|
|
|
|
CIAO DATE: 12/03
Poor Diagnosis, Poor Prescription: The Error at the Heart of the Sarbanes-Oxley Act
On The Issues
March 2003
Because they are so easily manipulated, audited financial statements are not adequate indicators of companies' health. Sarbanes-Oxley did a disservice to investors by enshrining this accounting measure as companies' central financial disclosure. Cash flow is a much more reliable indicator, and it would be of much greater use to investors if the Securities and Exchange Commission were to encourage companies to present this data in a usable form.
The key element of the Sarbanes-Oxley Act, which became law in 2002, was the establishment of a regulatory body to oversee and regulate the way accountants audit public companies. Cutting through all of the legislative language and the regulatory action it has spawned, the act's principal effect was to enshrine the audited financial statement—prepared under Generally Accepted Accounting Principles (GAAP)—including a balance sheet, income statement, and statement of cash flow, as the central financial disclosure of companies whose shares are traded in the public securities markets.
In adopting the act, Congress acted hastily, without adequate thought, and apparently without an understanding of the problem it was seeking to address. Although the obvious purpose of the act was to improve the auditing of financial reports by independent auditing firms, ineffective or incompetent auditing was not the most significant reason for the decline in investor confidence in financial reports, and improvements in the auditing process will not cure the problem that investors face. The deficiency associated with audited financial statements is not their accuracy—as Congress seemed to think—but their adequacy. The fact is that GAAP financial statements, even when competently audited, do not, by themselves, furnish adequate or reliable information to investors. By placing special emphasis on the importance of audited financial statements and inducing the Securities and Exchange Commission to focus even more attention on financial statements prepared under GAAP, Congress has made the problem worse.
In reality, the underlying cause of the collapse of investor confidence was the recognition that audited financial statements prepared under GAAP, or any other system of financial statement preparation currently in use, could never give a completely accurate picture of the prospects of public companies. As a result, the appropriate policy for Congress would have been to diminish the importance of audited financial statements by encouraging the disclosure of information that is more useful to investors and less subject to manipulation by corporate management.
Even as Congress was acting, there was evidence throughout the economy and the financial system that GAAP financial statements were of diminishing relevance to investors for several reasons:
1. GAAP and all other methods of financial reporting, including International Accounting Standards, are inherently malleable, and results can be easily adjusted by corporate managements to meet predetermined targets. It is possible, perfectly legally and within the rules of GAAP, to produce audited income statements showing results that are highly variable simply by changing predictions about the future—for example, by increasing or decreasing reserves, or depreciation rates.
2. GAAP financial reporting is based on cost—an element that is becoming increasingly less relevant as more and more companies rely on intangible assets to generate earnings. Under GAAP, many of the costs associated with creating intangible assets are written off as incurred. As a result, many companies are generating most of their earnings from assets not actually on their balance sheets.
3. There is strong evidence that investors are relying on many factors other than audited earnings in making judgments about the value of companies, particularly free cash flow. Surveys of investors show that audited earnings are not even at the top of the list.
Malleability of GAAP Statements
Contrary to what Congress seemed to believe, there is no such thing as a statement of income under GAAP that is inherently "correct" or that can be made inherently "correct" by an effective audit. All such statements—quite legitimately and necessarily—are the consequence of management's opinion about a vast number of events and circumstances that may or may not arise in the future. Thus, while better and more expensive auditing—which is surely coming—may improve the chances of catching actual fraud, it will not improve the "accuracy" of financial statements.
Auditors do not prepare the financial statements and are not capable of making the judgments that management must make in the course of doing so. The most auditors can do is question management estimates of such things as reserves or depreciation rates—items that can have major effects on reported earnings—if they seem so grossly out of line as to be unreasonable. These cases will be unusual. The notion that better regulation of accountants and more thorough audits will produce substantially more accurate financial statements is wrong.
That financial statements prepared under GAAP are inherently subject to manipulation should be obvious to anyone who follows the securities markets. The economy moves up and down, interest rates, exchange rates, and raw material costs fluctuate with supply and demand, consumer preferences change constantly, foreign competition and competition abroad for U.S. companies operating internationally is highly variable, and yet the earnings of companies did not fluctuate appreciably during the boom years of the late 1990s. During that period, the earnings of public companies grew steadily from year to year, frequently hitting to the penny the forecast for quarterly earnings per share made by the sell-side analysts.
