Accounting Reform

Accounting is often seen as a veil, a mere detail of measurement that does not affect the economic fundamentals of a firm. However, the intensity of the public debate surrounding accounting reforms in recent years suggests that there may be more at stake than obscure debates on measurement. Currently, attention is focused on the initiative of the International Accounting Standards Board (IASB) and the U.S. Financial Accounting Standards Board (FASB) to create global accounting standards based on market prices.

Beginning in the late 1990s and gathering momentum in the wake of accounting scandals at Enron and WorldCom, regulators have pushed for firms to record all assets and liabilities on their balance sheets at market values, a system referred to as "mark- to- market" accounting. Under traditional accounting methods, a company's assets are recorded at the prices at which they were originated and these prices are not adjusted to match current market prices, a system referred to as "historical cost" accounting.

In the study "Marking-to-Market: Panacea or Pandora's Box?", University of Chicago Graduate School of Business professor Haresh Sapra, Guillaume Plantin of Carnegie Mellon University's Tepper School of Business, and Hyun Song Shin of Princeton University present an economic analysis of the costs and benefits of the mark-to-market approach on financial institutions.

"Switching from historical cost accounting to mark-tomarket is not simply a case of switching between equivalents, such as measuring temperature in Celsius or Fahrenheit," says Sapra.

Under historical cost accounting, financial institutions may appear too conservative because their books do not reflect the appreciated value of their assets quickly enough. As a result, financial institutions may not necessarily have an incentive to participate in the most profitable projects.

Under the marking-to-market system, a company's assets and liabilities are recorded at current market prices. In theory, a measurement system that reflects the market values of assets and liabilities should provide better insights into the risk profile of firms. Such a system would offer investors greater transparency, which is in line with the goals of the Securities and Exchange Commission.

If market prices reflected only the fundamental values of all assets and liabilities, then the superiority of the markingto- market system would be undeniable. Unfortunately, when there are imperfections in the marke t, the superiority of the marking-to-market measurement system compared to the historical cost system is no longer so clear-cut.

The authors studied the consequences of the accounting measurement system on the market price of the asset and then investigated the impact of the asset's price on a financial institution's incentives to invest in this asset.

The central debate in current accounting reform is the extent to which marking-to-market injects volatility into financial markets, and therefore into asset prices. The authors show that market prices can serve as a double-edged sword in the economy. Market prices reflect the fundamental values of assets through the forces of supply and demand. As these fundamentals change, market prices change, which leads to price volatility. However, market prices also affect market outcomes through their influence on the actions of the firm.

The authors show that for certain assets of financial institutions, marking-to-market may lead to asset allocation decisions based on second-guessing of other firms' decisions rather than decisions based on perceived fundamentals. There is the danger of an additional source of price volatility that is purely a consequence of the accounting standards rather than volatility reflecting changes in the underlying fundamentals.

"When a firm tries to extract information from prices for a very illiquid asset under the marking-to-market system, the firm naturally becomes concerned about whether other firms will sell or hold the asset," says Sapra. "If a firm believes that others will sell and all firms have the same concerns, then those concerns can become self-sustaining. This strategic effect arising from marking an asset to market is the point that the FASB is missing."

One Accounting Method Does Not Fit All

Using their economic model, Sapra, Plantin, and Shin show that the choice of accounting measurement policy significantly alters a firm's reported earnings. Their model centers on the decision of a manager of a financial institution, such as a bank or insurance company, where the manager aims to maximize the interim earnings of the firm.

The earnings reported in quarterly and annual reports are typically the basis for executive compensation,especially in the financial services industry,so there is a natural incentive for the manager to maximize interim earnings. The manager must decide whether to sell a given loan or to carry this loan on the balance sheet using the prevailing accounting standard.

A manager who has to opt for one of these two measurement regimes is caught between the horns of a dilemma. On one hand, historical cost accounting makes too little use of the information contained in prices on the asset market and relies too heavily on outdated historical cost. On the other hand, in trying to extract the informational content of prices, which reflect the supply and demand for an asset, marking-to-market distorts the information by adding extraneous risk.

