A Stanford professor argues that the less that investors use fair value accounting to value companies, the better.
As investors learned the hard way during the financial crisis, we live in a much faster financial world than the one that spawned historical accounting principles. The value of an asset — say, a mortgage-backed security — can fall off a cliff in a matter of days. When that happens, the most recent quarterly income statement can seem like ancient history.
Prior acknowledgment of this reality had resulted in the Financial Accounting Standards Board’s Statement of Financial Accounting Standards 157, Fair Value Measurements (now known as ASC-820). The standard, which melded a scattered array of pronouncements into a single guidance, became effective for fiscal years beginning after November 15, 2007 — just in time for the fall of Lehman Brothers in September 2008.
Under fair value accounting, companies are now required to estimate the value of their assets and liabilities based on the price they could attract today, rather than what they originally cost. To be sure, such reporting is likely to have given stakeholders a more transparent picture of what was going on as huge financial institutions saw previously highly valued assets burning away during the crisis.
As could be expected, however, the reaction against fair value accounting started almost immediately. Bankers were in the front ranks of the protesters, claiming that the need to report assets at market prices when no markets for them existed actually made things worse by triggering fire sales. As time went on, traditionalists grumbled that growth in the use of fair value was interjecting too many estimates and predictions into what they considered the hard-and-fast science of accounting.
“As an investor, when I turn to financial statements, I want a trustworthy and interpretable account of what took place,” Charles Lee, a professor of accounting at the Stanford Graduate School of Business, told the school’s newsletter in July. “As soon as we start to anticipate future exchanges, we are in a world of speculation. And unfortunately, given dysfunctional managerial incentives and other moral hazard problems, it is often a world of fiction.”
Indeed, in a December 2013 keynote address delivered at an accounting conference in London, Lee made the case that the less investors rely on a company’s fair value to measure what their stocks are worth, the better.
True, he acknowledged, a standard definition of a company’s equity value is the “present value of expected future payoffs to shareholders,” a definition that sounds a whole lot closer to fair value than it does to historical accounting. And “valuation involves forecasting,” Lee stressed. Investors “need to predict future cash flows, dividends, and discount rates. In fact, the essential task in valuation is forecasting.”
But accountants should be extremely wary of getting into the business of prediction, Lee said. And fair value, which asks accountants to confirm corporate estimates of the market value of an asset or a liability even if no market exists for them, turns them into crystal-ball gazers rather than the historians of corporate finance they’re intended to be.
To be sure, the historical fact provided by traditional accounting isn’t in itself a reliable guide to future performance. On the other hand, predictions of corporations’ value based on expectations of future cash-flows are “subjective, speculative, quasi-educated guesses about the future,” Lee said in his speech.
A Language for Forecasting
So how can accountants and investors measure the value of a firm in a way that goes beyond guesswork, yet isn’t stuck in the past? Lee’s answer: cautious prognostication reined in by the strictures of historical accounting. In his speech, Lee outlined three important reasons why historical accounting numbers (and the accounting systems that generate them) are crucial to the task of valuing companies:
• Historical accounting numbers provide “a language for forecasting.” Because they represent a good measurement of corporate transactions within a given period, for example, GAAP earnings provide a solid accounting model for forecasting future earnings. The historical method of reporting profits can help “because it specifies what is to be forecasted, directs you to the information needed to make the forecast, and shows you how to convert a stream of expected payoffs into a value estimate,” said Lee.
• Historical numbers provide integrity to the forecasts by providing a means to confirm their accuracy after the fact. “Because accounting systems provide a structure for expressing what happened in a given time period, they impose discipline on market participants engaged in making forecasts about these time periods,” Lee explained. “Today’s earnings forecasts have credibility only because they can be compared to the actual (and audited) numbers reported in the future.”
• Accounting information itself can be useful in making forecasts. In the footnotes to their financial reports, for instance, companies provide disclosures of such forward-looking information as contingencies, off-balance-sheet assets and liabilities, and information regarding the prospects of various lines of business.
In the wake of the financial crisis, the likelihood of stemming the growth of the use of estimates, forecasts, and fair-value accounting is slim to none. Corporate transactions are simply moving too fast to be captured by the accounting language of the past. At the same time, however, valuation will always require the springboard of solid historical accounting. Perhaps, as Lee suggests, it can take some of the guesswork out of forecasting by revealing the ways in which history can repeat itself.
Source: www.cfo.com
