It was the 19th century English poet William Wordsworth who once warned that: "In modern business it is not the crook who to be feared most, it is the honest man who doesn't know what he is doing." (Goodman, 1997: p113)
In accounting rules there is still much room for interpretation, especially principal based rules like IFRS. It is argued that these principal based rules increase the value of financial statements. However, it also has some downsides. The most important downside is that it can create opportunities for earnings management as argued by Hellemans (2006).
1.1 Social and scientific orientation
An example of fraud mentioned by Bergstresser and Phillipon (2006) is Xerox. This is a company that whose executives have appeared to have manipulated the reporting income during the 1990s. During this period, Xerox' CEO was exercising large numbers of stock options and selling large numbers of shares. In April 2002 the SEC sued Xerox for manipulating reported earnings and forced them to restate their reported earnings from 1997-2001. The effect of this restatement on the reported revenues were $2.1 billion and $1.4 billion on the reported net income. The SEC accused Xerox that it used multiple methods to inflate their net income. Their accounting choices were also inconsistent with US GAAP. The value of the exercised options was over $20 million (Bergstresser and Philippon, 2006).
We endorse the view of Bergstresser and Philippon (2006) that CEO incentives for earnings management are an important item the last two decades. Many people expect a significant relation ship between these to variables.
1.2 Relevance of the research
Due to the social and scientific attention, an extent body of literature is written about earnings management. In the review article of Verbruggen, et al. (2008) is stated the vast majority of papers (72%) depend on financial and/or accounting data to find empirical evidence of earnings management (Verbruggen, Christaens and Milis, 2008: p. 4).
In this research papers, we would first like to create a structure in the motives for earnings management. Because to identify and cure the consequences, we still need to understand the underlying cause of which it origins, before taking a look on different opinions about detecting earnings management.
1.3 Research question
Ronen and Yari (2008), gives three alternative classifications of definitions of earnings management; classifying them as white, gray or black. Beneficial (white) earnings management enhances the transparency of reports; the pernicious (black) involves outright misrepresentation and fraud; the gray is manipulation of reports within the boundaries of compliance with bright-line standards, which could be either opportunistic or efficiency enhancing (Ronen and Yari, 2008).
For this research paper we will use the gray definition (Ronen and Yari, 2008) of earnings management given by Watts and Zimmerman (1990):
'Earnings management occurs when managers exercise their discretion… [over the accounting numbers with or without restrictions]…because (1) the exercised discretion increases the wealth of all contracting parties, or (2) the exercised discretion makes the manger better off at the expense of some other contracting party or parties (Watts and Zimmerman, 1990: p. 135).
Though the definition being used already includes motives for earning management, we want to focus this paper on the main research question:
"Can we detect managers incentives to exercise their discretion over the accounting numbers?"
To answer this question, we have to fulfil the sub question:
1. What is the importance of earnings?
2. What is the background of the management reporting accounting numbers?
3. Which different motives does literature describe?
4. Can we detect earnings management?
1.4 Plan of review paper
In the second chapter we will take a look into the importance of earnings and have a look into the background of the role of management. This chapter is followed by chapter three about the compensation cause. In this chapter we will discuss two mayor different motives found in literature for earnings management. In the following chapter the sub question if we can detect earnings management will be reviewed. We will end the paper with chapter 5, in which we will discuss our conclusion of the research question and state the possibilities of further research.
2. Management and earnings
In this chapter, we will investigate why earnings are so important that they could be the object of manipulation by the management, and will take a look into the background of the role of management.
2.1 The importance of earnings
At first glance earnings are important because accounting information in general is. The traditional view on the value of accounting information is that this information has a dual role: informativeness and stewardship (Ronen and Yaari, 2008).
The informativeness role arises from investors' demand for information to predict future cash flows and asses their risk.
The stewardship role of accounting comes from the separation between ownership and management in public firms, which puts the manager in a position of a steward to shareholders. Since managers act as self-interested individuals, goal congruence between the shareholders and managers is no longer assured. For shareholders, the remedy is to demand information to monitor the manager after he has acted and to provide him with incentives that align his interest with their own, before he acts.
As Watts and Zimmerman (1978, pp 113) state, "one function of financial reporting is to constrain management to act in the shareholders interest." (Watts and Zimmerman, 1978).
Despite that there are other financial statements' components being more important then earnings in certain industries, research by Francis, Schipper and Vincent (2003) provides evidence that reported earnings are more closely associated with prices then cash flows, sales and other financial statements' data.
Although Jensen and Meckling (1976), focus on the manager's excessive benefits, another theory of agency costs operates from the perspective that managers bear personal costs in their employment with the firm. We refer here to the so called principal-agent paradigm.
It is easy to see that the manager's consumption of a company's resources is likely to be excessive: the manager enjoys 100% of every dollar of perk, but he bears only a fraction of the costs in proportion to his (much) lower relative equity holding.
Shareholders can increase the cost of such private consumption by designing equity-based compensation.
