogenous shock that increases external takeover pressure. Specifically, we perform a set of DiD regressions using a sample of firms from countries that passed M&A laws during our sample period (treatment sample) and from countries that never passed M&A laws before or during the same period (control sample).

To test the impact of the enactment of M&A laws on auditor selection, we estimate a logistic regression for auditor choice (Fan and Wong 2005; Choi and Wong 2007; Guedhami et al. 2014).

$$\begin{array}{l}{BIG4}_{ijt}=\beta_0+\beta_1{POST}_{jt}\times{TREAT}_j+\beta_2{TREAT}_j+\beta_3{SIZE}_{ijt}+\beta_4{LEV}_{ijt}+\beta_5{ROA}_{ijt}+\beta_6{MB}_{ijt}\\\qquad\qquad\;+\beta_7{NGS}_{ijt}+\beta_8{NBS}_{ijt}+\beta_9{INVREC}_{ijt}+\beta_{10}{ISSUE}_{ijt}+\beta_{11}{LOSS}_{ijt}+\beta_{12}{CURR}_{ijt}\\\qquad\qquad\;+\beta_{13}{ATURN}_{ijt}+\beta_{14}{CRLST}_{ijt}+\beta_{15}GDP_{jt}+\beta_{16}REGQUAL_{jt}+\beta_{17}CH\_FR_{jt}+\beta_{18}CGRI_{jt}\\\qquad\qquad\;+\beta_{19}ASD_{j}+\beta_{20}{ACCENF}_{jt}+\beta_{21}{INF}_{jt}+Legal\;\&\;Year\;\&\;Industry\;Fixed\;Effects+\varepsilon_{ijt},\end{array}$$

(1)

where, for firm i in country j in year t, the dependent variable BIG4 is an indicator variable that equals 1 if the firm is audited by one of the Big 4 auditors (or one of the former Big 5, 6, or 8 auditors, in earlier years) and 0 otherwise. TREAT is an indicator variable that equals 1 for firms in the treatment sample and 0 for firms in the control sample. POST is an indicator variable that equals 1 for observations in the post-law period and 0 otherwise (i.e., for observations from the treatment sample in the pre-law period and for all observations from the control sample).¹¹ The key variable of interest POST × TREAT is an indicator variable that is coded 1 for firm-year observations in the post-law period (POST = 1) from countries that have enacted M&A laws (TREAT = 1) and 0 otherwise. Equation (1) does not include the standalone variable POST because it is redundant, due to the inclusion of POST × TREAT. (See Lel and Miller 2015, p. 1598.) We expect the coefficient on our DiD estimator POST × TREAT (i.e., β₁ in Eq. (1)) to be positive if the heightened takeover pressure, due to the enactment of M&A laws, induces greater demand for high-quality audit verification proxied by the use of a Big 4 auditor. Note that the coefficient on TREAT (i.e., β₂ in Eq. (1)) captures the difference in the demand for Big 4 auditors between the treatment and control samples in the pre-law period (POST = 0).

We include a series of client characteristics as control variables in Eq. (1). We control for client size (SIZE), financial leverage (LEV), profitability (ROA), and market-to-book ratio (MB). We also control for operating complexity, using the client firm’s number of geographic segments (NGS), number of business segments (NBS), and the sum of inventories and receivables divided by total assets (INVREC). Moreover, we add a long-term debt or equity issue indicator (ISSUE), a loss indicator (LOSS), current ratio (CURR), asset turnover ratio (ATURN), and a cross-listing indicator (CRLST) to the equation. Our regression model also includes country-level economic and institutional variables that may cause cross-country variations in the demand for and the provision of audit services. Specifically, we control for gross domestic product per capita (GDP), regulatory quality (REGQUAL), the change in financial reform index (CH_FR), corporate governance reforms (CGRI), the anti-self-dealing index (ASD), the strength of enforcement actions related to financial statement reporting (ACCENF), and the inflation rate of each country-year (INF).¹² Detailed variable definitions are provided in the Appendix Table 11.

We further add legal system fixed effects in Eq. (1) because legal origin may influence the demand for and the provision of audit services through its impact on investor protection, legal enforcement, and corporate governance (La Porta et al. 1997).¹³ We also include year fixed effects to capture the effect of staggered regulations other than enactments of M&A laws that were implemented in different years. We include industry fixed effects to control for potential variations in auditor selection across industries.¹⁴ To mitigate the effect of extreme values, we winsorize observations at the top and bottom 1 percent of the distribution for each continuous variable. We report z-statistics (or t-statistics) using robust standard errors corrected for heteroscedasticity and firm-level clustering throughout the paper.

