Deposit Insurance and the Effective Pricing of Risk: Regulators Could Mitigate Institutions' Excessive Risk Taking by Adopting a Risk-Sensitive Approach to Pricing Deposit Insurance Premiums. Those Premiums Would Have to Reflect the Complex Risks Associated with Trading and Off-Balance-Sheet Activities as Well as Traditional Credit Quality Considerations

Article excerpt

A basic principle of economics is that underpricing stimulates demand. A recent example was the huge surge in subprime mortgage lending as banks and nonbanks raced to originate loans for inclusion in the flood of collateralized debt obligations. In the process, interest rates were reduced (admittedly for a temporary term), and longstanding limitations on loan-to-value and income-to-value ratios were relaxed. The point was even reached where NINJA (no-income-no-job-or-assets) loans became a target of Internet humor.

Certainly, the recent episode has been extreme and was justified by a sense that the assets being originated would be held on the books only for a temporary period. (1) Nevertheless, some analysts say that banks have underpriced credit for many years, even--or perhaps especially--when it is expected to be held on the books to maturity.

How have banks been able to price credit so poorly and remain viable? One contributing factor is that the large retail deposit funding of banks is not competitively priced. In part, this pricing is the result of deposit insurance that underpins public confidence in the safety of these deposits. Under normal circumstances, this pricing allows banks to borrow at rates below those that would be consistent with default risk. (2) Of course, an additional cost of retail deposits is the insurance premium banks must pay to the FDIC.

Deposit insurance is arguably one of the great success stories of modern banking legislation. It has substantially reduced periodic bank runs that proved so disruptive to financial markets and the general economy before it was introduced. (3) Like all such successful innovations, however, deposit insurance has its drawbacks. The most serious is the moral hazard that arises when deposit insurance premiums are not sufficiently sensitive to differing risk levels across banks. This encourages excessive risk taking on the part of banks, as the owners reap the upside rewards while avoiding the full impact of downside losses. (4)


Regulators could mitigate such destructive--and potentially costly--behavior with a risk-sensitive approach to pricing deposit insurance premiums. Without such risk-sensitive pricing, bank shareholders benefit from society's willingness (in the limit) to bail out depositors, thereby reducing the return required to attract such deposits.

Despite its importance, setting appropriately risk-sensitive deposit insurance premiums is a daunting task. It is further complicated by the rapid evolution of banking in recent years. Certainly for banks that have an active market-making activity, a comprehensive risk analysis is required. Even many smaller banks, however, have moved toward an originate-and-distribute model, making their balance sheets far more dynamic than under the traditional originate-and-hold regime. In this context, truly risk-sensitive deposit insurance premiums would require considerably more detailed data-gathering and analysis than is reflected in current practice.

Ultimately, regulators need to develop granular estimates of expected and unexpected losses for each bank based on underlying risk pools and the additional risks associated with trading activities and securitization. This would require:

* New and more detailed data to be collected from financial organizations that goes beyond that available from bank call reports.

* A series of models appropriate for banks of different sizes and compositions.

* An adjustable scale of deposit insurance premiums tied to risk.

In effect, such risk estimates would have to become part of the regular examination process. Practical considerations mean that the complexity of this detailed risk assessment process would vary across banks, based on the potential systemic impact of their failure.

Such risk estimates would require reporting, at least for comparative purposes, of much broader fair-value estimates of both assets and liabilities. …