Fair-value accounting, in which the assets and liabilities of firms are reflected in financial reports at a price that would be received to sell an asset or paid to transfer a liability in an orderly transaction — often very close to their market value — has been increasingly popular over the last two decades, particularly in the banking industry. As the use of fair-value accounting has increased, so has criticism of the approach — criticism that has grown acute in the context of the financial crisis as credit markets constricted and formerly liquid assets dried up.
Fair-value accounting bases the reported value of any given asset (or liability) on the current market value of that and similar assets. This means that values move, to a large degree, with the market. At the other end of the accounting spectrum is historical-cost-based accounting, in which asset valuation is based on the asset’s historical transaction prices. Disregarding other benefits and detractors of each approach, historical cost is simply not subject to the same sensitivity of day-to-day market variations as fair-value prices are.
When markets become illiquid, fair value can set off a chain reaction: banks are forced to sell their assets at prices far below their fundamental values (the present value of expected future cash flows), and other banks with similar assets can be forced to mark those assets at a lower price. As prices for the asset class drop, the falling prices have the potential to trigger regulatory constraints — such as increased capital requirements — and otherwise interact with internal control mechanisms — such as value-at-risk (VAR) — or managerial incentives, such as how pay levels for executives are set. All the while, prices continue to fall in a vicious downward spiral.
With the rise of fair-value accounting, particularly in the years leading up to the financial crisis, it is natural to ask whether fair value is associated with systemic or system-wide risk in the banking industry. Professor Urooj Khan considers this question in a new paper, and looks for evidence of whether that association contributed to the financial crisis.
Using financial reporting data from bank holding companies, which publicly traded banks are required to file with the Federal Reserve, and security prices from the Center for Research Security Prices (CRSP), Khan looks at how the use of fair value in financial reporting has evolved from 1998 to 2008. The 20-year period coincides with the Basel I Accord, an internationally established set of common capital requirements for banks — allowing Khan to compare the accounting practices of banks operating in the same regulatory environment.
Khan first measures the amount of assets and liabilities reported at fair value, scaled by total assets in the banking industry, to determine the extent to which fair value is used in financial reporting. He then looks at whether bank contagion — spread of market shocks, especially on the downside and observed through co-movements in stock prices — is more common when fair value is in greater use, and finds that it is.
“But this only happens on the downside. I find that when markets are liquid, assets don’t seem to get sold below their fundamental value and contagion does not spread, as it does in illiquid markets.” In other words, it’s not that fair value causes contagion regardless of whether markets are liquid or illiquid but that fair value is more sensitive in bad times than in good times.
Prior research has found that using fair value might be a more timely and relevant approach for investors, but when markets are illiquid, the additional cost associated with fair value is increased contagion in the banking sector. That suggests a rationale for careful interpretation of fair-value financial reporting during periods of market illiquidity.
Khan also found that banks that are poorly capitalized or that have a higher proportion of assets and liabilities reported at fair value are more likely to be affected by the increased bank contagion associated with fair value.
As the financial crisis continued into 2009, some suggested that fair-value accounting rules should be suspended. But many users in many contexts depend on accounting information to inform their decisions. “Maybe the answer is not to suspend fair-value accounting rules but rather to understand the additional cost associated with fair value,” he suggests. “So when regulators, banks, and investors are taking into account the costs and benefits of a more fair–value-based regime and assessing what future accounting standards should look like, they should be prepared to consider that at some points in time fair-value accounting may not be conveying the most important information.
“I am not trying to document the superiority of historical cost based accounting or fair value,” Khan emphasizes. “I am trying to show there are unintended consequences and costs associated with fair-value accounting.”
Urooj Khan is assistant professor of accounting at Columbia Business School.
Watch Professor Khan discuss this research at the Program for Financial Studies’ No Free Lunch Seminar Series.
Urooj Khan is the Class of 1967 Associate Professor of Business at Columbia Business School. He is also the Assistant Director of Program of Financial Studies’ initiative on the future of banking and insurance and a faculty advisor to the India Business Initiative. Between 2014 and 2015, he was a part-time researcher with the U.S. Department of the Treasury’s Office of Financial...
Read the Research
"Does fair value accounting contribute to systemic risk in the banking industry?"