3.2 Sample and data

Table 1 presents the selection and distribution of the sample used in our main analysis. Panel A of Table 1 describes the sample screening procedures. To construct the dataset, we first obtain financial data from Worldscope for firms in countries that passed takeover laws during the period from 1986 to 2009 and from those that had never passed takeover laws before or during this period.¹⁵ The sample period starts in 1986, which is five years before the earliest M&A law enactment in our sample. (South Africa, Spain, and Sweden enacted M&A laws in 1991.) It ends in 2009, which is five years after the last M&A law enactment in our sample. (Switzerland enacted M&A laws in 2004.) We do not include countries that passed M&A laws before 1991 because, for these countries, there are insufficient data available for the pre-law period, and they are thus not appropriate for the DiD analysis. The initial sample consists of 202,761 firm-year observations. We exclude 31,054 observations for firms from financial industries, leaving 171,707 firm-year observations. We also exclude observations with missing data for auditor identity and other variables required in Eq. (1). This screening process provides our main sample of 117,369 firm-year observations from 32 countries.¹⁶

Table 1 Sample distribution

Panel B of Table 1 reports the number of observations, the M&A-law status including enactment year, and the name of the M&A law for each of our sample countries. We obtain information on takeover laws around the world from Lel and Miller (2015), Nenova (2006), and other sources. Among the 32 sample countries, 16 treatment countries enacted takeover laws during our sample period (N = 43,945) and 16 control countries never passed takeover laws before or during the same period (N = 73,424).¹⁷ Japan has the largest number of observations (N = 40,550), and China and Taiwan follow Japan in terms of sample size (N = 9,454 and 7,478, respectively).

Table 2 reports descriptive statistics for our research variables, beginning with summary statistics for variables used in the auditor selection model. The mean of BIG4 is 0.504, indicating that 50.4 percent of observations are audited by one of the Big 4 auditors. We find that the mean of POST × TREAT is 0.307, suggesting that 30.7 percent of the observations in our full sample are subject to M&A laws. The mean of TREAT is 0.374, indicating that 37.4 percent of our firm-year observations are retrieved from firms in countries that enacted M&A laws during the sample period. Table 2 also presents descriptive statistics for country-level control variables and variables used in additional analyses.¹⁸

Table 2 Descriptive statistics

4 Baseline results

4.1 The effects of M&A laws on auditor selection

To formally test whether the enactment of M&A laws affects auditor selection, we estimate the DiD regression in Eq. (1) and report the results in Table 3. Column (1) shows the results using the full sample from 1986 to 2009. We find that the coefficient on POST × TREAT is positive and significant (0.225, z-statistic = 3.443), indicating that the demand for Big 4 auditors increases following the enactment of M&A laws.¹⁹ Also, the estimated coefficient of 0.225 is economically significant. For average firms in treatment countries (i.e., those that enacted M&A laws), the predicted probability of hiring a Big 4 auditor is 5.6 percentage points higher in the post-law period than in the pre-law period.²⁰ The coefficient on TREAT is also positive and significant, implying that there is a significant difference between treatment and control countries in terms of the demand for Big 4 auditors in the pre-law period (POST = 0). The full sample results in Column (1) indicate that the enactment of M&A laws leads to a significant increase in the demand for Big 4 auditors; this change in demand holds, even after controlling for differences between the treatment and control samples in the pre-law period.²¹

Table 3 The effect of M&A laws on auditor selection

Columns (2) to (4) present the regression results for Eq. (1) using alternative samples. In Column (2), we use the same sample period (1986–2009), but we further require at least one observation in both the pre- and post-law periods for each treatment firm. The coefficient on POST × TREAT remains positive and significant (0.120, z-statistic = 2.046). In Columns (3) and (4), we shorten the sample period to mitigate the influence of confounding events on our results; to this end, we restrict the sample period to 1988 − 2007, starting three years before the first enactment year of treatment countries (i.e., 1991) and ending three years after the last enactment year (i.e., 2004). Similar to Column (2), Column (4) further requires at least one observation in both the pre- and post-law periods for each treatment firm. The results in Columns (3) and (4) are consistent with those in the previous columns. In summary, our findings in Table 3 support the hypothesis that the demand for Big 4 auditors increases in treatment countries after the passage of M&A laws and to a greater extent, compared to the corresponding changes in the demand for Big 4 auditors in control countries. This evidence is consistent with the view that takeover pressure encourages managers and boards to increase their demand for high-quality auditors.²²

4.2 Parallel trends assumption and treatment effect heterogeneity

In the preceding analyses, we employ standard DiD regressions to test the impact of external takeover threats on auditor selection. To further examine whether the standard DiD tests capture the causal effects of M&A laws correctly, we employ several alternative specifications, including parallel trends analyses and event-study DiD estimations.

