What is earnings management and is it good or bad? Can positive accounting theory and signalling theory helps use to understand earnings management?

Definition of earnings management

Earnings management is the process by which management can potentially manipulate the financial statements to represent what they wish to have happened during the period rather than what actually happened (Scott 2009).

Factors that motivate earnings management (internal targets, income smoothing, external expectations and window dressing)

Management wants to manage earnings because of internal and external pressures.

The main internal reason is to meet targets. The targets may be there for a number of reasons. Some may just be budgeted numbers, which if they are not met will look unfavorable on the person, department, or company that “blew the budget.” Others may be “required” numbers, which if not met will mean that a person doesn’t get his or her bonus.

The external factors are a bit more diverse. The company may have previously projected numbers that external parties are now expecting the company to meet or exceed. External analysts may have made their own predictions public, which the company would now like to achieve.

Investors, and potential investors, like to see continual upward income growth. It looks really positive and looks as if the company is doing well in the charts found in annual reports. Hence, income smoothing is the second external factor potentially contributing to earnings management.

Finally, if a company is looking for new financing, they will have an easier time obtaining it (or obtain better terms if it is debt financing) if they have good looking financial statements. Therefore, window dressing is the final factor listed.

Techniques of earnings management

The big bath – This form of income manipulation can be thought of as part of income smoothing. What it usually does is effectively accelerate expenses and losses into a single year with already poor results so that future income looks better and smoother. Even though the FASB has issued fairly recent statements to reduce the magnitude for taking a big bath, companies can, and do, still use this technique. Examples may include recognizing losses on assets that have a fair market value below the current book, or carrying, value, cookie jar reserves (to be discussed in 3. below); and doing a restructuring (taking the expenses allowed under SFAS No. 146) that a company may not otherwise have done.

Creative acquisition accounting – As the number of acquisitions has decreased (since the late 1990s) and with the advent of SFAS Nos. 141 and 142, this doesn’t seem to be as much of a problem as it once was. Still, when a company has made an acquisition or acquisitions during the year, the transactions should be looked at closely to see exactly how they were accounted for and what effect the treatment has on current, and will have on future, earnings.

Cookie jar reserves – These can go along with the big bath and are a form of income smoothing. Earnings are managed under this method by selecting the period in which a revenue or expense item is taken. This is usually done for expenses that are based on estimates. If a company is having a particularly good year and next year’s results are uncertain, they can over-accrue some reserves in the current year and then have the ability to under-accrue them in the next year if needed. By doing so, they effectively inflate the following year’s income at the expense of this year’s.

Income, thus, appears smoother, and the company may be able to publicly forecast higher profits for the following year even if their business isn’t actually going to do any better in the following year. This may temporarily be good for the stock price, but it isn’t good for those wanting to know how the company is actually performing.

Materiality – This topic may not be a big deal to small companies since nearly everything is material and, hence, should be accounted for. But for large, publicly traded companies with revenues and assets in the billions of dollars, they can potentially get away with millions of dollars worth of misstatements and merely write them off as being “nonmaterial” in nature. Auditors are primarily concerned with material misstatements. Materiality has the potential to allow companies to slightly fudge their numbers, just enough to get them to where the analysts forecasted.

Revenue recognition – Sort of the flip-side of cookie jar reserves, improper revenue (or expense) recognition can lead to inflated financial statements now at the expense of future earnings. Some companies that have dabbled with this earnings management technique then have to inflate revenue in the next period even more to make up for the shortfall caused by the prior period’s accelaration of revenue. It becomes a never-ending game of covering up for the previously improperly recorded revenue and can fairly easily lead to outright fraud. Several of the bigger scandals of the past few years have been the result of companies improperly, and/or prematurely, recording revenues in order to meet or exceed forecasts, only to have the house of cards eventually come tumbling down, resulting in massive restatements to the prior financial statements, new management, new auditors, and very low stock prices (if not bankruptcy).

