Chapter 4
Insurance Coverage and Financial Operations of the FDIC

The past 50 years have witnessed many changes in the operations of the FDIC. Some have been the result of legislation, while others have been due to the experience gained in providing deposit insurance. In retrospect, the changes have been relatively minor considering the economic climate and the level of experience with deposit insurance prevailing in 1933. This chapter focuses on the changes in the financial and internal operations of the FDIC since 1933.

Financial Operations

Many informed observers in 1933 felt that a system of federal deposit insurance, especially if substantive coverage were provided to virtually all banks, could not remain viable without direct support from the Treasury. The banking crisis of the early 1930s had left the banking system in a weakened condition. There was concern that another banking crisis could result in an accelerating rate of bank failures, and that already low bank earnings would not be sufficient to finance a deposit insurance system. At the same time, the use of tax revenues to finance a deposit insurance scheme was viewed as unacceptable, and in fact formed one of the primary bases for the Roosevelt Administration's opposition to federal deposit insurance.

The concern regarding federal involvement in financing deposit insurance led to an initial organization that closely paralleled a typical casualty insurance company. Because of the weakened condition of the banking system, however, it was recognized that at least some of the initial capitalization would have to be supplied from government sources. It was anticipated, although with some reservations on the part of many, that expenses, losses and future additions to reserves (net worth) would be covered by insurance premiums levied on insured banks and by income from investments.

As discussed in Chapter 3, the 1933 Act provided for two deposit insurance plans: a temporary plan and a permanent plan. Funding to support the temporary plan was provided by an assessment of one-half of one percent of total insured deposits, half of which was payable upon admittance to the program and the remainder subject to call by the FDIC. If this proved to be inadequate to cover expenses and losses, the FDIC had the authority to levy one additional assessment not to exceed the amounts already paid by insured banks. The Act also provided for one reassessment based on changes in insured deposits during the existence of the interim plan.

The financing of the permanent plan was somewhat more complex and potentially very burdensome to the banking system. Basically, the system would have involved an initial capital contribution (capital stock purchase) upon joining the program and an assessment (insurance premium) effectively to pass all insurance losses directly to insured institutions.¹ The basis for both the initial contribution and subsequent assessments was to have been shifted from insured deposits to total deposit liabilities.

During the 20 months that the Temporary Federal Deposit Insurance Fund was in operation, the banking situation improved significantly. Attention thus shifted to the specific insurance provisions of the 1933 Act. Most of those who had originally opposed deposit insurance legislation apparently had been convinced that the existence of the FDIC was a major contributing factor to the drastic reduction in bank failures. However, various provisions of the original permanent plan were viewed as not being appropriate in the new environment.

The banking industry did not like the potential for virtually unlimited assessments and generally felt that the assessment rate should be set at a relatively low level. Large banks took exception to shifting the assessment base from insured to total deposits, contending that they would be unduly penalized because of the relatively large portion of uninsured deposits held in larger institutions. State chartered, nonmember banks objected to mandatory membership in the Federal Reserve System as a precondition for retaining deposit insurance coverage.

For its part, the FDIC was faced with a dilemma. Although the bank failure rate had dropped precipitously and the capital rehabilitation program of the RFC and FDIC had been moderately successful, the banking system was not strong and the prospects for bank earnings were not bright. Additionally, the fears and uncertainties regarding the bank failure rate had not been dispelled by 1934 and indeed would not recede for more than two decades. The FDIC thus was faced with the problems of protecting the earnings of insured banks until capital and reserve positions could be rebuilt while, at the same time, conserving what was by historical standards a modest deposit insurance fund.

During 1934, FDIC staff began drafting what was to become Title I of the Banking Act of 1935. In hearings beginning in February 1935 before the House Committee on Banking and Currency, FDIC Chairman Leo Crowley articulated his plan for the future of federal deposit insurance. In addition to an assessment rate lower than historical experience would suggest, his plan consisted of a combination of stricter entrance standards for new banks and expanded authority over the actions of existing banks, expanded powers regarding the handling of failing banks, a reduction in insurance exposure (i.e., retention of the $5,000 insurance coverage rather than the higher limit envisaged in the original permanent plan) and other provisions that would tend to conserve the deposit insurance fund.² From a practical point of view, the program advocated by Mr. Crowley consisted of attempting to strengthen the banking system, while using every legal means available to conserve FDIC financial resources. This philosophy dominated FDIC behavior until the mid-1960s.