The shareholder-manager relationship is characterized by moral hazard. To induce the work-averse, risk-averse manager to exert more effort and not shirk his duty, shareholders must impose risk on him at a higher level than some first-best level. The means is to offer options. Options impose risk on the manager because they are valuable only if the (volatile) market price of shares has risen when they are exercised. Thus a manager who is reluctant to bear personal cost will still take the right action in order to increase the firm's value and enrich himself in the process (Ronen and Yaari, 2008).
Bergstresser and Phillipon (2006) mention the enormous increase in stock-based and option based executive compensation. These forms of compensation are often described as a way to align upper management incentives with the interests of shareholders. On the other hand, this strategy may have had some side effects. In particular, it has been suggested that large option packages increase the incentives for managers to manipulate their firm's reported earnings (Bergstresser and Phillipon, 2006). This we will discuss in chapter 3.
The recognition that managers, not the owners, makes most of the firm's decisions motivates the conflict of interests between the two parties. The decision of the management do not necessarily coincide with the wishes of the shareholders. Being rational, each party takes actions that he considers to be beneficial to him personally, without necessarily taking into account any benefits to the other.
Ronen and Yaari (2008) presents three approaches: the contracting approach, the decision making approach and the legal approach. Figure 1 classifies the three approaches according to two lexicographic criteria: knowledge and power.
The costly-contracting approach highlights the importance of formal contracts. Earnings and other accounting numbers provide summary statistics that are valuable for designing efficient contracts, given that contracts fail to specify all future contingencies. When an unforeseen contingency does occur, the contracting parties may be stuck with the old contract. In such a situation earnings change from means to ends, and firms may manage earnings to comply with figures specified in their contracts (Ronen and Yaari, 2008).
The decision making approach regards earnings as providing valuable information for making decisions. If all stakeholders are fully rational, earnings management cannot take place without explicit or implicit consent from investors (Ronen and Yaari, 2008).
The legal political approach recognizes that shareholders lack tools to control management effectly. Earnigns numbers are a valuable performance measure that summarises the activities of the firm and allow shareholders to make better use of their limited set of tools. Earnings management in the good sense- providing a signal on future value- is an efficient means to bridge the information asymmetry between management and shareholders. Earnings management in the bad sense-distorting the truth- is the result of poor governance (Ronen and Yaari, 2008).
2.2 The background of the role of management
An earnings management perspective requires a focus on those senior officers who are responsible for reporting the firm's earnings: the chief executive officer (CEO), the controller, and the chief financial officer (CFO). Literature has different perspectives about the influence of these senior managers and earnings management. Kasznik (2003) find that the distinctions among the different definitions of senior management might be innocuous, while Geiger and North (2006) find that incoming CFO's tend to be associated with a lower level of earnings management.
Due to the fact that the Sarbanes-Oxley Act (SOX) has redefined management's fiduciary duty, increasing senior management's responsibility for the financial reports we follow Kasznik's findings in this literature review.
3. Compensation motives
Bergstresser and Philippon (2006) provides evidence that 'the use of discretionary accruals to manipulate reported earnings is more pronounced at firms where the CEO's potential total compensation is more closely tied to the value of stock and option holdings'.
In this section we review the components of typical compensation package of CEO's and other senior managers which can be influenced by earnings management. Typical compensations in general are cash compensations (eg. salary and bonuses), stock and options (equity holdings, options), CEO turnover, insider trading and management buyouts.
3.1 Extent of earnings management vs height of accruals
Bergstresser and Philippon (2006) indicate that there is a correlation between the extent of earnings management and the height of accruals. Dechow et al. (1995) suggest that first a distinction should be made between non-discretional and discretional accruals.
The first category contains obligatory expenses not yet realised but recorded in the account books, e.g. next month's salary.
The second category are non-obligatory expenses, such as anticipated bonuses of management.
In their study, Dechow et al. (1995) elaborate on the second category, since these accruals seem to be most likely to be manipulated. They aim to find a model which can prove the detection of earnings management. Which will be discussed in the next chapter.
Earning management include the misrepresentation of profits, revenues, or costs. In previous research different methods are described. Earnings management can be done by using accruals "a manager may increase or decrease the levels of accounting accruals (such as accounts receivables, inventory, accounts payable, deferred revenue, accrued liabilities, and prepaid expenses) in order to reach a desired profit" (Dharan, 2003). Earnings management can also take place by altering the future expected revenue of pension assets. Another form of earnings management is delaying or rushing large investment plans.
3.2 The effect of bonus schemes
This sub-article analyzes the typical bonus schemes. Bonus schemes that are based on earnings are a popular means of rewarding corporate executives. Fox (1980) reports that in 1980 ninety percent of the largest US company's uses this kind of earning-based bonus plan (Fox in: Healy, 1983).
The rest results shows us that these bonus schemes create incentives for managers to select accounting procedures and accruals to maximize their bonus. There is a strong relationship between accruals and managers income-reporting incentives under their bonus contract. Managers often choose accruals that decrease the income when upper or lower bands are binding, if these bands are not binding they are more likely to choose income-increasing accruals.