First, we examine whether our DiD analysis satisfies the parallel trends assumption. A critical assumption underlying the standard DiD analysis is that the treatment and control samples have parallel trends in the pre-event period. To evaluate whether this assumption is satisfied, we examine the dynamic effects of the M&A laws’ passage on auditor selection (Bertrand and Mullainathan 2003; Kim et al. 2019a; Cannon et al. 2020). Specifically, we replace POST in the preceding regressions with three indicator variables—YEAR_(t−1), YEAR_(t), and YEAR_(≥t+1)—that indicate one year before the enactment of M&A laws, the year of enactment, and one or more years after the enactment, respectively; we add the interaction terms of the three indicator variables with TREAT to Eq. (1).²³

Table 4 presents the results for the parallel trends test.²⁴ As shown in Column (1), we find that the coefficient on TREAT × YEAR_(t−1) is statistically insignificant, whereas those on TREAT × YEAR_(t) and TREAT × YEAR_(≥t+1) are positive and significant. This finding is consistent with the view that the demand for high-quality auditors increases in the post-law period but not in the pre-law period. We find similar results when we expand the model by adding TREAT × YEAR_(t−2) and replacing TREAT × YEAR_(≥t+1) with TREAT × YEAR_(t+1) and TREAT × YEAR_(≥t+2). As reported in Column (2), while the coefficients on TREAT × YEAR_(t−2) and TREAT × YEAR_(t−1) are insignificant, those on TREAT × YEAR_(t), TREAT × YEAR_(t+1), and TREAT × YEAR_(≥t+2) are all positive and significant. The results suggest that our baseline findings are unlikely to be driven by the violation of the parallel trends assumption.

Table 4 Parallel trends analysis

Second, we conduct the event-study DiD estimation to address the potential bias in our results from the staggered DiD regressions. According to Baker et al. (2022), staggered DiD analyses can lead to biased results in the presence of treatment effect heterogeneity; the bias may arise when already treated units act as effective comparison units and treatment effects vary across units and over time.²⁵ We expect our results to be less affected by this bias because (1) our sample excludes observations from countries that passed M&A laws before the beginning of our sample period and (2) our control sample includes many observations from countries that never passed M&A laws by the end of our sample period (63% of the full sample). Nevertheless, we further check the influence of potential bias by employing Callaway and Sant’Anna’s (2021) estimation suggested by Baker et al. (2022). For this event-study DiD estimation, we construct two types of control groups. The first control group consists of firms in countries that never passed M&A laws during our sample period (i.e., never-treated firms), while the second control group consists of firms in countries that never passed M&A laws during our sample period and those that did not pass M&A laws until year t (i.e., never-treated and not-yet-treated firms).

Figure 1 presents the results from the event-study DiD estimation for the period of [t − 5, t + 5], where t is the enactment year of the M&A laws. This method estimates each cohort-time-specific treatment effect and aggregates them to measure the average treatment effect on the treated (ATT). Our sample has eight cohorts based on eight unique enactment years of M&A laws (i.e., 1991, 1992, 1997, 1998, 2000, 2001, 2002, and 2004). Panel A illustrates the results when we use the never-treated firms as a control group. While the ATT for the pre-law period is − 0.002 with a z-statistic of − 0.42, the corresponding ATT for the post-law period is 0.037 with a z-statistic of 2.86. The difference in ATT between the pre- and post-law periods is significant at the 1% level (Chi-square = 7.99). The results indicate that our main finding of a significant increase in the demand for Big 4 auditors after the passage of M&A laws is robust to this alternative estimation. We find similar results when we use the never-treated and not-yet-treated firms as a control group. As depicted in Panel B, while the pre-law period ATT is − 0.007 with a z-statistic of − 0.77, the post-law period ATT is 0.033 with a z-statistic of 2.60. The difference in ATT between the pre- and post-law periods is significant at the 1% level (Chi-square = 7.31). The results in Fig. 1 reveal that our main findings remain robust, even after the treatment effect heterogeneity is accounted for.²⁶