The deposit insurance provisions of the Banking Act of 1935, with few exceptions, were identical to the draft legislation prepared by the FDIC. From a financial point of view, one of the most significant revisions to the original permanent plan related to the calculation of assessments levied on insured banks. The 1935 Act provided that assessments were to be based on a flat annual rate of one-twelfth of one percent of total deposits; the net effect of this change was to shift the relative burden of the deposit insurance system to the larger banks while protecting the level of assessment income to the FDIC. Additionally, the requirement for initial and subsequent capital subscriptions by insured banks was deleted, and the payment of dividends on capital stock held by the U.S. Treasury was eliminated. To provide for emergency situations, the FDIC was given authority to borrow up to $975 million from the Treasury.³

By year-end 1946, the deposit insurance fund (net worth) had increased to over $1 billion. Because of the highly liquid condition of the banking industry, the legislation passed in the 1930s to reduce risks in many sectors of the economy and the recent low bank failure rate, many observers felt that a $1 billion fund was sufficient to cover almost any economic contingency. In fact, three years later, in connection with the Congressional hearings relating to the Federal Deposit Insurance Act of 1950, Jesse Jones, former chairman of the RFC, advocated an effective assessment rate that would maintain the deposit insurance fund at the $1 billion level. Apparently, Congress felt that the fund was adequate at that time and legislatively mandated repayment of the original capital subscriptions. The $289 million initially subscribed by the Treasury and the Federal Reserve Banks was fully repaid by the end of 1948.

Bankers also had voiced concern that the assessment rate was too high. By 1950, the deposit insurance fund had reached a level of over $1.2 billion, despite the repayment of capital completed two years earlier. Assessment income had been growing at a high rate, reflecting the rapid growth in bank deposits during the World War II and post-war years. Moreover, because of low interest rates during this same period, bank earnings lagged increases in prices and deposit insurance expenses.

On September 9, 1947, FDIC Chairman Maple Harl (right) presented to Under Secretary of the Treasure A. L. M. Wiggins a check for $146 million, repaying more than half of the governemnt's initial funding of the FDIC. The balance ware repaid in 1948. Photo: Harris & Ewing

The FDIC was reluctant to support a permanent reduction in the basic assessment rate. There still was concern that accumulated earnings would be insufficient to handle the increased rate of bank failures that many thought would occur during the 1950s. This fear was reinforced by the decrease in capitalization of the banking industry due to low earnings and rapid asset expansion since 1940.

As a compromise, deposit insurance charges were effectively reduced by the Federal Deposit Insurance Act of 1950. Rather than lowering the basic assessment rate, however, the reduction was accomplished through a rebate system. After deducting operating expenses and insurance losses from gross assessment income, 40 percent was to be retained by the FDIC, with the remainder to be rebated to insured banks. This procedure meant that losses were to be shared by insured banks and the FDIC on a 60 percent - 40 percent basis. This provision has tended to stabilize FDIC earnings during periods of fluctuating loss experience.

The 1950 Act also required the FDIC to reimburse the Treasury for interest foregone-on the initial capital contributions. This requirement was the result of an exchange between FDIC Chairman Maple T. Harl and Senator Paul Douglas of Illinois during hearings on the 1950 Act. The exchange went as follows:

Senator Douglas: . . . Mr. Harl, on page 2 [of your prepared statement] you speak of making final payment to the Treasury on August 30, 1948, when you paid the Treasury out in full for the loans [capital] which were advanced. Do I understand that to be your statement?

Mr. Harl: We paid them for the money advanced.

Senator Douglas: Would that include the interest upon the Goverment loan which was made?

Mr. Harl: It did not. The law provided that there should be no dividend upon the capital stock.

Senator Douglas: In practice, the Government has made an advance to the FDIC which has not been repaid; namely, the interest on the bonds which the Government issued, but for which it was not reimbursed. . . .

Mr. Harl: . . . This Corporation stands ready to reimburse the Government, or anyone else, provided it is legally authorized to do so. Senator Douglas: You are ready to pay the interest, is that right?

Mr. Harl: Yes. If we have an obligation we are ready to pay it.

