حفاظت از سرمایه گذار، نظریه چشم انداز، و مدیریت سود: مقایسه بین المللی صنعت بانکداری

This paper raises three issues related to the earnings management (EM) of banks across 48 countries. First, does earnings management of banks exist in all 48 countries? Second, what is the incentive of banks to manage earnings? Third, why does EM vary across countries? To answer these three questions, two thresholds (viz., a threshold of zero earnings and a threshold of zero earnings change) are employed. The answer to the first question above is that banks in more than two-thirds of the 48 countries sampled are found to have managed their earnings. With respect to the second question, prospect theory is used to provide an answer. The relationship between return and risk is positive for high earnings groups, but is negative for low earnings banks. Finally, as to the last question, stronger protection of investors and greater transparency in accounting disclosure can reduce banks’ incentives to manage earnings. Also, higher real GDP per capita decreases the degree of earnings management. It is seen that stronger enforcement of laws can counter intuitively result in stronger earnings management. However, this effect appears in low-income countries only, and not in high-income countries.

Recent allegations of accounting fraud at Enron, followed by similar allegations at WorldCom, Xerox, Royal Ahold, HealthSouth, and so on, have triggered a closer examination of the topic of earnings management. Schipper (1989) and Healy and Wahlen (1999) state that earnings management is the alteration of firms’ reported economic performance by insiders to either “mislead some stakeholders” or to “influence contractual outcomes”. For instance, insiders can use their discretion in financial reporting to overstate the true level of earnings and hide unfavorable earnings realizations (e.g., earnings losses or earnings decreases) that would prompt outsiders to take action against insiders. In the presence of extensive earnings management, financial reports inaccurately reflect firm performance and consequently weaken outsiders’ ability to govern the firm (Leuz et al., 2003). To inferentially measure the extent of earnings management, Degeorge et al. (1999) and Burgstahler and Dichev (1997) present evidence that managers of US firms use accounting discretion to avoid reporting small losses. Employing annual earnings (scaled by beginning market value) on US firms for the years 1976–1994, Burgstahler and Dichev (1997) demonstrate a relatively smoothed single-peaked, bell-shaped distribution except in the area of zero earnings. Their graph of “drop-at-zero” is reproduced in panel A of Fig. 1 where earnings slightly less than zero occur much less frequently than would be expected given the smoothness of the remainder of the distribution, and earnings slightly greater than zero occur much more frequently than would be expected. The evidence suggests that firms manage reported earnings so as to avoid losses in earnings when losses are small. Namely, while non-financial firms can hide small losses, they cannot hide big losses. Burgstahler and Dichev (1997) also find that managers of US firms use accounting discretion to avoid reporting small earnings decreases. Degeorge et al. (1999), using statistical earnings management measures, also find evidence of earnings management that exceeds each of three “thresholds”: reported positive profits, sustained recent performance, and the meeting of analysts’ expectations. Full-size image (40 K) Fig. 1. The distribution of earnings of US non-financial and financial industries. (A) Non-financial industries. Earnings: Annual net income scaled by market values at the beginning of the year. (Graphs are taken from Burgstahler and Dichev, 1997.) (B) Financial institutions. Earnings: Annual net income scaled by year-end common equity for US banks for the sample period from 1993 to 1999. Figure options While the above studies provide convincing evidence of earnings management, their samples typically exclude financial institutions and firms in other regulated industries, such as the utility industry. As to Burgstahler and Dichev’s (1997) viewpoint, for regulated firms, conflicting incentives to report lower earnings or decreases in earnings arise whenever there are economic benefits from reporting lower earnings to regulators; for financial institutions, incentives to avoid earnings decreases or losses may be (negatively) linked to (the extent of) regulatory oversight. While the banking industry also falls into this category, it often plays a crucial role in the capital market; for example, banks’ market capitalization ratios in the capital market are typically large. Aside from this, there are three more reasons why banks have special incentives to manage earnings, relative to the general industry, and thus it is worthwhile to test if earnings management exists in the banking industry.2 