Fig. 1

Event-study DiD estimators. This figure shows the average treatment effect on the treated (ATT) estimated from the Callaway and Sant’Anna (2021) method. Panel A uses never-treated firms as a control group, while Panel B uses never-treated and not-yet-treated firms as a control group. ^***, ^**, and ^* indicate statistical significance at the 1%, 5%, and 10% levels, respectively (two tailed)

5 Cross-sectional tests and channel analyses

5.1 Cross-sectional tests

The preceding analyses suggest that the takeover pressure heightened by the passage of M&A laws increases the demand for high-quality audits by Big 4 auditors. To provide the context for our main results, we examine whether the effects of M&A law enactment are associated with country-level takeover intensity. We conjecture that the positive effects of M&A law enactment on the demand for Big 4 auditors are more pronounced when firms operate in countries with takeover activities that are more intense, and thus where takeover threats are more credible. We measure the intensity of takeovers for each country-year using the number of M&A deals (NDEALS) and the dollar amount of M&A deals ($DEALS) in the previous year.²⁷ We retrieve the M&A data from the Thomson SDC Platinum database.

Table 5 presents the results for subsample regressions. We split the full sample using the sample median of NDEALS in Columns (1) and (2) and by the sample median of $DEALS in Columns (3) and (4). Consistent with our expectation, we find that the positive effects of M&A law enactment on the demand for Big 4 auditors are concentrated in the subsamples exposed to more intense M&A activities. The coefficient on POST × TREAT is positive and highly significant in the subsamples with a large number and dollar amount of M&A deals in Columns (1) and (3), respectively. In contrast, the corresponding coefficient is statistically insignificant in the subsamples with a small number and dollar amount of M&A deals in Columns (2) and (4), respectively.²⁸ As shown in the bottom row of Table 5, we further find that the differences in the magnitude of the coefficients on POST × TREAT between each pair of subsamples are both statistically significant at the 10% level or less. Collectively, the results suggest that the impact of the enactment of M&A laws on the demand for Big 4 auditors is stronger when firms face greater takeover pressure.

Table 5 Cross-sectional analyses for the effect of M&A laws on auditor selection

5.2 Channel analyses

In this subsection, we examine two potential channels through which the enactment of M&A laws may affect auditor selection. We expect that the effects of M&A laws on auditor selection are more pronounced when managers have greater incentives to reduce agency costs (managerial commitment channel) and when the board has stronger incentives to monitor managers in the presence of a takeover threat (board monitoring channel).

First, we examine the managerial commitment channel through which the passage of M&A laws increases the demand for Big 4 auditors. We expect that managers under takeover pressure have greater incentives to appoint high-quality auditors as a commitment to high-quality financial reporting when their firms have more severe agency problems. To test this expectation, we measure the extent of agency conflicts using (1) the number of business segments in the previous year (LNBS), (2) the change in leverage (∆Lev), and (3) the change in investment (∆Investment). Firms with larger LNBS tend to bear greater agency costs because corporate outsiders have greater difficulty in monitoring managers and evaluating firm performance (Berger and Hann 2003). In addition, managers who face takeover threats have stronger incentives to increase financial leverage and cut inefficient capital investments to reduce agency conflicts and to mitigate the chances of hostile takeovers (Jensen 1988; Berger et al. 1997; Servaes and Tamayo 2014). Thus the increase in leverage and the reduction of investments in response to the passage of M&A laws reflect managerial efforts to address severe agency problems (Khurana and Wang 2019).

Panel A of Table 6 presents the results for the managerial commitment channel. Columns (1) and (2) report the regression results for the subsamples partitioned by the sample median of LNBS. The coefficient on POST × TREAT is positive and significant for the subsample with the above-median LNBS (Column (1)) but insignificant for the subsample with the below-median LNBS (Column (2)). Columns (3) − (4) and (5) − (6) present the regression results for the subsamples partitioned by the sample medians of ∆Lev and ∆Investment, respectively. The coefficients on POST × TREAT are positive and significant in the subsamples with the above-median ∆Lev and the below-median ∆Investment (Columns (3) and (5), respectively) but insignificant in the other subsamples (Columns (4) and (6), respectively). As shown at the bottom of Panel A, the differences in the coefficients on POST × TREAT between each pair of subsamples are significant at the 5% level or less for all three pairs. The results indicate that managers facing greater agency problems are more likely to increase their demand for Big 4 auditors as a commitment to higher-quality financial reporting. Overall the results in Panel A of Table 6 are consistent with the prediction for the managerial commitment channel.

