Most empirical research following Merton’s (1977) proposal to value deposit insurance with an option pricing model have used the market prices of bank securities to calibrate the model’s bank risk characteristics. The required model inputs most often include the ratio of the institution’s market value of assets to liabilities, the volatility of the institution’s assets, and in some cases a measure of the institution’s interest rate risk. Studies, such as Marcus and Shaked (1983), Pennacchi (1987b), and Duan, Moreau, and Sealey (1995) show how these model inputs can be calculated using the value and volatility of a bank’s publicly traded common stock, as well as the covariance between its stock returns and changes in interest rates. This is a natural approach for depository institutions having publicly traded shareholders’ equity because stock market values are likely to reflect the economic value of the financial institutions in a forward-looking manner. However, equity market prices are not available for most institutions insured by the Federal Deposit Insurance Corporation (FDIC). While the several hundred publicly held banks and bank holding companies hold the majority of the total deposits of the United States financial system, thousands of institutions are privately held. To apply an option model to privately held financial institutions, an alternative method is required.
This paper develops such a method. It begins by outlining a model that is the basis for valuing the cost of insuring any institution’s deposits. Similar to previous models, its three main bank risk characteristics can be directly estimated only if data on an institution’s market value of shareholders’ equity is available. For privately held institutions, an indirect method is needed. The indirect method we propose involves two steps. The first step uses both supervisory accounting data and equity market data from publicly held depository institutions to explore the statistical relationships between a bank’s financial statement variables and its market derived risk characteristics. The second step then uses these relationships to predict the risk characteristics of private depository institutions based on their supervisory accounting data.