Market reactions to audit committee director's gender: Evidence from US-traded foreign firms

This study investigates whether the appointment of a female to the audit committee of a foreign issuer in the US is positively associated with subsequent market price reaction. We hypothesize that female members on the audit committee can strengthen corporate governance by their conservative and ethical qualities. Accordingly, such appointments deliver a positive message to capital market participants. In order to observe the impact of audit committee gender diversity on foreign fi rms, we include all audit committee appointments for UStraded foreign fi rms from 2002 to 2009. We fi nd that the appointment of female audit committee members has signifi cant positive cumulative abnormal returns compared to the appointment of male audit committee members.


Introduction
The primary purpose of this study is to investigate whether the appointment of a female to the audit committee of a foreign issuer in the US is positively associated with subsequent market price reaction. Research shows that female members on the audit committee can strengthen corporate governance by their IJBF conservative and ethical qualities. Accordingly, when a fi rm announces the appointment of a female to its audit committee, it delivers a positive message to capital market participants. In order to observe the impact of audit committee gender diversity on foreign fi rms, we include all audit committee appointments for U.S.-traded foreign fi rms from 2002 to 2009 in our sample.
Over the past two decades, foreign issuers have found the US capital markets to be an important gateway for capital-raising activities. However, the attractiveness and growth of other international stock exchanges (e.g., Hong Kong, Tokyo and London) represent signifi cant challenges to the competitive position of the American markets. Indeed, the sluggish economic conditions in the country and the challenging global economic environment also present signifi cant obstacles to the competitiveness of our capital markets. 1 These conditions signal the need for the US to attract new listings by foreign fi rms while retaining the existing population of foreign listings.
In response to a series of fi nancial scandals in the early 2000s, the Sarbanes-Oxley Act (SOX) was signed into law in 2002 to restore investor confi dence in the capital markets and fi nancial reporting in general. The provisions of SOX encompass four major areas: business ethics, internal controls and procedures, governance and independence, and reporting and disclosures. Although particular sections of SOX focus on strengthening the corporate governance of public fi rms, evidence suggests that these strict requirements and onerous compliance costs present barriers to foreign fi rms since SOX does not provide them with a general exemption from its provisions. For example, Hostak, Karaoglu, Lys and Yang (2009) note that foreign fi rms, which voluntarily deregistered from exchanges, had weaker corporate governance and experienced, a signifi cantly negative stock market reaction when SOX was passed. These fi rms also suffered a signifi cant price decline when they announced their intention to deregister. Given this environment, Balakrishnan (2007) suggests that the future challenge for the SEC is to balance two competing goals which include: (1) protecting investors by setting sound standards of corporate governance, and (2) enhancing the attractiveness of the capital markets, particularly with respect to foreign issuers.
SOX includes several provisions that focus on strengthening the role of the audit committee as a critical monitor over the fi nancial reporting process. In particular, Section 301 (SOX 301) states that the audit committee ''shall be directly responsible for the appointment, compensation, and oversight'' of the auditor. Under SOX 301, audit committee members are expected to be concerned about their personal reputations and ensure that the external auditors have fulfi lled their responsibilities. Moreover, audit committee members are required to be independent and are prohibited from accepting compensatory or advisory fees from the fi rm. Eaddy et al. (2003) indicate that the audit committee requirements raise signifi cant problems for most foreign issuers because they deal with technical structures, relationships and processes rather than with ultimate results, as do most of the disclosure requirements. Again, given the lack of a general exemption for public foreign issuers, it is important to understand how the characteristics of audit committee members affect foreign fi rms.
Prior research (e.g., McMullen & Raghunandan, 1996;Klein, 2002;Bedard, Chtorou & Courteau, 2004;DeFond, Hann & Hu, 2005) has explored the following characteristics of audit committee members: independence, meeting frequency, size, and fi nancial expertise. Moreover, specifi c research concerning the effect of gender diversity of audit committee members has recently emerged. Ittonen, Miettinen and Vahamaa (2009) fi nd evidence that fi rms with female representation on the audit committee reduces the inherent risk of misstatements. Their results have implications for external auditing since they also fi nd that gender diversity is associated with lower audit fees. Huse and Solberg (2006) fi nd that female directors are better prepared for board meetings than male directors which results in improved board behaviour and effectiveness. These benefi ts should also be realized by audit committees with female membership.
Research on ethics fi nds that women are less likely to engage in unethical behaviour in the workplace to obtain fi nancial rewards (Bernardi & Arnold, 1997;Betz, O'Connell & Shepard, 1989). Ruegger and King (1992) fi nd that females are more ethical than males in their perceptions of ethical conduct in business situations. Moreover, related studies (e.g., Krishnan & Parsons, 2008) fi nd that earnings quality is positively associated with gender diversity in senior management, that females are more risk averse than men, and are less likely to engage in unethical behaviour in the workplace.
We fi nd that the appointment of female audit committee members in U.S.-traded foreign fi rms has signifi cantly positive cumulative abnormal market returns compared to the appointment of male audit committee members. Our fi ndings contribute to the extant literature in several ways. First, the results confi rm that share prices react favourably to the appointment of females to the audit committee. This reinforces the belief that gender diversity enhances the perceptions of market participants with respect to the audit committee's independence and the overall performance of the governance function. Second, this study has implications for global investors and other users of fi nancial information who are contemplating future investments in the US capital markets. If the SEC's major goal is to improve the quality of corporate fi nancial reporting, it should encourage gender diversity on audit committees.
Finally, this study contributes to the emerging literature that examines the effects of audit committee gender diversity. We provide additional evidence from the market participants' perspective that the appointment of female audit committee members contributes to shareholder value by improving the reliability of fi nancial reporting through enhanced monitoring.
The remaining sections of this paper are organized as follows. Section 2 outlines the related literature and hypotheses. Section 3 provides an explanation of the research method and data. In section 4, is provided the empirical tests and explanations of the results are presented. Section 5 is the conclusion.