Table 6 Channel analyses for the effect of M&A laws on auditor selection

Second, we investigate the board monitoring channel through which the M&A law enactment increases the demand for Big 4 auditors. Strong legal protection for shareholders increases potential litigation risk for board members and hence incentivizes them to scrutinize managers more closely in response to increased takeover threats. Also, the board’s monitoring via the appointment of high-quality auditors becomes more effective when auditors have stronger incentives to improve the quality of their assurance services to reduce their exposure to litigation risk (Khurana and Raman 2004). Therefore we predict that the disciplinary effect of takeover pressure on increasing the board’s demand for Big 4 auditors is more pronounced in countries with stronger investor protection and countries with higher auditor legal liability. To test this conjecture, we employ three country-level factors that strengthen the efficacy of board monitoring: (1) the anti-director index (ANTIDIR), (2) the anti-self-dealing index (ASD), and (3) the auditor legal liability index (AUDLIAB).²⁹ We obtain the first two indexes from Djankov et al. (2008) and the last one from La Porta et al. (2006).

Panel B of Table 6 shows the results for the board monitoring channel. Columns (1) to (6) report the regression results for subsamples partitioned by the country-level medians of ANTIDIR, ASD, and AUDLIAB, respectively. We find that the coefficients on POST × TREAT are positive and significant in the subsamples with high values on the anti-director index, anti-self-dealing index, and auditor legal liability index in Columns (1), (3), and (5), respectively. In contrast, the corresponding coefficients for the subsamples with low values on these indexes are positive and significant only in Column (2) and nonsignificant in Columns (4) and (6). More importantly, the test statistics at the bottom of Panel B reveal that the differences in the coefficients on POST × TREAT between each pair of subsamples are consistently significant at the 10% level or less for all three pairs. This finding suggests that the heightened takeover threat in the post-law period increases the demand for Big 4 auditor verification mainly in countries with higher (i.e., above-median) investor legal protection and auditor legal liability.³⁰ These results support the prediction of the board monitoring channel and suggest that the M&A law passage strengthens the board’s incentive to monitor managerial reporting, which in turn boosts its demand for high-quality auditor verification and assurance in the post-law period.

6 M&A laws and the external demand for high-quality audits

The results in the preceding sections suggest that corporate insiders (i.e., managers and boards) increase their demand for high-quality audits following the enactment of M&A laws. In this section, we turn our attention to whether corporate outsiders (i.e., creditors and investors) change their demand for external audits in response to the passage of M&A laws.

6.1 Creditor demand for high-quality audits: Debt covenant analysis

Creditors have strong incentives to monitor borrowers under takeover pressure because borrowing firms tend to increase financial leverage as a defense strategy in an active takeover market (Berger et al. 1997; Safieddine and Titman 1999; Servaes and Tamayo 2014). As discussed in Section 2, takeover pressure also plays a governance role in enhancing the quality of financial reporting (e.g., Khurana and Wang 2019; Balachandran et al. 2020).³¹ If the external governance from takeover markets improves reporting quality—for example, by reducing the likelihood that firms misstate their accounting reports—creditors are likely to rely more on accounting information in screening and monitoring borrowers. The greater use of accounting numbers in debt contracts can further induce creditors to demand higher-quality audits (Baylis et al. 2017). Therefore, to the extent that the external governance from the market for corporate control increases financial reporting quality, one can predict that creditors are more likely to use accounting-based covenants in debt contracting than non-accounting covenants. To test this prediction, we estimate a logistic regression model.

$$\begin{array}{c}{ACOV}_{ijt}(or\;{NACOV}_{ijt})=\beta_0+\beta_1{POST}_{jt}\times{TREAT}_j+\beta_2{TREAT}_j+\beta_3{SIZE}_{ijt}+\beta_4{LEV}_{ijt}+\beta_5{ROA}_{ijt}\\\qquad\qquad\;+\beta_6{MB}_{ijt}+\beta_7{PPE}_{ijt}+\beta_8{DB\_AMT}_{ijt}+\beta_9{DB\_MAT}_{ijt}\\\qquad+Legal\;\&\;Year\;\&\;Industry\;Fixed\;Effects+\varepsilon_{ijt},\end{array}$$