First, a banking system faces a potential illiquidity problem and thus is exposed to the risk of widespread bank runs, i.e., not all banking system obligations can be met if all holders of those obligations simultaneously claim what they have been promised (Diamond and Dybvig, 1983). Morgan (2002) says that: “… banks are black boxes, and the risks taken in the process of intermediation are hard to observe from outside the bank. Therefore, the opacity of banks exposes the entire financial system to bank runs, contagion, and other strains of ‘systemic’ risk.” Thus, in order to keep depositors from losing confidence in banks, banks have a strong incentive to prevent their earnings from being negative. Second, Morgan (2002) also says that: “… uncertainty over the banks stems from their assets, loans and trading assets in particular, the risks of which are hard to observe or easy to change. Banks’ high leverage compounds the uncertainty over their assets; their assets present bankers with ample opportunities for risk or asset substitution, and their high leverage inclines them to do so.” Therefore, bank insiders have a high incentive to hide asset substitution behavior through earnings management. Lastly, banks are highly regulated firms, whose non-performing loan ratio, capital adequacy ratio, liquidity ratio, etc. are strictly regulated. Thus, earnings management is one of the management skills that banks adopt to avoid violating regulations.3 An example, taken from Wall and Koch (2000), is provided to see why bank earnings management is the concern of researchers, regulators and market. There was once a heated debate about the use of allowance for loan loss (ALL) to conduct earnings management in the US. In the fall of 1998, the Securities and Exchange Commission (SEC) of the USA was questioning the overstated loan-loss allowance of SunTrust Bank to conduct earnings management. Thus, investors had no correct information about the bank’s earnings and capital adequacy. The stock price of the bank was thus distorted. Bank regulators, instead, had the opposite concern. They preferred banks to have high loan-loss allowance to absorb more unexpected losses without imposing losses on the Federal Deposit Insurance Corporation (FDIC). Some bank analysts criticized the SEC’s action by arguing that SunTrust Bank’s earnings stability was a function of a conservative credit culture. Banks themselves are worried that they might be caught in a conflict between the SEC and bank regulators. Finally, to provide banks with some guidance about appropriate reserves, the SEC and bank regulators issued joint letters to stress that banks should have prudent, conservative, but not excessive, loan loss allowances that fall within an acceptable range of estimated losses. Hence, banks’ earnings management is the concern of the SEC, bank regulators and banking organizations. The aim of this paper is to study EM and the related issues of banks across 48 countries. We explore the following three questions. First, is bank earnings management a global phenomenon? We answer this question by first providing graphical evidence similar to Burgstahler and Dichev (1997) and Degeorge et al. (1999). We then use four measures of EM to evaluate the degree of earnings management. We find that bank earnings management indeed does exist in our sample countries. Our next question is related to the incentives of banks to manage earnings. Prospect theory, from Kahneman and Tversky (1979), is raised to account for this behavior (Burgstahler and Dichev, 1997 and Degeorge et al., 1999). The theory suggests that individuals derive values from gains and losses with respect to a reference point, rather than being from absolute levels of wealth. It also suggests that individuals’ value functions are concave in gains and convex in losses (S-shape). Thus, for a given increase in wealth, the increase in value is greatest when the wealth of the individual increases from a loss to a gain relative to a reference point. If the preferences of the stakeholders are consistent with the prospect theory, then the manager has an incentive to report earnings that exceed the threshold, or the reference point, such as zero earnings levels or zero earnings changes, to obtain more rewards. Although Burgstahler and Dichev (1997) and Degeorge et al. (1999) theoretically infer that prospect theory is a possible motivation for EM, they do not test it directly. In fact, the S-shape prospect theory suggests that the decision maker will be risk-averse above the reference point and risk-seeking below it, which implies that the relationship between the risk and return is positive above the reference point and negative below it. Employing 85 US industries, Fiegenbaum (1990) finds that asymmetric risk-return association indeed exists, i.e., firms below the industry target are risk-lovers while those above it are risk-averters. Similar to the above notion, this