This was possible because, wholly legally, companies could hit earnings targets by adjusting one or more of the variable elements involved in the preparation of financial statements under GAAP. This gave rise to claims that companies were engaged in "earnings management," but to no effective cure. In fact, what seems to have been occurring was a game in which analysts and investors were testing the quality of a company's stated earnings by determining whether management could hit its targets. If it could, that meant the company's earnings were probably growing, although not necessarily as stated. If it could not, that was a signal that the company had run out of ways of adjusting its results to produce earnings growth, and that in turn suggested that its earnings had really fallen quite dramatically. It was because of this that we saw the strange market phenomenon in which companies that missed their earnings targets by a penny or two saw 20 or 30 percent declines in their share prices.
Another effort to address the variability of GAAP financial statements was the informal use of EBITDA—earnings before taxes, depreciation, and amortization. The objective here was to show the company's GAAP earnings before many of the major variables that are affected by management decisions. However, the SEC and some commentators have objected to the widespread use of this number because it diminishes the comparability of earnings thought to result from all companies using the complete GAAP disclosure system.
In a sense, neither management nor analysts should be blamed for the earnings game, or for taking actions that will smooth earnings growth over time. Since there is no "correct" statement of income, and corporate managements were and are in a position to show earnings results within a broad range, smoothing earnings so that they grow gradually over time is not necessarily dishonest, and it is certainly rational. The problem then is not dishonest managements—although there are some—it is excessive reliance on a financial disclosure mechanism—GAAP—that inherently permits a variety of outcomes. The effective cure for this problem was clearly not the Sarbanes-Oxley Act, with its reliance on more effective auditing. That simply placed even greater emphasis on a flawed system. The real cure should have been changes in the securities laws that would encourage the development and greater use of other kinds of financial and nonfinancial disclosure, some of which are discussed below.
By placing greater emphasis on the GAAP audit as the primary means of measuring past financial results and providing clues about a company's future prospects, the Sarbanes-Oxley Act has stalled the development of other more useful and effective indicators and measures, probably for a long time. The act's adverse effect can now be seen clearly in the SEC's recently proposed Regulation G, which requires companies that use any non-GAAP measures of financial performance to reconcile them to their GAAP financial statements and warns that any such non-GAAP measures must not be materially misleading. This will limit the ability and willingness of companies to develop and use new measures of performance-in part out of fear that these will be viewed in hindsight as misleading to investors and because some of them cannot be reconciled to GAAP without introducing judgmental or predictive factors that in hindsight can be the basis for lawsuits.
Intangible Assets
All of this retrograde activity at the SEC is occurring at just the time when many in the accounting industry have begun to recognize that GAAP financial statements are inherently unable to produce accurate measures of assets and earnings for companies that rely on intangible assets.
Intangible assets are knowledge assets—patents, trade secrets, pharmaceutical designs, computer software, brands, customer relationships, employee intellectual capital, and the like. Increasingly, intangible assets are the assets generating revenues and profits for the many companies engaged in what has come to be called the postindustrial, knowledge, or information economy. It has been estimated that 80 percent of the assets of the Fortune 500 consist of intangible assets. Since the early 1980s, market-to-book ratios for public corporations have grown from approximately one-to-one to about five-to-one today. This growth has been attributed to the increasing importance of intangible assets. The discrepancy arises because the costs of creating internally generated intangible assets like pharmaceuticals designs or software programs—unlike buildings and machinery—tend to be written off as incurred and thus do not appear on balance sheets.
Obviously, if a company is writing off the development cost for a new software program, when the program is ultimately marketed there are fewer offsetting costs, such as depreciation, to reduce its profits. The traditional idea that income statements should endeavor to account for both revenues and costs in the same period simply does not work. If the computer program the company has developed is successfully marketed, a great deal of revenue and income will be recognized with relatively little in the way of balance sheet assets. This accounts for the growing gap between balance sheet values and company values since the early 1980s.