(2)

where, for firm i in country j in year t, the dependent variable is either the use of accounting covenants (ACOV) or the use of non-accounting covenants (NACOV). We define ACOV (NACOV) as an indicator variable that equals 1 if a debt contract contains at least one accounting (non-accounting) covenant and 0 otherwise.³² The proportion of tangible assets (PPE) is defined as the property, plant, and equipment scaled by total assets. The debt amount (DB_AMT) is measured by the natural logarithm of the amount of debt in millions of US dollars. The debt maturity (DB_MAT) is defined as the natural logarithm of debt maturity measured in months. All other variables are as defined earlier. We obtain loan data from Dealscan and bond data from Thomson One, and we merge these two datasets with our main dataset using all available company identifiers.³³ Following Ball et al. (2015), we match debt-level variables to firm-level variables in the year immediately before the debt issuance. We require that all data for debt issuance date, debt amount, and debt maturity be available from the data sources identified above.³⁴

Table 7 presents the results for the effects of M&A laws on the use of debt covenants. The first two columns show the results for the sample of bank loans. In Column (1), where the dependent variable is ACOV, the coefficient on POST × TREAT is positive and highly significant. In contrast, in Column (2), where the dependent variable is NACOV, the corresponding coefficient is not significant. The results indicate that banks are more likely to impose accounting covenants on borrowers in the post-law period than in the pre-law period, but they do not change their use of non-accounting covenants over the same period. Moreover, the coefficient on POST × TREAT is larger in magnitude for the ACOV regression in Column (1) (1.509, z-statistic = 7.126) than for the NACOV regression in Column (2) (0.306, z-statistic = 1.039). As shown at the bottom of the table, this difference in coefficient magnitude is highly significant (Chi-square = 11.01; p < 0.001). Our results are similar when we use the sample of public bonds, as shown in Columns (3) and (4); while the coefficients on ACOV and NACOV are both positive and significant, the difference in coefficient magnitude between the two regressions is highly significant (Chi-square = 139.48; p < 0.001). Taken together, the results in Table 7 show that creditors are more likely to use accounting-based covenants in their debt contracting after the passage of M&A laws, suggesting that their demand for external audits increases from the pre- to the post-law period (Baylis et al. 2017).³⁵

Table 7 The effect of M&A laws on debt covenants

6.2 Investor demand for high quality audits: Earnings informativeness analysis

Active takeover markets create strong incentives for investors to use financial statement information for performance evaluation. In particular, potential acquirers rely heavily on financial statements as a key source of information to identify underperforming firms and estimate the expected value creation in takeovers (Raman et al. 2013). Moreover, cross-border M&A markets further increase the demand for high-quality public disclosures, particularly because foreign acquirers have, at best, limited access to private information about potential targets.

Takeover pressure and investor monitoring associated therewith create investors’ demand for high-quality financial reporting and thus improves external reporting quality. In this environment, the M&A law passage, which heightens takeover pressure and thus enhances financial reporting quality, would increase investors’ reliance on accounting information in their valuation decisions. We therefore expect to observe stock returns more closely related to accounting earnings in the post-law period than in the pre-law period. To test this expectation, our analysis focuses on whether the price informativeness of accounting earnings increases following the passage of M&A laws. To this end, we estimate a return-earnings regression model.

$$\begin{array}{l}{RET}_{ijt}=\beta_0+\beta_1{POST}_{jt}\times{TREAT}_j+\beta_2{TREAT}_j+\beta_3E_{ijt}\times{POST}_{jt}\times{TREAT}_{j}+\beta_4E_{ijt}\times{TREAT}_j\\\qquad\qquad+\beta_5E_{ijt}+Legal\;\&\;Year\;\&\;Industry\;Fixed\;Effects+\varepsilon_{ijt},\end{array}$$

(3)

where, for firm i in country j in year t, the dependent variable, RET, is the stock return for firm i minus the value-weighted market return over the fiscal year and E is earnings per share scaled by the beginning-of-year share price. All other variables are as previously defined.