Different tests comparing firms with upper or lower bands and firms excluding upper or lower bands, supports this theory. Accruals are lower for firms with binding bonus plans than for firms with no upper bound.
3.3 Benefit on pension plans
Bergstresser, Desai and Rauh (2006), search for evidence of managers changing decisions in order to capitalize on these decisions. Examples are found at IBM, under CEO Louis Gerstner, Jr. the expected returns on defined benefit pension plans were raised while stock market returns declined. By raising the assumed returns, the share of income due to the pension plan results raised to almost 5% by the year 2000. On its own this is no proof of capitalization on these decisions, but when compared to the stock options and shares paid to L. Gerstnera, it becomes clear. While the number of options exercised grew, the pension plan made a larger part of the firm's returns.
Bergstresser, Desai and Rauh (2006) have found evidence of such behaviour by testing 3.350 firms in their research. Their conclusion is that a manager will use earnings management more often when higher returns of the firm are capitalized in stock prises and the reported earnings are highly affected by the altered returns on the defined pension plan. These can be incentives for the manager to alter the expected earners, whether these incentives are in the best interest of the company or only in the interest of the decision maker is not always clear.
4. Can we detect earnings management?
As given in the previous chapter, that managers with more accruals are more likely to use earnings management to manipulate their accruals, is there a way for gatekeepers like accountants to detect earnings management?
4.1 Detecting earnings management tests by Dechow et al.
Dechow et al. (1995) reviewed five prior models to show if and how earnings management can be detected. They gather their own set of variables and use the existing models to see if earnings management can be detected.
The models researched and reviewed by Dechow et al. (1995) did indeed detect earnings management, which is the same conclusion as Bergstresser and Philippon (2006).
Though, Dechow et al. (1995) also found some other issues. The power of the different models is relatively low for earnings management of economic magnitude. They only showed earnings management in a smaller scale. Also when the models were tested in years when firms showed high performance, they led to misspecified tests, thus making data unreliable to forecast earnings management. This implies further research on the subject is needed, which will be illustrated in the next paragraph.
4.2 Detecting by the use of tax expense accruals
Other researchers focus on determining the use of tax expense accruals as a way to detect earnings management. Phillips et al. (2002) define deferred tax expense as follows:
"Deferred tax expense is a component of a firm's total income tax expense and reflects the tax effects of temporary differences between book income (i.e., income reported to shareholders and other external users) and taxable income (i.e., income reported to the tax authorities) that arise primarily from accruals for revenue and expense items that affect both book and taxable income, but in different periods."
This is, in our opinion, a good definition for this paper since it manages to show the opening for earnings management in the deferred tax expense. Phillips et al. (2002) also explained why they used the deferred tax expense as their empirical proxy for book-tax differences. This is because various previous research showed that the tax law, in general, allows less discretion in accounting choices relative to the discretion that exists under the general accepted accounting principles.
To ensure whether these deferred tax accruals are useful in measuring earnings management in three specific cases they created the following hypotheses:
H1: Deferred tax expense is incrementally useful to accrual measures in detecting earnings management to avoid an earnings decline.
H2: Deferred tax expense is incrementally useful to accrual measures in detecting earnings management to avoid a loss.
H3: Deferred tax expense is incrementally useful to accrual measures in detecting earnings management to avoid failing to meet or beat analysts' earnings forecasts.
All of these three cases are incentives for managers to increase their earnings. Especially in when the managers have option or share based compensation they want to meet the investors/analysts' expectations. If the deferred tax expense is useful for detecting earnings management it would endorse the conclusions made by Bergstresser and Philippon (2006).
4.3 The 'Jones Model' by Phillips et al.
The relationship between the probability of earnings management in the above mentioned cases and the deferred tax expense are empirically tested in two versions of the 'Jones model' (Jones, 1991). Although there are several critics about the 'Jones model' like Bernard and Skinner (1996), who argue that abnormal accruals estimated using "Jones-type" models reflect measurement error due in part to the systematic misclassification of normal accruals as abnormal accruals.
Phillips et al. (2003) however, take a different tact and argue that this error can be reduced by focusing on deferred tax expense instead of attempting to decompose accruals into normal and abnormal components.
The Jones model (Jones, 1991) contains a variable called DTEit (DTE = firm i's deferred tax expense in year t, scaled by total assets at the end of year t-1).
Phillips et al. (2003) expect that the beta coefficient of this variable is positive and significant. This would imply that there is a positive and significant relationship between earnings management and deferred tax liabilities.
Results of the empirical research support the first two hypotheses, but reject the last hypothesis. So, according to Phillips et al (2003) deferred tax expense is generally incrementally useful to the various accrual-based measures in detecting earnings management to avoid an earnings decline (H1) and to avoid reporting a loss (H2), but not in detecting earnings management with regard to avoiding failing to meet or beat analysts' forecasts (H3).
Because of this positive relationship between DTEit and earnings management, Phillips et al (2003) show that there is a relationship between the use of accruals and earning management.
This also supports the research done by Bergstresser and Philippon (2006), as described earlier.