This can also have a highly distortive effect on GAAP earnings, completely apart from the problem of adjustment or smoothing of earnings discussed above. For example, let us consider an Internet company that is attempting to develop a large list of subscribers to its service. The more subscribers the company has, the more income it will generate, since advertisers go where the eyeballs are. But how do we treat the costs of developing this list of subscribers? If we capitalize these costs—on the theory that the company is creating an asset in its subscriber list—the company's expenses during its start-up period will be low. Thus, the company may show early profits that could be misleading to investors who do not understand that many of its costs are not being charged against revenues. On the other hand, if we expense the costs and the company ultimately develops a strong subscriber base, it will generate a lot of income from the use of this asset without many of the associated costs one would normally expect—suggesting a high rate of operating profit that could also be misleading to investors.
Facts like these can make price-earnings ratios—a very popular way of judging whether companies are over or under-valued—virtually useless. For example, in the case of companies with substantial amounts of intangible assets, investors would be correct to assume that certain costs the company has written off in the current period were actually investments—since they created revenue-producing assets—and should have been capitalized. If they recast the company's balance sheet in this way, investors will logically conclude that the company's real earnings-which have been subject to reduction through the expensing of costs associated with the development of the company's intangible assets—are higher than its reported earnings. For this reason, P/Es can be considerably higher than historical levels and still not reflect a so-called bubble.
The problem here is inherently insurmountable; whether the subscriber development costs should be capitalized as an asset or treated as an expense depends on whether the company is successful in developing a profitable subscriber base. In other words, to judge the past, we must know the future.
This does not happen as readily with the typical industrial company for which GAAP accounting was initially developed. Accounting for these companies is based on cost, principally the cost of assembling the plant and equipment that is the source of their revenues and profits. These companies buy their plant and equipment at a market price and carry these income producing assets at their depreciated cost. The depreciation is matched each year against the revenues, producing an operating profit. There may be many other ways that management can manipulate or smooth out this profit, but the foundation of the company's earnings are the initial cost of the plant and equipment or other hard assets that it carries on its balance sheet. The assets may not produce profits, and may ultimately be devalued for that reason, but their initial cost and whether they should be capitalized or expensed is not an issue.
Thus, as the knowledge economy grows and intangible assets increasingly become the basis of company values, GAAP accounting becomes less and less relevant as a way of measuring either the past profitability of companies or for estimating their profitability or cash flows in the future. Instead, commentators who have been looking to the future have been recommending the development of new indicators and measures that will give investors a better picture of how companies are doing and what their prospects are. These indicators and measures are sometimes called "value drivers" and consist of facts about the company's business that provide part of the picture of what will make it successful. Examples of value drivers are employee retention rates, customer turnover rates, product development cycle time, employee and customer acquisition costs, new product success rates, and product reject rates.
Needless to say, the renewed focus on audited GAAP financial statements demanded by Sarbanes-Oxley sets back the development of these indicators. The act made very clear that Congress wants companies, accountants, and financial experts to spend their time trying to obtain marginal improvements in auditing GAAP financial statements rather than developing the new indicators and measures that will eventually make it possible for investors to get a better picture of the value of companies—which today is most public companies—that use intangible assets to generate their earnings.
Cash Is a Fact
There is a consensus among financial experts that companies are valued, not on the basis of their audited earnings, but on the basis of their current and estimated future cash flows. Many studies have shown that multiples of audited earnings—despite the attention paid to this number by the media and sell-side analysts—are not as good a predictor of the market value of companies as discounting their current and estimated future cash flows. As Warren Buffett has described the process, in his practical way: "You just want to estimate a company's cash flows over time, discount them back, and buy for less than that."
There are a number of examples that show the importance of cash flows rather than audited earnings. Foreign companies that have listed their shares on the New York Stock Exchange or NASDAQ have been required to restate their audited earnings in accordance with GAAP. In some cases, this has resulted in substantial declines in reported earnings from what they had reported under the accounting systems they had previously used. Nevertheless, their share prices did not substantially decline.