Table 8 reports the estimated results for Eq. (3). As shown in Column (1), the coefficient on E × POST × TREAT is positive and highly significant (0.222, t-statistic = 7.606), indicating that the informativeness of earnings increases following the passage of M&A laws. To further investigate the positive impact of takeover pressure on earnings informativeness, we split the full sample into two subsamples based on the median value of accruals (TACC). Columns (2) and (3) present the results for the subsample regressions. We find that the increase in the informativeness of earnings from the pre- to the post-law period (as reflected in the positive coefficient on E × POST × TREAT) is significantly greater for the subsample with high accruals (Column (2)) than for the subsample with low accruals (Column (3)). This finding suggests that accruals have higher information quality after the passage of M&A laws; thus investors incorporate more accruals information into their valuation decisions in the post-law period. Given that accruals require higher-level verification than operating cash flows, the evidence is consistent with the view that investors’ demand for external audit verification increases in the post-law period and, to a greater extent, for firms with higher accruals (Ball and Shivakumar 2008).³⁶

Table 8 The effect of M&A laws on the informativeness of earnings

Collectively, the evidence from both the debt covenant and earnings informativeness tests supports the view that outside financial statement users rely more on reported accounting numbers after the passage of M&A laws.³⁷ This greater reliance boosts their demand for high-quality audits (Ball and Shivakumar 2008; Baylis et al. 2017).

7 Further analyses

7.1 The effect of M&A laws on audit quality

In this subsection, we investigate the impact of M&A law enactment on the quality of audit services as a further consequence of heightened takeover pressure. Our preceding analyses indicate that both corporate insiders (managers and boards) and outsiders (creditors and investors) increase their demand for high-quality audits following the passage of M&A laws. To the extent that the enhanced demand induces auditors to provide better assurance services, one can expect that the enactment of M&A laws would lead to an increase in audit quality.

To test this expectation, we measure audit quality using the likelihood that auditors issue erroneous unqualified audit opinions. Specifically, we first run a logistic regression of the issuance of a qualified audit opinion (QAO) on its determinants and then estimate the predicted value of QAO. The logistic regression includes all explanatory variables used in Eq. (1), a Big 4 indicator (BIG4), the issuance of a qualified audit opinion in the previous year (LQAO), and firm age (AGE). Then we define an indicator variable Erroneous_Opinion that equals 1 for the client-years that fall above the 95^th percentile of the predicted value of QAO but receive an unqualified opinion and 0 otherwise. Erroneous_Opinion captures potential audit failure in which an auditor issues a clean opinion for a client with high misstatement risk (i.e., type II error).³⁸

Table 9 presents the results of our logistic regressions with Erroneous_Opinion as the dependent variable. We find that the coefficient on POST × TREAT is negative and significant in the full sample (Column (1)). Additionally, the corresponding coefficient is significantly more negative in countries with high auditor legal liability (Column (2)) than in those with low auditor legal liability (Column (3)). The results indicate that the enactment of M&A laws curtails the likelihood that auditors issue unqualified audit opinions for clients with high misstatement risk, and the effect is more pronounced when auditors face higher litigation risk. This finding aligns with the view that the heightened takeover pressure induces auditors to improve their governance role beyond its impact on auditor selection.

Table 9 The effect of M&A laws on audit quality

7.2 The effect of M&A laws on audit fees

In this subsection, we test whether the passage of M&A laws affects the pricing of audit services and, if so, how. The results in the previous sections suggest that takeover pressure heightened by the enactment of M&A laws is likely to influence audit fees in two opposing ways. On one hand, increased takeover pressure disciplines self-serving managers and improves financial reporting (Khurana and Wang 2019; Balachandran et al. 2020), thereby reducing auditor litigation risk embedded in the pricing of audit services (i.e., supply-side effect). On the other hand, the passage of M&A laws can increase financial reporting quality and thus boost the demand for external audit verification; specifically, the improved reporting quality induces the board of directors, creditors, investors, and other stakeholders to rely more on accounting information when making their contracting and valuation decisions. This in turn increases auditor litigation risk and thus audit fees (i.e., demand-side effect). Given the two countervailing effects, the (net) effect of M&A law enactment on audit fees depends on which effect dominates.

We test this by estimating an ordinary least squares (OLS) regression (Simunic 1980; Choi et al. 2008; Kim et al. 2012).

