“CEO educational backgrounds and non-GAAP earnings disclosures”

Non-GAAP earnings have received attention recently. Existing literature suggests CEOs’ educational backgrounds affect the financial reporting quality. Thus, the paper analyzes whether the educational background of CEOs affects the disclosure of non- GAAP earnings. Using logit regression to examine the probability of non-GAAP earnings disclosures, this study finds the coefficient value of MBA is 0.4171, which suggests that CEOs with an MBA degree are more likely to disclose non-GAAP earnings than other CEOs. In addition, the moderating effect of audit committee quality on the asso- ciation between CEO educational backgrounds and non-GAAP earnings disclosures is investigated. The coefficient value of MBA×ACC_QUA is –2.809, which suggests that audit committee quality negatively moderates a positive association between MBA-holding CEOs and non-GAAP earnings disclosures. By focusing on a company’s non- GAAP earnings, this study contributes to the financial reporting literature. The results provide evidence that CEO education backgrounds and audit committee quality influ- ence firms’ non-GAAP earnings disclosures.


INTRODUCTION
Non-GAAP earnings, also known as "pro forma" or "street earnings" in the earnings news, have garnered attention (Choi et al., 2007;Black & Christensen, 2009; Guillamon-Saorin et al., 2017). Non-GAAP earnings are performance indicators disclosed voluntarily by managers. When calculating these alternative performance measures, managers exclude certain one-time or unusual charges from GAAP net income. The Securities and Exchange Commission indicates that non-GAAP reporting increases the risk of fraud. Literature investigates managers' strategies for disclosing non-GAAP earnings and the significance of non-GAAP earnings data. However, few studies examine the managerial characteristics that lead to non-GAAP earnings disclosure. Consequently, this study investigates how the educational background of CEOs influences non-GAAP earnings disclosures.

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
Research indicates that non-GAAP measures convey more relevant information than GAAP earnings ( Doyle et al. (2013) find that when balance sheet limitations are strong, managers exclude more expenses from non-GAAP earnings calculations. Christensen (2009) indicates managers exclude not only one-time expenses such as restructuring charges, but also recurring expenses such as depreciation, amortization, research and development, and stock-based compensation to achieve these strategic goals. Bhattacharya et al. (2003) find that some managers report adjusted earnings measures that more accurately reflect long-term core earnings, while others may report adjusted earnings measures that overstate operating results.
The financial executives regard non-GAAP earnings as one of the most important performance metrics disclosed to investors (Graham et al., 2005 2017) indicate that regulation has generally led to less aggressive non-GAAP reporting, but some companies still release non-GAAP earnings numbers that could be misleading even after SOX regulation. Prior study has found a drop in the incidence and magnitude of non-GAAP exclusions after the implementation of regulation (Marques, 2006;Entwistle et al., 2006). Heflin and Hsu (2008) find evidence that SOX and Regulation G decreased the frequency and scope of both special and recurring exclusions. Isidro and Marques (2013) indicate, from the perspective of corporate governance, that when board director compensation is tied to firm performance, companies may disclose non-GAAP figures, and make supplementary adjustments for recurring items to avoid reconciliations. This reporting style has been associated with opportunistic disclosures. According to Isidro and Marques (2015), countries with strong investor protection, competent law enforcement, developed financial markets, increased communication and information distribution are more likely to use non-GAAP results to meet or beat strategic performance goals.
Literature suggests that stakeholders benefit from corporate governance rules that limit opportunistic managerial activity (Ashbaugh-Skaife et al., 2006), and independent directors can limit opportunism related to non-GAAP earnings exclusions (Frankel et al., 2011). One of the responsibilities of the audit committee is to monitor a company's GAAP and non-GAAP earnings (Warner, 2006). Non-GAAP earnings are not audited and can therefore be ma-nipulated by managers (Bruce & Bradshaw, 2004;Frankel et al., 2011). Given that the audit committee is an internal control mechanism for monitoring firms' financial reporting, this study examines the moderating effect of audit committee quality on the relationship between CEO educational background and non-GAAP earnings disclosures. Audit committees with financial expertise reduce the incidence of internal control problems (Krishnan, 2005) and non-GAAP earnings exclusions (Seetharaman et al., 2014). This study therefore hypothesizes that the quality of the audit committee decreases the positive relationship between the CEO's educational background and non-GAAP earnings disclosures. Consequently, this study develops the following hypotheses: H1: Firms led by CEOs with MBAs are more likely to disclose non-GAAP earnings than are other firms.
H2: The positive relationship between CEOs with MBA degrees and non-GAAP earnings disclosures is weaker for firms with high quality audit committees. Analyst following (ANALY_FOLL) is the log of the number of analysts who follow the company as a proxy for analyst influence on non-GAAP reporting. CON_BEAT and ANALY_FOLL are expected to be positively related to non-GAAP disclosures (Isidro & Marques, 2013). Intangible assets (INTAN), calculated as the value of intangibles scaled by total assets, is anticipated to have a positive coefficient due to firms' inclination to revert the amortization of goodwill to its original value (Marques, 2006). Special item (SPECIAL) is an indication variable that is coded as 1 if the company reports special or extraordinary items or discontinues operations, and 0 otherwise. The literature indicates that firms with more special items disclose non-GAAP earnings more often (Isidro and Marques, 2013;Marques, 2006). Firm risk (RISK) is the quarterly return on assets standard deviation over the sample period and is expected to be positively related to non-GAAP disclosures (Lougee & Marquardt, 2004  The sample's industry composition is shown in Table 2, and the year distribution is shown in Table  3. Table 2 shows that business services make up the largest portion (13.75%), while Table 3 shows that the year distribution changes from the lowest level, 1.11% in 2017, to the highest level, 11.16% in 2010.  The findings of the correlations between the variables are presented in Table 4, with certain coefficients deserving special consideration. The correlation between SIZE and ANALY_FOLL is 0.7, which suggests that larger firms have larger analyst following. Overall, the correlations are low, and there are few correlations between any of the independent variables, which suggests that the regression model does not have an issue with multicollinearity. Table 5 displays the results for Hypothesis 1, the coefficient of MBA is significantly positive. The coefficient value of MBA is 0.4171, which suggests that firms managed by CEOs with an MBA are more likely than other firms to disclose non-GAAP earnings. The result shows that CEOs with an MBA are more likely than other business executives to disclose the firm's non-GAAP earnings as a strategic opportunity. The result is similar to the prior literature that managers with MBA degrees may adopt more aggressive strategies such as spending more capital investments, taking on more debt, undertaking more diversification acquisitions, and paying out fewer dividends (Bertrand & Schoar, 2003). In addition, Graham and Harvey (2001) find that CFOs with an MBA use the more sophisticated valuation methodologies than those without an MBA. Thus, Hypothesis 1 is supported. The results provide evidence that how CEO education background influences firms' non-GAAP earnings disclosures.   Note: The Pearson correlation coefficients are reported along the lower diagonal; n = 2,429; *** , ** , and * represent significance at the 1%, 5%, and 10% levels, respectively. Note: n = 2,429; *** , ** , and * indicate significance at the 1%, 5%, and 10% levels, respectively; one-tailed for all coefficients except for those without predicted signs. er-quality audit committee. Hypothesis 2 is interested in the interaction term of ACC_QUA×MBA, which measures how firms with a higher-quality audit committee are affected differently. Table 6 shows that the coefficients of the interaction term ACC_QUA×MBA is significantly negative. The coefficient value of ACC_QUA×MBA is -2.809, which suggests that audit committee quality is important for moderating the positive relationship between CEOs with MBA degrees and non-GAAP earnings disclosures. The second hypothesis is therefore supported. The result is supported by Seetharaman et al. (2014) that non-GAAP earnings exclusions are reduced by audit committees with financial expertise.