Senator Douglas: That is a possible source of revenue that I had not thought of. This brief conversation, which I at first thought was going to be unprofitable, might yield the Government as much as $40,000,000. I first thought it was love's labor lost. It may turn out that there was gold in "them there hills."⁴

During 1950 and 1951, the FDIC paid about $81 million to the Treasury for the interest foregone on the initial contribution of both the Treasury and the Federal Reserve banks.⁵

The 1950 Act also removed the law governing FDIC operations from the Federal Reserve Act, and created a separate body of law known as the Federal Deposit Insurance Act. Although of only symbolic significance, this change over the years has reinforced the FDIC's separate identity.

To compensate certain banks for the effect of a technical change in the computation of the assessment base, net assessments were further reduced in 1960, when the rebate percentage was increased to 66 2/3 percent. In 1980, the basic percentage was lowered to 60 percent, with mandatory adjustments to be made if the ratio of the deposit insurance fund to estimated "insured" deposits were to exceed 1.40 percent or were less than 1.10 percent. The FDIC sought this latter provision to help rebuild the fund if abnormally high losses were experienced, and to inhibit excessive growth of the fund in periods of low losses.

Income and Expenses of the FDIC

The major sources of income to the FDIC have been assessments collected from insured banks and interest on its portfolio of U.S. Treasury securities. In recent years, interest on capital notes advanced to facilitate mergers and deposit assumption transactions and to assist open insured banks has become an increasing, although not major, source of income.

Expenses incurred by the FDIC are normally grouped into two categories. Administrative expenses include expenditures not directly attributable to bank closings and the subsequent liquidation of assets. The other major expense category, insurance expenses and losses, includes expenses associated with bank closings, liquidation activities and the FDIC's share of losses on acquired assets.

Table 4-1 presents the major income and expense items for each year since 1933. For over half of this period, assessments accounted for the largest share of income to the FDIC. However, continued favorable loss experience allowed the securities portfolio to grow so that, in 1961, investment income exceeded assessments. This relationship has continued since that time and, absent abnormally large cash demands or drastic reductions in interest rates, the relative importance of interest income probably will increase.

In addition to the absolute size of the securities portfolio, investment income also is sensitive to the interest rate environment and the investment strategy followed by the FDIC. This phenomenon first became apparent in the mid-1960s, when market rates started to exhibit some degree of short-term instability.

In the mid-1970s, the FDIC started to pursue an active role in managing its investment portfolio;⁶ prior to this time the FDIC had assumed a passive role and, in essence, allowed the Treasury to invest the funds in whatever issues it felt appropriate. About this same time, the FDIC started to shorten the average maturity of its portfolio and generally to achieve a better maturity balance. As the earnings problems faced by mutual savings banks became more apparent, the FDIC sharply reduced the average maturity of its portfolio in anticipation of large cash needs and as a hedge against rising interest rates. While the need for the amount of liquidity originally envisaged never materialized, a highly liquid position, coupled with historically high short-term interest rates, resulted in extraordinarily high earnings from investments and helped to offset unprecedented insurance expenses during 1981 and 1982.

Assessment income has paralleled the growth of deposits in the banking system. The assessment rebate system adopted in 1950 has resulted in a lower level of assessments being retained by the FDIC. In most years since 1950, the FDIC has retained slightly in excess of 40 percent of gross assessment income. In 1981 and 1982, however, the large insurance losses resulted in retention of about 90 percent of gross assessments. Since a sliding scale of rebates was mandated in 1980, the ratio of the fund to insured deposits has remained within the statutory limits and the rebate has remained at 60 percent of net assessment income.

Administrative expenses of the FDIC have grown roughly in proportion to changes in the price level and staffing requirements.⁷ The one exception occurred in 1976, when substantial losses ($105.6 million) on sales of securities were realized in connection with the shift in investment strategy mentioned earlier. Normally, gains and losses on securities transactions are considered to be part of interest income; however, this loss (and a smaller loss realized in 1978) was incurred as a result of a change in operating procedures, and it was decided at the time that the loss was more appropriately an operating expense.

Insurance losses and expenses are related to the number and size of banks requiring financial intervention by the FDIC. Periodically, the expected loss to the FDIC from each active closed bank or assisted merger case is revaluated, and adjustments are applied to the appropriate loss reserve and expense accounts. For accounting purposes, the adjustments are combined with current year losses, and the net is charged to insurance expense. This practice can result in a misleading impression, and can compound the difficulties experienced by readers of FDIC financial statements. Perhaps the best example of the magnitude of the distortion that can occur is the insurance loss of $100 million reported by the FDIC in 1974. Essentially this entire amount was attributable to a revision to the expected loss on the United States National Bank (San Diego) failure that had occurred the previous year. Again in 1982, reported losses included a $158 million reduction in losses associated with assisted mergers of mutual savings banks during 1981. The negative losses reported by the FDIC in 1979 and 1980 also were the result of revisions to original cost estimates.