Pros (Goods) of earnings management

It is good, from an efficient contracting perspective. This argument assumes, however, that the earnings management was anticipated by the principal when the bonus contract was being negotiated, so that it is allowed for in setting the bonus rate. Is to lower contracting costs in the face of rigid and incomplete contracts.A second, and more controversial, side is that earnings management can reveal inside information to investors. A provocative discussion question here is to ask if earnings management can be thought of as an extension of the accrual process. That is, if accrualssmooth out lumpy cash flows to produce a more useful measure of quarterly and annual performance, why can’t earnings management be used to smooth out annual accrual-based earnings to produce a more useful multi-year measure of persistent earning power? Such a measure may help investors better predict future firm performance, which is a major goal of financial reporting.To pursue the argument that earnings management as a vehicle to release inside information can be “good,” the case of General Electric Co. (Problem 9 of this chapter) works well to get the point across. The steady increase in GE’s reported earnings over the years is quite impressive. I point out the complexity of GE to the point where even financial analysts have difficulty in understanding the whole company. As a result, it is very difficult to estimate GE’s persistent earning power. I also point out that a simple announcement by GE of its persistent future earnings is ‘blocked.” Such announcement lacks credibility since, for such a complex company, the market has little ability to verify it. This sets up the role of “good” earnings management as a credible way to reveal this information. An argument that GE does engage in earnings management for this purpose is supported by both the variety of earnings management devices available to it and the steadily increasing pattern of its earnings over time.

It is interesting to note that GE’s earnings management came under suspicion in the market in the early 2000s, due to the severe apprehension of post-Enron investors about earnings management in general. According to an article “General Electric: Big game hunting” in The Economist (March 14, 2002) investors may have interpreted GE’s increased reported earnings for 2001 as evidence of bad earnings management, since poor economic conditions during 2001 suggest that earnings should have declined. In addition, GE appointed a new CEO in late 2001. The Economist suggests that the market may have less trust in the new CEO than in Jack Welch, the highly regarded former CEO, simply because he is less of a known quantity. As a result, the market may have felt that there is a higher likelihood that GE will use its considerable potential for earnings management for bad purposes rather than good.GE’s response to these market concerns is worth noting. It started to release considerably more information. Further discussion of how GE

worked to overcome investor scepticism is given in Problem 21 of Chapter 12.Theoretical and empirical evidence in favour of good earnings management is given inSection 11.5.2, which I have marked as optional reading for those that wish to pursue goodearnings management in greater depth. Suffice it to say that there is considerable evidencein this regard.

Cons (Bads) of earnings management

Despite the above arguments, most people would likely regard earnings management withsuspicion, reinforced by revelation of serious abuses of earnings management by Enron andWorldCom and numerous other corporations in the early 2000s. Consequently, studentsshould not be left with the impression that it is necessarily good. A useful place to start isHanna’s 1999 article in CA Magazine, which is well worth assigning and discussing. Theimportant point to get across from this article is that management is tempted to provideexcessive unusual, non-recurring and extraordinary charges, to put future earnings in thebank. Furthermore, these future earnings are buried in operations. This makes it difficult forinvestors to diagnose the reasons for subsequent earnings increases. Nortel Networks’reversals of its excess accruals (see Theory in Practice vignette11.1 in Section 11.6.1)provide a vivid example of Hanna’s argument. Also, the effect on future profits of putting earnings in the bank has been recognized by an article in The Economist (“A world awashwith profits, Business is booming almost everywhere,” February 18, 2005, pp. 62-63). Thisarticle states that one reason for the dramatic increase in firm profits during 2002-2004 is thatthey are an “accounting fiction,” which apparently means that they are a consequence ofearlier writeoffs.I find that to drive home these various considerations, an example of how earnings management can go too far is instructive. An excellent case in point is the downfall of “Chainsaw Al” Dunlap at Sunbeam Corp. Jonathan Laing’s 1998 article in Forbes is reproduced in Question 10. Laing demonstrates that Sunbeam’s 1997 reported earnings were almost completely manufactured by means of discretionary accruals. The

substantial first quarter, 1998, loss reported by Sunbeam supports Laing’s analysis, and the “iron law” of accrual reversal.I think that Laing’s analysis of the effects of the $17.2 million drop in Sunbeam’s prepaid expenses for 1997 is backwards–see part a of Question 10. If I am not correct in this, presumably other instructors will let me know. However, even taking this error into account does not substantially alter Laing’s conclusion that 1997 earnings were manufactured.