Gender Research in Corporate Governance
Traditional gender studies in the accounting literature have focused on the dissatisfaction of women in the workplace and the lack of high-level female directors 2 (e.g., Pillsbury, Capozzoli and Ciampa, 1989;Neidermeyer, Tuten and Neidermeyer, 2003). Behavioural studies in business have shown that women in the accounting profession are generally more ethical than men when performing accounting tasks. For example, Schaefer and Welker (1994) indicate that CPAs disciplined for work-related ethics violations are more likely to be males who practice in small fi rms. Bernardi and Arnold (1997) fi nd that female managers are at a signifi cantly higher average level of moral development than male managers in the Big Six accounting fi rms. Cohen, Pant and Sharp (1998) indicate that women have consistently different ethical evaluations, intentions and orientation than men. More specifi cally, women consider questionable actions to be less ethical than men, and indicate that they would be less likely to engage in certain actions than men. Neidermeyer et al. (2003) suggest that women believe that the practice of low-balling audit fees is less acceptable to women than to men, and women also agree signifi cantly more often that low-balling is a violation of the "independence in appearance" clause of the AICPA Code of Professional Conduct. 3 Chung and Monroe (2001) fi nd signifi cant gender differences with respect to information processing abilities and that females exercise more accurate audit judgment and decision-making abilities than males when faced with complex accounting decision tasks. Due to these unique characteristics of women, researchers have recently begun to investigate how gender diversity infl uences the effectiveness and effi ciency of corporate governance, particularly with respect to the board of directors and audit committees. For example, Carter, D'Souza, Simkins and Simpson (2008) indicate that gender diversity has a positive effect on fi nancial performance primarily through the audit function of the board of directors. Ittonen et al. (2009) suggest that female representation on audit committees may decrease audit fees by affecting the auditor's assessment of the integrity of the fi nancial reporting process, as well as by reducing the inherent risk of misstatements. Gul, Srinidhi and Tsui (2008) fi nd that boards with female directors are more likely to demand higher monitoring in the form of more audit effort. Levi, Li and Zhang (2008) document that the bid premium over the preannouncement target share price is statistically and economically smaller when there is a larger proportion of women on the target company's board, provided that the female directors are independent appointees. Abbott, Parker and Presley (2009) study the relationship between board diversity and fi nancial restatements and fi nd that a female board presence is indicative of a more control-conscious CEO and a more vigilant board of directors. In summary, extant gender studies document the positive infl uence of female representation on the corporate governance function.

Market Reaction to the Appointment of Audit Committee Members
Investors respond to important announcements and events that are believed to impact a fi rm's value at a point in time. The appointment of a new audit committee member can be viewed as critical information to market participants. Previous studies document a signifi cant increase in the market's reaction to earnings reports subsequent to the formation of the audit committee (Wild, 1996). Davidson, Xie and Xu (2004)  These studies demonstrate the existence of information content in the capital markets for audit committee related announcements. Given this discussion and the positive infl uence of female representation on corporate governance, we expect that the appointment of a female to the audit committee provides signifi cant information content regarding: (1) enhanced oversight over the fi rm's fi nancial reporting process, (2) more effective supervision over the external auditor, and (3) expected improvements in the quality of the fi rm's fi nancial reporting to market participants.
We further hypothesize that this signal simultaneously induces favourable market price reactions as follows: Hypothesis 1a: Market reaction in response to the appointment of a female audit committee member is positive.
Hypothesis 1b: Market reaction in response to the appointment of a male audit committee member is neutral/negative.

