ICBA - Publications - FDIC Options Paper on Deposit Insurance: The Pricing Issue

FDIC Options Paper on Deposit Insurance: The Pricing Issue

WWR ARTICLE
SEPTEMBER 8, 2000

 

Editor’s note: This is the first in a series of articles that will take a more in-depth look at the three major issues raised in the FDIC’s options paper on deposit insurance reform. The FDIC would welcome banker comments on these issues, as would we.

“The current deposit insurance system has several features that work against the effective and equitable functioning of the system,” the FDIC said in the introduction to the options paper on deposit insurance reform issues it released last month. One is that the current pricing system creates “inappropriate incentives and raises fairness issues.” For example, currently 93 percent of all banks and thrifts (more than 9,500 institutions) pay no premiums. The result, the paper notes, is that not only is the rate paid by vastly disparate banks identical, but the dollar amount as well. A bank with $100 billion in deposits and a complex risk profile is billed the same as the smallest and most conservatively run community bank.

Institutions in the 1A premium category with higher risk profiles cannot be charged higher premiums under the current system as long as they remain well-capitalized, and maintain a CAMELS 1 or 2 rating. Yet, in an economic downturn, many of these institutions would deteriorate faster than others, the paper notes. If deposit insurance were priced according to risk, it is likely that every bank would pay something, the FDIC posits, for the same reason every bank pays at least some spread over Treasuries for unsecured debt. The FDIC’s basic argument is that the current system is not sufficiently risk-based, despite the legislative changes in the 1980s and 1990s mandating a risk-based system.

Also, because of the zero premium in the 1A category, new or rapidly growing banks can add exposure to the FDIC funds without paying assessments. Over 840 banks chartered in the last five years have insured deposits totaling $44 billion, but have never paid a penny in premiums. All institutions are eventually assessed to cover deposit growth at the fastest growing 1A-rated institutions. And, in a downturn, assessment rates would be raised earlier or in a greater amount to make up for the dilution in the reserve ratios of the BIF and SAIF attributable to unfunded insured-deposit growth.

Merrill Lynch’s two subsidiary banks are the prime examples of rapidly growing banks that are paying no premiums. Sweeps from Cash Management Accounts into the two banks have resulted in insured deposit growth of about $19 billion just this year. Estimates are that the Merrill Lynch banks could grow as much as $100 billion, dropping the reserve ratio of the BIF by five basis points, and bringing the industry closer to mandatory rate increases for all banks.

In light of these issues, the first question posed by the paper is: “How should the FDIC price risk?”

Options for Pricing Risk

The most straightforward conceptual approach to price deposit insurance, according to the FDIC, is for a bank to pay an amount equal to the “expected loss.” An “expected loss” pricing system would reflect risk differences among banks and generate premium revenue ultimately equal to the average long-term losses of the FDIC. The expected loss price is derived from the bank’s probability of default, exposure (i.e., insured or total deposits) and severity (size) of the loss as a percentage of the exposure.

In theory, the paper states, every bank could be assigned a probability of default similar to a credit rating. In the FDIC’s experience, good indicators of likely severity of loss in a particular bank’s failure include the bank’s business mix, loan concentration, and the structure of its liabilities.

Options for better assigning risk ratings outlined by the FDIC include:

Supervisory evaluations. One concern with this approach is that a greater reliance on subjective information and judgment could affect consistency in determining premiums across individual banks. Another issue is the timeliness of supervisory ratings, particularly with longer exam cycles for well-performing banks.

  • Composite CAMELS ratings. Instead of only three different risk categories in the current premium matrix for CAMELS ratings, greater risk differentiation could be achieved by separating the CAMELS 1 and 2 banks into two categories. (Currently they are together in supervisory Subgroup A).

  • Component CAMELS ratings. Component CAMELS ratings could play a greater role in setting premiums. For example, 2-rated banks could be rated “2 minus” if they had a sufficient number of components rated 3 or worse.

  • Risk matrix results. For large, complex banks, the OCC and the Fed assign ratings based on a variety of risk-management factors. Banks with the same CAMELS rating can have substantial differences in their risk ratings that are not currently reflected in the premium pricing scheme.

Other objective factors. Advantages of this approach cited by the FDIC include better risk-related information without subjective distinctions among healthy banks, and better timeliness of information. Disadvantages include lack of qualitative factors such as adequacy of internal controls or quality of underwriting standards.

  • Non-public bank specific information. The FDIC could create a basic package of objective information to differentiate risk profiles among banks. For example, the Canada Deposit Insurance Corporation uses a scoring system based on a variety of risk-related factors to assign premium categories. The factors include capitalization ratios, profitability ratios (e.g., ROA), efficiency ratios, significant credit concentrations, percentage of past due loans, and exam ratings.

  • Call Reports. Banks could be rated based on Call Report data and how certain financial ratios compare either with peers or with an absolute benchmark. Ratios such as high loan concentrations, rapid loan growth and high dependence on volatile liabilities have historically been associated with higher probabilities of failure. Institutions in the 1A premium category show significant differences in a number of these factors. For example, the top ten percent of 1A banks average zero percent non-performing loans, while the bottom ten percent average 3.2 percent. The average charge-off rate for the top ten percent is zero, and 10.2 percent for the bottom ten percent. Volatile liability growth is –41.4 percent for the top ten percent of banks, and 721 percent for the bottom ten percent. The average equity to assets ratio for the top ten percent is 23 percent and 6 percent for the bottom ten percent.

  • Market information such as credit agency ratings or subordinated debt prices. Use of market information by its nature would have to be limited to larger publicly traded banks. Only 750 institutions issue debt or equity instruments that are traded in organized markets. And smaller, privately held institutions do not have credit agency ratings.

    The FDIC does have authority to establish two separate premium systems based on bank size, but it has not exercised it. The FDIC suggests it is time to reconsider the one-size-fits-all approach to premium pricing in light of the growing differences between large, complex, global institutions and privately-held, community-oriented institutions focused on a limited number of core businesses.

The FDIC has posted the options paper, along with a series of questions designed to elicit public comment, on its Web site (www.fdic.gov). Viewers can respond to the questions directly on line. Among the questions posed on pricing are: should 1A-rated Institutions be differentiated more on risk, what information should be used to differentiate the risk profiles of healthy banks, how can Call Report information be used to differentiate risks among banks, and should large banks be required to issue subordinated debt to assist the FDIC in pricing deposit insurance.

Next issue: funding deposit insurance losses and maintaining the deposit insurance funds.