Marking-to-market may result in excess volatility because gains and losses to a loan on the market will flow into the firms' equity and earnings numbers and make the asset seem more volatile.

At the firm level, marking-to-market mechanically creates "artificial" volatility on the firms' balance sheet, simply because its books are updated more frequently. Based on their model, the authors suggest that this extraneous effect impacts firms' trading behavior, thereby modifying the asset prices and injecting further volatility in the market for the asset. If banks look risky in the short-run, the bank may decide to stop making these types of illiquid long-term loans, such as mortgages. Because most banks invest in similar classes of assets with the same risks, switching to marking-to-market can create volatility in asset prices if many firms decide to sell the assets.

"If all assets are very similar across banks, and market prices are unreliable, many banks may start selling off their assets because they don't want to be holding worthless assets after prices start dropping and put themselves at risk for bankruptcy," says Sapra.

The authors found that by trying to extract information from prices, marking-to-market may create "beauty contests" in which financial institutions become concerned with selling their assets because they are concerned that other firms will do so first.

The Role of Financial Institutions

Sapra, Plantin, and Shin's results show why opposition to marking-to-market accounting has been led by the banking and insurance industries. For these financial institutions, a large portion of their balance sheets consists of precisely the assets and liabilities with cash flow claims that have a long duration, are relatively illiquid, and are senior in nature.

Senior claims by a financial institution have a relatively larger downside risk but a limited upside risk. Conversely, junior claims have a relatively larger upside risk but a limited downside risk.

A large proportion of banks' senior claims appear on the asset side of their balance sheets. For insurance companies, the focus is on the liabilities side of their balance sheet. Insurance liabilities are long-term, illiquid, and have a limited upside from the insurance company's point of view.

"Marking-to-market could take apart the whole foundation of banks and insurance companies," cautions Sapra. "These industries would be forced to hold shorter-term loans to look better on the books, but their role in the economy is that of liquidity transformation,to make long-term illiquid loans and have short-term deposits as liabilities."

The Future of Accounting Regulation

At present, marking-to-market has been limited to hedge funds that deal only with very liquid claims,a small segment of the financial sector. Implementing the proposed changes has been hindered by the lack of reliable prices for all assets. However, concerns over unreliable data will soon change due to the development of new markets in credit derivatives that remove the practical barriers to marking loans to market. The most significant change will come when the markingto- market system is applied to loans and other previously illiquid assets.

Sapra, Plantin, and Shin's results suggest that a careless, rapid shift to a full mark-to-market system may be detrimental to financial intermediation and therefore to economic growth.

However, the authors note that in the long run, the markingto- market system will become the norm because of the push for transparent accounting information. The problem lies in the timeline for implementation, since liquid markets do not exist for all assets. New derivative markets may need to be developed, along with new types of securities and loans. Switching to an entirely marking-to-market system also may require changing the way accounting numbers are tied to executive compensation, and companies may need to create new incentive arrangements.

In a world where informational frictions limit the scope for arms-length contractual relationships, accounting numbers serve an important certification role. These audited numbers carry quasi-legal connotations that can serve as the publicly verifiable quantities on which to base contracts, including the compensation contracts for managers. As such, accounting n umbers serve as a justification for actions. If decisions are made not only because the underlying fundamentals are right, but because the accounts provide the external validation to take such decisions, then accounting numbers take on greater significance.

"Corporate governance is much more than how the board is organized. It also encompasses how the manager's behavior is affected by external accounting rules," says Sapra.

While these new regulations are still being developed, Sapra questions the usefulness of a one-size-fits-all measurement rule. Sapra argues in favor of a hybrid system, as currently used in the United States, whereby the marking-to-market system is used for very liquid, short-term assets, and illiquid assets are still measured by historical cost.

"The most important lesson from our paper is that accounting ceases to be just a veil if real economic decisions are influenced by how quantities are measured and disclosed," says Sapra. "While good corporate governance and mark-tomarket accounting are often seen as two sides of the same coin, regulators must understand that simply moving to a mark-to-market regime without addressing the other imperfections in the financial system need not guarantee a welfare improvement."