$$\begin{array}{c}AUDFEEijt=\beta_0+\beta_1{POST}_{jt}\times{TREAT}_j+\beta_2{TREAT}_j+\beta_3{SIZE}_{ijt}+\beta_4{LEV}_{ijt}+\beta_5{ROA}_{ijt}\\\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;+\beta_6{MB}_{ijt}+\beta_7{BIG4}_{ijt}+\beta_8{NGS}_{ijt}+\beta_9{NBS}_{ijt}+\beta_{10}{INVREC}_{ijt}+\beta_{11}{ISSUE}_{ijt}\\\begin{array}{c}\;\;\;\;+\beta_{12}{LOSS}_{ijt}+\beta_{13}{CURR}_{ijt}+\beta_{14}{ATURN}_{ijt}+\beta_{15}{CRLST}_{ijt}\\\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;+\beta_{16}{GDP}_{jt}+\beta_{17}{REGQUAL}_{jt}+\beta_{18}{CH\_FR}_{jt}+\beta_{19}{CGRI}_{jt}+\beta_{20}{ASD}_j\\\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;+\beta_{21}{ACCENF}_{jt}+\beta_{22}{INF}_{jt}+Legal\;\&\;Year\;\&\;Industry\;Fixed\;Effects+\varepsilon_{ijt},\end{array}\end{array}$$

(4)

where, for firm i in country j in year t, the dependent variable AUDFEE is measured by the natural logarithm of audit fees in US dollars.³⁹ This regression equation includes all explanatory variables in Eq. (1) and a Big 4 auditor indicator (BIG4) as an additional control variable. We expect the coefficient on POST × TREAT (i.e., β₁) to be negative (positive) if the supply-side (demand-side) effect dominates.

Panels A and B of Table 10 present the selection process and the distribution of the audit-fee sample by country, respectively. The number of sample countries is reduced to 15 (six treatment countries and nine control countries) because few firms in other countries disclose audit fees. The sample period starts from 1992, which is five years before the earliest M&A law enactment in the audit-fee sample (i.e., India and Ireland enacted M&A laws in 1997), and ends in 2009, which is five years after the last M&A law enactment in the audit-fee sample (i.e., Switzerland enacted M&A laws in 2004).

Table 10 The effect of M&A laws on audit fees

Panel C of Table 10 presents the regression results for the effects of M&A law enactment on audit fees. Column (1) shows that the coefficient on POST × TREAT is negative and significant (− 0.209, t-statistic =  − 4.226). Because the average audit fees exhibit an increasing trend in both treatment and control countries during our sample period (in untabulated analyses), this result indicates that audit fees increase significantly less in treatment countries after the passage of M&A laws, compared to the corresponding increase in audit fees in control countries.⁴⁰ We obtain consistent results in Columns (2) to (4), where we require at least one observation in both the pre- and post-law periods for each treatment firm, use a shortened sample period, 1994–2007, or both. Collectively, the results in Panel C show that the enactment of M&A laws is associated with a smaller increase in audit fees, suggesting that the disciplinary effect of takeover pressure on mitigating auditor litigation risk (i.e., supply-side effect) dominates its effect on increasing auditor litigation risk through the increased demand for external audit verification (i.e., demand-side effect).⁴¹

8 Conclusion

The market for corporate control mitigates agency problems by disciplining the management and the board. Exploiting the staggered enactments of M&A laws as a natural experiment, we find that firms are more likely to appoint Big 4 auditors in response to enhanced takeover pressure. We also find that the positive effect of takeover pressure on the appointment of Big 4 auditors is more pronounced in countries that experience more and larger takeover transactions. Our channel analyses reveal that the impact of M&A laws on auditor selection runs through managerial commitment and board monitoring channels.

Next we find that the enactment of M&A laws leads creditors and investors to rely more on accounting information for their contracting and valuation decisions. This evidence is consistent with the greater demand for high-quality audit verification in the presence of heightened takeover pressure (Ball and Shivakumar 2008; Baylis et al. 2017). In line with the increased demand for audit verification, we also find that the external governance from takeover markets improves audit quality. Finally, we find that audit fees increase less steeply after the passage of M&A laws than before, suggesting that an effective market for corporate control improves auditor governance without substantial increases in audit fees.

Our paper is among the very few, if not the first, to study the market for corporate control from an auditing perspective. Our results provide novel evidence on the significant effect of external governance from active takeover markets on auditor selection, audit quality, and audit fees. However, these findings should be interpreted cautiously because our analyses are subject to caveats. First, we cannot rule out the possibility that our sample countries could have endogenously passed M&A laws after weighing the pros and cons of the laws. Second, although we perform a battery of robustness tests, our results might be influenced by country-level, time-varying omitted variables. Third, although we show that the effect of takeover pressure on auditor selection runs through managerial commitment and board monitoring channels, our evidence is far from conclusive, due to limitations on our measurement of channel variables. Future research may further explore the channels through which takeover threats affect audit variables.

Appendix

Table 11 Variable Definition