Techniques of earnings management

Revenue recognition. revenue recognition is an accruals-based earningsmanagement policy Is effective because recognition criteria under GAAP are vague and general. A company can speed up, or slow down, revenue recognition but disguise the change through vague wording of its revenue recognition accounting policy disclosure. Also, as in the case of Coca-Cola, revenue recognition can be speeded up by stuffing the channels to unconsolidated subsidiaries or customers, without any formal change in revenue recognition policy.Stuffing the channels can be difficult for investors, or even auditors, to detect.

A disadvantage of revenue recognition as an earnings management device is that accruals reverse. Consequently, it is difficult to maintain increased reported revenue over time. Also, stuffing the channels becomes quite costly if it is necessary to compensate the subsidiary or customer for carrying costs, as Coca-Cola did.

Reasons for income smoothing

They may feel that the market rewards share prices of firms that report steadily increasing earnings, consistent with the findings of Barth, Elliott, and Finn (1999).

They may want to keep earnings for bonus purposes between the bogey and cap of their bonus plan.

They may want to reduce the probability of violation of debt covenants.

They may want to convey inside information about persistent earning power by smoothing reported earnings to an amount they feel can be sustained.

They may smooth earnings because of implicit contracts, consistent with the findings of Bowen, Ducharme, and Shores (1995). The firm may be able to secure better terms from suppliers and other stakeholders with which it has a continuing relationship if it reports steady earnings.

Do managers accept securities market efficiency?

Evidence of good earnings management is consistent with managers’ beliefs that marketsare reasonably efficient. Why use earnings management to reveal inside information if themarket cannot interpret it? However, evidence of bad earnings management may or may notbe consistent with efficiency.On the one hand, managers may feel that they can fool the market by managing theirearnings, which seems to have been the case with Sunbeam management. Emphasizing proformaearnings (Section 7.4.2) is another tactic that seems inconsistent with acceptance ofefficiency. It is hard to believe that managers would continue to attempt to manipulateinvestors’ beliefs if an efficient market immediately detected and penalized such behaviour.On the other hand, bad earnings management may hide behind poor disclosure. If the marketis not aware that reported earnings are being managed, it can hardly be concluded that themarket is inefficient. Rather, the question is whether the market will react once it suspects or becomes aware of the earnings management. The market’s negative reaction tothe frequency of non-recurring charges as an indicator of possible earnings management, asdocumented by Elliott and Hanna (1996) (see Section 5.5) suggests considerable efficiency,for example. Also, the market’s post-Enron suspicion of GE’s earnings management,discussed above, is also consistent with efficiency.The text concludes that at least some managers do not accept market efficiency. However, italso concludes that markets are sufficiently close to full efficiency that improved disclosurewill reduce bad earnings management.

To Summarize the Strategic Aspects of Accounting Policy Choice

I end my discussion of earnings management with two main points:(i) I emphasize the concept of strategic accounting policy choice, wherebymanagers choose accounting policies to achieve certain objectives. These objectivesmay include efficient contracting, such as avoiding excess earnings volatility forcompensation and debt covenant reasons, which may conflict with accounting policiesthat are most useful to investors. This greatly expands the role of financial reporting,since we now formally recognize two main roles of financial reporting– reporting toinvestors and reporting on manager performance. Both roles matter since the qualityof manager effort and the well-working of managerial labour markets is as important tosociety as the quality of investor decisions and the well-working of securities markets.The conflict between these two roles, I hope, validates to the class the time spent onbasic game and agency-theoretic concepts of conflict in Chapter 9.(ii) I emphasize that managers have a legitimate interest in accounting policychoice, since their operating and financing policies, and even their livelihoods, are atstake. This view is in contrast to many discussions of standard-setting wheremanagement seems to be the “bad guys,” opposing every new standard that comesalong. The theory provides several legitimate reasons why managers will beconcerned about changes to GAAP.