paper also investigates whether or not an asymmetric risk-return association between banks exists, by using banks’ earnings threshold as the reference point. If an asymmetric relation is found, then the prospect theory may be empirically proved as an explanation for earnings management. The third and last question asks how the degree of EM varies across countries, if EM exists. While most banks do engage in earnings management, the degree of management is not the same. It is interesting to investigate reasons behind the wide variations of EM across countries. Leuz et al. (2003) explore a similar issue. They highlight the role of laws and enforcement as important determinants of earnings management, but focused only on non-financial industries across 31 countries. They suggest two competing hypotheses, diversion and penalty hypotheses, to account for the variations. The diversion hypothesis claims that an insider plans to manage earnings so that he can divert a firm’s resources to himself, such as excessive compensation to managers, perquisite consumption, and so on. A good set of laws and strong enforcement may mitigate diversion activities, but they also lessen insiders’ incentive to manage earnings due to the fact that the insider has less to hide from outsiders. As a result, the incentive to manage earnings decreases in a strong legal protection environment. The penalty hypothesis alternatively argues that a strong legal environment encourages earnings management, because negative earnings incur an authority’s penalty. The insider thus has a greater incentive to hide a profit loss when faced with greater expected penalties. Therefore, earnings management increases with a strengthening of a country’s legal protection. Leuz et al. (2003) find that the legal protection of outside investors – including “antidirector rights”, “quality of legal enforcement”, and “accounting disclosure”, which are all extracted from La Porta, Lopez-de-Silanes, Shleifer, and Vishny (1998; LLSV hereafter) – is indeed a key determinant of the quality of financial information communicated by insiders to outsiders. They find that earnings management decreases in stronger legal protection countries. Therefore, the penalty effect is dominated by the diversion effect. Again, since financial institutions are ignored in Leuz et al. (2003), it’s worthwhile to test if earnings management also decreases in the banking industry. Following Leuz et al. (2003), we employ “accounting disclosure” as another factor to explain the variation in the extent of managed earnings across countries, since stringent accounting standards can affect the reliability of financial reports and thus lower the degree of managed earnings. In addition, we introduce an “insider trading” index to explain variations in the extent of managed earnings across countries, since insiders have higher incentives to manage earnings when it’s easy to acquire benefits from insider trading. The paper proceeds as follows. In addition to the first section, the next section shows the earnings histogram of banks in the US and 47 other countries. We then use three earnings management measures to examine the null hypothesis of no bank earnings management across the 48 countries. Section 3 tests whether the prospect theory can be a motivation for banks to manage earnings so as to exceed thresholds. Section 4 tests whether outside investor protection can explain the variation in the earnings management measures across our sample countries. Section 5 provides the conclusion.

This paper finds that the distributions of banks’ net income are half-normally distributed for more than two-thirds of the countries in our sample, suggesting the possibility of earnings management. We use three measures to assess the null of no earnings management. The three measures were originally developed for non-financial industries. Our results show that bank earnings management is common and indeed exists for nearly all of the sample countries regardless of the measures. Conflicting evidence, however, also exists. For example, using conventional measures, US banks show no sign of earnings management, yet earnings management seems to exist with respect to the distribution of banks’ net income. Further research is needed. Our results also demonstrate wide variation in the extent of the earnings management across countries. It appears to be strongly driven by the elements of prospect theory. Furthermore, we find that in order to weaken banks’ incentives to manage earnings and thus improve the reliability of financial reports, stringent accounting disclosure requirements appear to be more effective than developing strong antidirector rights. But, it is striking that stricter law enforcement contrarily results in more earnings management, since managers feel the need to avoid earnings decreases; thus possibly lowering the quality of financial reports of the banking industry.