Another example is what has occurred when companies change their inventory valuations from FIFO (first in first out) to LIFO (last in first out). Ordinarily this switch reduces reported earnings, because the most recently purchased raw materials and other inventories are likely to be more expensive than those purchased earlier. Yet, according to studies, many of these companies receive higher market valuations after the switch, probably because the higher inventory valuations will reduce taxes and result in higher cash returns. Conversely, studies have found that companies that switch from LIFO to FIFO receive lower market valuations.
Finally, when analysts have studied the effects of accounting for mergers as purchases rather than poolings, they have found that while the goodwill that arises in purchase accounting reduces earnings per share, it does not have an adverse effect on the stock price of the merged companies. These companies generally had higher price-earnings ratios after the merger—because their audited earnings were reduced by the goodwill charge. But when the goodwill charge is excluded through the use of EBITDA, analysts found that these companies had the same EBITDA multiples as others in their industries.
These examples suggest that the market-that is, sophisticated investors—sees through the GAAP accounting to the underlying economic reality, which are the actual cash flows of these companies. This is logical for two reasons. First, GAAP earnings and earnings multiples make no effort to predict the future, except insofar as current results can be projected forward in time. The investor is left with no mechanism for estimating the value of a company except a guess about whether its profit margins will be sustained. Discounted cash-flow analyses, however, put a value on a company based on what is known in the present about the likely future cash flows of the company, discounted for the riskiness associated with that prediction. Since discounted cash-flow calculations take account of the cash investments that must be made to produce a stream of earnings, they automatically differentiate between companies that are inherently more profitable than others because they have to invest less to produce their cash flows. This information is far more useful than GAAP earnings for investors who want to assess the relative value of companies.
Second, from what has been said earlier, audited earnings are simply too easily manipulated to provide a reliable guide to a company's success; by making adjustments of various kinds-wholly within the rules of GAAP—company managements can smooth earnings over time and also obscure for a period of time what is essentially a weakening competitive position.
In one striking example, the price of Enron's stock began to fall six months before there was any indication that the company was tinkering with its financial results, and while it was reporting record revenues and strong earnings. The reason for this is that investors were looking at Enron's cash flows, which were not keeping pace with its reported earnings. Now we know why—the company was generating paper earnings that did not result in actual cash income—and the investors who were smart enough to catch this discrepancy sold the stock before those who were less astute, or relied on the statements of the sell-side analysts who focused completely on reported earnings.
This outcome points to another reason for the superiority of cash-flow analysis over audited earnings. It is an old securities market saying that "Earnings are an opinion, but cash is a fact." Even honest managements may be enticed to smooth earnings in order to show steady growth; dishonest managements, such as those formerly at Enron and Worldcom, may go further and obscure competitive declines by manipulating results. The fact is, however, that cash results are not as easily manipulated as audited earnings. Of course, managements that are willing to engage in fraud can probably fake cash results as well as audited earnings, but for most companies it is far more difficult to manufacture cash than to manufacture earnings that even diligent auditors will certify.
For this reason, Sarbanes-Oxley's sole focus on audited earnings as the key to corporate financial disclosure was wholly misplaced. After Enron, Worldcom, and others, Congress and many investors were concerned about the accuracy of financial statements. The correct solution was not to create a massive regulatory structure to improve the auditing process but to encourage companies and the SEC to develop methods that will provide investors with information on company cash flows, making it easier for analysts and investors to understand how effectively and efficiently the company is able to generate cash.
What to Do Now
Although Sarbanes-Oxley has made it significantly more difficult, and the proposed Regulation G continues to reflect the agency's unwavering focus on audited earnings and GAAP financials, it is necessary for the SEC to take a broader view of financial and other disclosure. With a new chairman coming in, there is some hope that the commission's disclosure policies can be redirected so that they actually help investors rather than hinder them.
The various measures and indicators that could help investors and analysts better understand the value of intangible assets do not yet exist. Industry groups must be convened to develop them. The SEC has enormous convening power and could accomplish a great deal simply by asking public companies to begin the process of developing these indicators. It will be a time-consuming and arduous process, but that is an argument for beginning sooner rather than later.