RESULTS AND DISCUSSION
The overall results show that the CEOs with MBA degrees are more likely to disclose the non-GAAP earnings than other CEOs. In addition, the audit committee quality negatively moderates the positive relationship between MBA-holding CEOs and non-GAAP earnings disclosure. The results provide implications to stakeholders that the characteristics of managers and the quality of audit committee affect the firm's financial reporting. This study is the first to examine whether CEOs' educational backgrounds affect a company's subsequent non-GAAP earnings. Thus, this study adds to the existing body of work on the topics of non-GAAP earnings reporting and the managerial characteristics. This study also contributes to the literature on the impact of audit committee quality by measuring it based on audit committee characteristics. This study addresses these issues by emphasizing the function of the audit committee in monitoring non-GAAP financial disclosures. Thus, these findings should help regulators and boards of directors evaluate audit committee qualities.

CONCLUSION
Since management knows more than outside investors about the current and predicted future performance of their companies, they may decide to provide investors with non-GAAP earnings to provide additional information regarding earnings performance. However, earlier research has investigated the managerial characteristics underlying the disclosure of non-GAAP earnings. Consequently, the primary objective of this study is to investigate the association between the educational background of CEOs and non-GAAP earnings disclosures. Given that the audit committee is an internal control mechanism Note: n = 2,429; *** , ** , and * indicate significance at the 1%, 5%, and 10% levels, respectively; one-tailed for all coefficients except for those without predicted signs.
for monitoring firms' financial reporting, this study also examines a moderating effect of audit committee quality on the relationship between CEO characteristics and non-GAAP earnings disclosures. Using logit regression to examine the probability of non-GAAP earnings disclosures, this analysis reveals that firms managed by CEOs with an MBA are more likely than other firms to disclose non-GAAP results. In addition, the quality of the audit committee decreases the positive relationship between MBA holding CEOs and non-GAAP earnings disclosures.