Table 4-2 presents a summary by year of the number and total assets of failed insured banks, and the losses realized by the FDIC in connection with these failures. Because of the periodic revaluation of loss estimates, the losses reported for accounting purposes (Table 4-1) cannot be traced easily in this table.

Another source of distortion arises from the FDIC's past policies with respect to explicit interest charges on funds advanced in connection with insurance operations. The policy has been not to adjust cost estimates to reflect foregone interest, and this has significantly understated reported losses. Beginning in 1983, the FDIC changed its policy so that explicit interest will be factored into all future cost estimates.

The FDIC's practice of not allocating administrative costs to insurance expense also has tended to understate reported losses. In 1984, the FDIC will begin allocating overhead expenses to each failed bank receivership.

The understatement of historical costs notwithstanding, the loss experience of the FDIC has been modest. A majority of failures of insured banks (360) occurred before World War 11, resulting in reported losses slightly less than nine percent of assessments collected over this same period. It was not until the mid-1970s that losses again approached and surpassed this level.

The Deposit Insurance Fund

The deposit insurance fund is the net worth of the FDIC, and represents accumulated earnings retained since 1933. In every year except 1947, when the FDIC retired a majority of the capital stock originally issued to the Treasury and Federal Reserve Banks, the fund has increased and was approximately $14.3 billion in mid-1983.

The fund is often compared to various definitions of deposit liabilities in insured banks in an attempt to measure its ability to absorb losses in the banking system. The relationship that probably has received the most attention is the ratio of the fund to total insured deposits. As a practical matter, however, the concept of an aggregate level of insured deposits has little meaning.

Since the mid-1960s, the FDIC has handled most failed banks in a way that all depositors, and indeed all general creditors, have been afforded de facto 100 percent insurance.⁸ It is only in cases where the FDIC pays off the depositors of a failed bank that the basic insurance limit becomes relevant. However, even in the case of a payoff, many uninsured depositors are either collateralized or have an offset against an outstanding credit. Thus, the ratio of the fund to insured deposits probably represents an underestimate of the exposure of the fund.

Additionally, the measurement of total insured deposits within the system with any precision has become extremely difficult, if not impossible. The complexities in the law pertaining to the definition of deposits, the method of aggregating individual depositors' accounts within a bank for insurance purposes and the increased activity of brokers, who specialize in gathering funds from many individuals and placing them in fully insured deposit accounts, all contribute to measurement problems.

In Table 4-3, the ratios of the fund to both insured and total (domestic) deposits are presented. Although there have been some fluctuations in these ratios, they have remained remarkably stable over time. This is a reflection of the ability of the FDIC to generate sufficient income to cover operating expenses and insurance losses, and to contribute enough to the fund to maintain a stable relationship to deposit liabilities. Even in 1981-1982, years when record losses were absorbed by the FDIC, the fund increased both in absolute terms and in relation to total deposits.

There are several reasons to believe that the historical relationship of the fund to deposits will continue into the future. Market interest rates tend to move with bank deposits. Over the past 25 years, interest rates on three-to five-year Treasury securities have increased at an annual average compound rate of one to one-and-one-half percent less than deposits in the banking system. While this same relationship has not been constant over time, it is probable that the positive correlation will continue into the future. Whatever the shortfall of interest income, retained assessment income is the other source available to stabilize the ratio of the fund to deposits. The magnitude of this income depends importantly on the volume of insurance losses.

In general, losses incurred by the FDIC in connection with failed banks have been modest. From 1934 to 1980, reported losses and insurance expenses accounted for less than five percent of assessment income. The record losses reported in 1981 and 1982, when losses accounted for approximately 74 percent of assessment income, are not expected to continue over any protracted period of time. While future losses may be higher than those experienced through 1980, losses even greater than the more recent levels would have to persist for several years before the ability of the fund to generate substantial income would be compromised. Although 1981 and 1982 cannot be considered to represent a normal period, it must be recognized that the fund grew by about 25 percent during this period despite the enormous losses absorbed by the FDIC.