However, studies have shown that investors consider cash-flow data for companies to be more important than audited earnings in determining company values. And we can see that excessive attention to audited earnings has created an unhealthy situation in which companies feel impelled to smooth or otherwise manipulate their audited results in order to meet targets set by analysts. Moreover, sophisticated investors already know how to extract cash-flow information from a company's financial data and apply to it the various formulas that yield an estimated present value. This process is so well developed that financial experts can show a strong correlation between discounted cash flow and the day-to-day values of listed companies—a far stronger correlation than can be obtained from the use of earnings multiples. One reason for this is that discounted cash flow takes account of the investments that will be necessary in the future to generate the projected cash flow and when those investments must be made. Another reason is that cash flows are far more difficult to manipulate than audited earnings, so analysts are generally starting from a firmer base.
This suggests that the SEC could accomplish a great deal immediately simply by making cash-flow data more accessible to ordinary investors. The necessary information is not obscure, although it must be estimated and discounted appropriately for the various risks associated with the production of that cash flow over time. To do a projection of the cash returns that can be expected from a business over time, the principal information one needs consists of actual and projected sales growth, margins, working capital needs, and capital spending needs.
Most of this information is available for the current year and past years from the cash flow and income statements required by GAAP but has to be restated into a form that subtracts net investment—the amount of cash necessary to continue the company's operations and any projected growth—from the cash flow from operations. This permits the calculation of free cash flow for any given reporting period. Free cash flow is the amount of cash actually available to investors, including both equity and debtholders. To calculate the company's value, it is then necessary to discount the free cash flow projected over all subsequent periods by the weighted average cost of capital of the equity and debt investors—that is, the amount they could earn with investments of equivalent risk, weighted according to the relative amounts represented by the debt and equity. For companies with various subsidiaries or lines of business, it is necessary to obtain this data for each of the lines of business because each is likely to have a different projected free cash-flow result.
It is obvious that a lot of this data is not available from the GAAP financial statements, including the statements of cash flow, and most investors would not have the skills necessary to develop and project it into the future. In order to reduce the reliance of investors solely on GAAP earnings, the SEC should encourage companies to provide this information in a usable form. This would not require much additional cost for companies, because well-managed companies use exactly this data to evaluate alternative investments. Nor would it be necessary for companies to predict or project results in order to provide useful data. Alternative scenarios for the most important data could be provided, and investors and analysts could choose among them. The company might also state what the current market price of the company's shares implies about the company's future cash flows. This would give investors a sense of how others are valuing the company. In the Enron case, in particular, investors would have had fair warning that the company's earnings were being generated largely through paper transactions.
Obviously, such disclosures would be forward-looking statements within the meaning of the Private Securities Litigations Reform Act of 1995, but this seems to be exactly the kind of information Congress was intending to encourage companies to produce with that amendment to the Securities Act of 1933. The SEC has done very little to implement the intent of Congress by further defining the scope of the safe harbor the Reform Act created for forward-looking statements, and as a result these statements have become formulaic boilerplate in most company registration statements and other filings. Without better indications from the SEC that the safe harbor is meaningful, companies are understandably reluctant to venture beyond textual statements of belief about their future prospects. The SEC must become more aggressive in encouraging companies to provide the kind of cash-flow projections that can be truly helpful to investors. If these are provided in good faith, and in a range of possibilities, and if they are accompanied by appropriate caveats, they should not give rise to liability.
Some will undoubtedly object that the introduction of projections of this kind will dilute the comparability across companies that results from all companies reporting under the same set of GAAP rules. But it has become clear that this process is not providing investors with the information they need and is not producing useful comparability. GAAP financial reporting is simply too malleable; by making different and entirely reasonable judgments about the future, corporate managements can produce results within a wide range almost at will.
The objective of the SEC should be to ensure that investors get the most useful information possible. If GAAP statements achieved true comparability, they would meet this test, but they do not. Accordingly, if the SEC is to meet its obligations, ordinary investors should not be left with the impression—as they are today—that the audited earnings of a company, prepared under GAAP, are the sole or even the most important measure of a company's value.
Peter J. Wallison is a resident fellow at AEI.
Email this citation