Hypothesis 2a:
Market reaction is positively associated with the number of female audit committee members appointed.
Hypothesis 2b: Market reaction is neutrally/negatively associated with the number of male audit committee members appointed.

Model Specifi cation
To examine our hypotheses, the following OLS regression models are employed (e.g. Chen & Church, 1996;DeFond et al. 2005;Choi, Kim, Liu & Simunic, 2008): (1) Where: CAR = Cumulative abnormal market returns; CHPRICE = Changes in stock price prior to the issuance of the auditor's report; LnMVE = Natural log of the market value of equity; LEVERAGE = Total debt divided by total assets; LnREGIME = The strictness of a country's legal regime, measured by the natural log of Wingate's 1997 litigation index for each country; LnGDP = Natural log of gross domestic product (GDP) per capita (in thousands of US dollars); FDI = Foreign direct investment scaled by GDP; LnEQUITY = The importance of each country's equity market, measured by the natural log of extent to which each country's fi rms rely on equity fi nancing (the data are extracted from Choi and Wong, 2005) (1).
Our main tests consist of a traditional event study that examines: (1) three-day (−1 to +1); (2) seven-day (−3 to +3), and (3) eleven-day (-5 to +5) CAR around the appointment disclosure dates of the new audit committee members. In the OLS regression models, CAR is regressed on variables of interest representing gender diversity, such as FEMALE, FSIZE, and MSIZE. We control for the potential effects of the size of the company and total market environment by using the following three variables: CHPRICE, LnMVE, and LEVERAGE as in previous studies (Chen & Church, 1996;DeFond et al., 2005). Country-specifi c litigation and governance variables, including LnREGIME, LnGDP, FDI, LnEQUITY and DISCL, are controlled as in Choi et al. (2008). As in Choi et al. (2008), each country has one litigation index, GDP, FDI, equity index, and disclosure level index. Therefore, US-traded foreign fi rms from the same country will be assigned the same governance indices. The GDP and FDI are different by years, but the litigation index, equity index, and disclosure level index are the same across the sample years due to data availability.

Sample Selection
Panel A of Table 1 provides a summary of the sample selection process. The initial sample of 476 audit committee appointments (from 2002 to 2009) by foreign fi rms were obtained from the AuditAnalytics database. To clearly identify that the CAR is driven by male or female appointments separately, we excluded 182 companies that appointed both female and male members on the same day. After matching with the Compustat and CRSP databases, the missing fi nancial data reduces the available observations to 107 for our regression analysis. There are no delisted fi rms in our fi nal sample. Panel B shows that the selected fi rms are distributed among 14 countries/economic entities. The highest representation is by Canadian fi rms (56) followed by fi rms from the United Kingdom (20) and China (12).

Descriptive Statistics
Table 2 provides descriptive statistics on the full sample tested in this study. 4 We use Eventus to calculate the CAR for all of our 107 sample fi rms. The mean and median CAR in all three event windows are negative. Sixty-three per cent of our selected fi rms were traded on AMEX, NASDAQ or NYSE (LISTANN). Ten per cent of our sample fi rms reported the appointment of a female audit committee member (FEMALE).    Table 3 presents the CAR analysis for both male and female groups. The mean and median of CAR in event windows (-3, +3) and (-5, +5) are positive for the appointment of female audit committee members. Conversely, the mean and median of CAR are negative in all three event windows for the appointment of male audit committee members.
The difference in CAR between male and female groups is signifi cant (p<0.10, two-tailed) in window (-3, +3). Accordingly, the empirical results provide some evidence that the market reacted positively (negatively) to the appointment of female (male) committee members. Overall, however, the univariate tests provide weak evidence to support our hypotheses. Table 4 shows the correlation matrix of our variables and reports signifi cant results in window (-3, +3). Both FEMALE and FSIZE are positively correlated with CAR and signifi cant at the conventional level (p<.10, twotailed). In window (-5, +5), FSIZE is also positively correlated with CAR and signifi cant (at p<.10, two-tailed). For male appointments, MSIZE is negatively correlated with CAR (-3, +3) and signifi cant at the conventional level (p<.05, two-tailed). Accordingly, the correlation analysis provides consistent evidence to support our major hypotheses   Table 2. . Table 5 reports the OLS regression results for the primary tests. 6 Several control variables are signifi cant in window (-5, +5), including CHPRICE, FDI, DISCL, and TECH. As expected, CAR is positively associated with the fi rms' prior market returns (CHPRICE). Technology fi rms with higher industrial risks experience greater market reactions after the appointments of female directors. This is perhaps due to the positive infl uences of female representation on the fi nancial reporting process and the quality of fi nancial reporting, particularly in fi rms exposed to higher risks. The coeffi cients of FDI and DISCL are both negatively signifi cant in event window CAR (-5, +5), which suggests that a country's foreign direct investment (scaled by total GDP) and disclosure level reduce the impact of the appointment of female audit committee members on the market price of foreign fi rms. A higher disclosure level (DISCL) is found to reduce the volatility of market reactions to the announcements of female appointments. The results of our variables of interest show that the appointment of female audit committee members has information content to the stock market in two event windows after controlling for the effects of cross-country variables.

