Fair-value accounting (FVA), also known as “mark-to-market” accounting, has been blamed for contributing significantly to the current financial turmoil. But those charges are largely unfounded, according to new research by Christian Leuz, a professor of international economics, finance and accounting at the University of Chicago Booth School of Business, and Christian Laux of Goethe-University in Frankfurt.
The conclusions that people draw about the role of FVA in the crisis are important as they will shape regulations and accounting standards that are beginning to emerge from the financial crisis. Recently, the Financial Accounting Standards Board announced a proposal to tighten FVA, which would lead to more instruments being recognized at fair value on the balance sheet. Many in the financial sector, by contrast, have pled for more relief from FVA, arguing that it led to excessive sales of assets as their market prices fell. Lawmakers in Congress have also taken a keen interest in this issue.
Yet in their study, “Did Fair-Value Accounting Contribute to the Financial Crisis?,” Laux and Leuz analyze the role of FVA for U.S. banks and conclude that there is no evidence that fair-value accounting added to the severity of the current financial crisis in a major way.
One reason that the impact of FVA is more limited than often claimed is that the treatment of financial assets under U.S. accounting standards (GAAP) is more flexible than many in the public realize.
Laux and Leuz stress that current FVA standards provide several circuit breakers that allow reported values to deviate from market prices under important circumstances. For example, instead of using fire sale prices, the FVA standards allow banks to value assets using fundamental cash flow models and unobservable inputs (Level 3 assets) as markets become illiquid. Thus, the argument that FVA forces banks to mark to market asset prices even when they are distorted is misleading.
Using U.S. banks’ financial statements, Laux and Leuz provide evidence that, as the crisis deepened, banks dealt with assets that had been valued using quoted prices in active markets (Level 1 assets), or that had been valued using observable inputs from similar assets (Level 2 assets), by reclassifying them as Level 3 assets and valuing them according to what their models said made sense. This switch from “mark-to-market” to “market-to-model” was consistent with the flexibility that is built into the FVA standards, and allowed banks to avoid reporting their assets solely according to market prices as those prices plunged during the crisis. In fact, many mortgage-related securities that later turned out to be toxic were never reported as Level 1 assets to begin with. For them, the notion of pure mark-to-market accounting is largely a myth.
There also seems to be little evidence of excessive write-downs, or that fair values reported by banks were systematically too low. Laux and Leuz point to a number of studies that compare banks’ reported fair values of assets with the market prices of those assets, to find out whether FVA forced banks to report figures that were artificially low. A consistent finding in these studies is that investors priced a dollar of reported assets valued during models (Level 3 assets) significantly below a dollar of reported assets valued by quoted prices in active markets (Level 1 assets). This finding suggests that the use of “mark-to-model” accounting allowed banks to report Level 3 asset values that, if anything are too high, rather than too low, compared with investors’ valuations.
On the whole, the evidence presented by Laux and Leuz suggests that, far from being a main contributor to the crisis, FVA rules actually allowed banks to use discretion to keep values high. Rounding out the analysis, they also point to similar evidence for other assets on banks’ balance sheets, notably loans and goodwill.
The important conclusion that emerges is that, although lenient standards may help to avoid contagion effects in time of crisis, relaxing the rules and giving bank management flexibility in interpreting FVA standards opens the door for manipulation. It can also have negative effects before a crisis hits. Making banks recognize losses early on forces them to take steps to fix problems quickly, and may even encourage them to be more cautious in the first place.
In short, besides finding evidence the FVA did not contribute negatively to the financial crisis as many critics have claimed, Laux and Leuz argue that any negative efforts from fair-value (and other) write-downs during a crisis should be weighed against the desirable effects that FVA and timely recognition of losses can have on banks’ all-around behavior.