OLS Regression Results
The coeffi cient of FEMALE is positive and signifi cant (+0.050, p<.05, twotailed) in event window CAR (-3, +3) and (+0.077, p<.05, two-tailed) in event window CAR (-5, +5). These fi ndings support our major hypotheses that the appointment of female audit committee members in U.S.-traded foreign fi rms has signifi cant positive cumulative abnormal market returns compared to the appointment of male audit committee members. The coeffi cient of FSIZE is also positive and signifi cant (+0.070, p<.05, two-tailed) in event window CAR (-5, +5). The evidence shows that the greater the number of female audit committee members appointed, the greater the positive market return incurred. Conversely, the coeffi cients for MSIZE are negative and signifi cant or insignifi cant in all event windows. This indicates that the number of male audit committee members appointed has a negative or no signifi cant impact on the company's market price. This empirical evidence consistently supports our major hypotheses.

Additional Analysis: Gender Diversity and Restatements
If the presence of female audit committee members enhances audit committee effectiveness, it is reasonable to expect that such fi rms would experience fewer fi nancial restatements subsequent to the appointment of female audit committee members. We investigated the association between audit committee gender diversity and fi nancial restatements for the 294 fi rms indicated in Panel A of Table 1 by dividing our sample fi rms into: (1) female audit committee member appointment group (N=23), and (2) male audit committee member appointment group (N=271). We found that only 1 of the 23 foreign fi rms (4.3%) restated their fi nancial statements in the year subsequent to the appointment of female audit committee members. However, 23 of the 271 U.S.-traded foreign fi rms (8.4%) restated their fi nancial statements in the year subsequent to the appointment of male audit committee members. Accordingly, the percentage of restatements in the female group is signifi cantly lower than that of the male group.

Sensitivity Analysis
The market reaction of fi nancial fi rms may be different from other industries due to their unique characteristics. To control for this potential effect, an additional dummy control variable, FINANCE (fi nance fi rms, defi ned as SIC= 6000-6999), were included in the two OLS regression models. The results show that FINANCE is insignifi cant. The results of gender variables remain consistent with the results presented in Table 5.
We also designed a 1 by 1 matched sample based on the fi rms' SIC code as well as size and re-tested our mean and median differences of CAR between female and male groups in Table 3. The difference in CAR between the male and the female groups is consistently signifi cant (p<.05, two-tailed) in window (-3, +3) and (p<.10, two-tailed) in window (-5, +5).

Conclusion
Researchers have recently begun to examine the important impact of gender diversity on audit committee effectiveness (two recent studies are Huse & Solberg, 2006;Ittonen et al. 2009). This is particularly important given the enhanced role of the audit committee function under SOX. Also, the increased importance of foreign issuers to the American capital markets and concerns over competition from other exchanges (Balakrishnan, 2007) suggests that it is extremely important to understand how the characteristics of audit committees affect the quality of fi nancial reporting of foreign fi rms.
In this study, we investigated the relationship between the appointment of female and male audit committee members and related market-price reaction for foreign fi rms. We found that the appointment of female audit committee members has signifi cant positive cumulative abnormal returns compared to the appointment of male audit committee members.
Furthermore, the greater the number of female audit committee members appointed, the greater the positive market returns incurred. Our evidence reinforces the belief that gender diversity increases investors' confi dence in a fi rm's audit committee and its ability to provide effective oversight over the fi nancial reporting process as well as the external audit. These results have implications for global investors who seek to invest in the U.S. capital markets. These fi ndings also suggest that, if the SEC's major goal is to improve the quality of corporate fi nancial reporting, it should further stress the importance of female presence on the audit committee.

Limitations and Avenues for Future Research
The following limitations apply to this paper. First, this study focused on the audit committee gender diversity of foreign fi rms traded in the US. The limited sample unavoidably reduced the power of the models and the signifi cance of our variables. Second, we only examined the market reactions in three event windows following the appointment of a female audit committee member. The three event windows may not fully refl ect the overall impact of this type of event.
Future studies can explore whether audit committee gender diversity affects the fi rm's corporate governance by facilitating better earnings quality for US-traded foreign fi rms.