borrow from the United States Treasury. It should probably not be assumed that in the event of difficulties in any one of these banks the disbursement by the Corporation would be equal to the total deposits, or even to the insured deposits. It is more reasonable to assume that if a very large insured bank were to become involved in serious financial difficulties an effort would be made to have its liabilities assumed by another insured bank, or to have the bank reorganized with the help of the Corporation. Apart from the administrative difficulties of direct payoff in such a case is the fact that the closing of a very large bank might result in the nearly simultaneous closing of many banks keeping their correspondent balances with the distressed bank.

Disbursements to date by the Corporation in the cases of banks which have Deen handled on other than a payoff basis have averaged 53 percent of total deposits. This includes principal disbursements, advances for asset protection, and liquidation expenses. If we assume, conservatively, that a disbursement in the case of an exceptionally large bank would be 30 percent of its deposits, it is evident that there is still a considerable concentration of risk to the Federal Deposit Insurance Corporation. Under such circumstances there are three insured banks, the failure of any one of which would more than exhaust the deposit insurance fund. There are six other banks, the disbursement for any one of which would require from two-fifths to one-half of the deposit insurance fund; and there are still six other banks, for any one of which a disbursement equal to 30 percent of total deposits would require over one-fourth of the deposit insurance fund.

Table 6 shows the total deposits of the 25 largest insured banks at the end of 1956 and the disbursements which might be necessary, should any one of these banks become involved in serious financial difficulties. Revealing as these figures are, they do not reflect the fact that a very large commercial bank in serious financial difficulties may be a symptom (or a cause) of fundamental difficulties in the banking system, which difficulties may result in other substantial disbursements by the Corporation.

Size of bank is not the only kind of concentration of risk to which the Corporation is subjected as insurer of deposits. In table 7 the total deposits and insured deposits are given for two groups of banks with a relatively thin capital cushion. The first of these groups includes the insured banks that, on June 30, 1956, had total capital accounts amounting to less than 5 percent of their assets. The deposits of these banks were more than 5 times, and their insured deposits more than 3 times, the amount of the deposit insurance fund.

The other group of banks with thin capital margins, shown in the same table. includes those with total capital accounts amounting to less than 10 percent of "assets at risk." The term, "assets at risk," as used here, is narrowly defined; it excludes not only the types of assets usually excluded in tabulations of "risk assets," such as cash and balances with other banks and United States Government obligations, but also loans that are insured or guaranteed by agencies of the Federal Government. It is evident that a bank with total capital of less than 10 percent of "assets at risk" is in a vulnerable position. A relatively small depreciation in these assets, say 2 or 3 percent, would result in serious capital impairment, while a decline of 10 percent would wipe out the bank's capital. It is to be expected that many of these banks would be among the cases requiring Corporation disbursements in the event of even a minor depression. On June 30, 1956, there were 226 insured banks with total capital less than 10 percent of "assets at risk." The insured deposits in these banks amount to 28 percent more than the entire deposit insurance fund, while their total deposits are more than twice the amount of the fund. The vulnerable position of these banks becomes even more apparent when we consider the fact that the assets not tabulated as “at risk” are not free from risk. Judged by current market values, these and other insured banks have large losses on their holdings of United States Government obligations. These "paper" losses do not show on the banks' books, and will not materialize if the banks hold the obligations to maturity. But whenever a bank that is close to the margin of safety runs into adverse circumstances, it is likely to face the necessity of selling United States Government obligations at market values, thus taking losses on those assets as well as on "assets at risk."

There is no mechanical or automatic method of translating these data on the concentration of bank deposits into a working rule as to the size of deposit insurance fund needed. It could reasonably be maintained that the deposit insurance fund should be at least as large as the amount of deposits insured under the $10,000 maximum in the largest bank participating in deposit insurance.

This would mean a fund equal to about 2.5 percent of all deposits in insured banks, or three times as large as the present deposit insurance fund. However, objection may be made to this figure on the ground that the Corporation is not likely to find itself in the position of paying individual depositors in the case of a very large insured bank in financial difficulties. An alternative proposition, therefore, might be that the deposit insurance fund should be at least equal to the minimum probable disbursement in the event of serious financial difficulties in any one of the largest insured banks. If this is placed at 30 percent of the deposits of such a bank, it means that the deposit insurance fund should be about 1.25 percent of the deposits in all insured banks. The present fund is less than two-thirds of this size.

Needed fund in the event of a wave of bank failures accompanying a deep depression. The United States has had numerous periods of banking troubles when large numbers of banks have failed. Among these, there are three in which seriously depressed business conditions and widespread banking difficulties were more intense and prolonged than the others: the late 1830's and early 1840's, the 1870's, and the early 1930's. Tabulations of the number and obligations of banks that failed or suspended during the first of these periods are not available.

In the 1870's the deposits of the banks that failed in a 6-year period amounted to approximately 7 percent of the deposits of all operating banks at the beginning of the period; and in the early 1930's, in a 4-year period, failed banks had about 13 percent of the deposits of operating banks at the beginning of the period. The losses to depositors on accounts up to $5,000, which we may take as roughly equivalent to the present coverage of $10,000, adjusted for assessments collected from stockholders, are estimated at 2.4 percent of deposits in operating banks for the period of the 1870's, and 2.3 percent for the early 1930's. Table 8 contains data on deposits and losses to depositors in failed banks during the 1870's and the early 1930's.

There is no question that the present deposit insurance fund would be entirely 'inadequate should, for example, a situation similar to that of 1930-33 recur. After a careful analysis we have concluded that in order to make the necessary disbursements in such a situation the Corporation would need to have at its disposal available funds equal, as a minimum, to 5 percent of the total deposits in all operating banks. This figure assumes that the necessary principal disbursements would have been only 37 percent of the deposits in the closed banks in comparison with the Corporation's experience of 50 percent. Since the fund today is only 0.80 percent of total deposits (and with the assured borrowing power only 2.2 percent of total deposits) the inadequacy is obvious. As a matter of fact, it would require all of the present deposit insurance fund plus all of the $3 billion borrowing power to absorb only the losses that would occur in such an emergency.

To what extent can we expect a situation such as that of 1930-33 to recur? Certainly, we can conceive of the possibility of a severe economic downturn, accompanied by large numbers of bank failures. Neither the public confidence engendered by the existence of Federal deposit insurance nor the improvements in banking or bank supervision would be sufficient to prevent these failures, which would be a consequence of economic dislocations of a fundamental nature. However, because the Federal Government is committed, under the Employment Act of 1946, to follow policies which will stimulate full employment, and in view of the knowledge and authority now possessed by various agencies of the Federal Government, it is reasonable to assume that we will be able to avoid the prolongation of a serious depression.

Needed fund to handle contingencies other than deep depressions.-Waves of bank failures are not necesarily connected with deep depressions. They may occur during periods when the economy as a whole is stable or prosperous, because of a set of circumstances affecting a particular region or industry. The underlying causes of such a situation may be deep-rooted so that the failures tend to extend over a number of years. In such a situation there is the danger that the deposit insurance fund will come to consist largely of low quality assets acquired through disbursements to protect depositors.

The most recent such period, and the one for which the most data are available, is that of 1922-29. This 8-year period was a time of great prosperity interrupted only slightly by the mild recessions of 1924 and 1927, which were comparable in severity to that of 1954. Some of the bank failures of those years were due to mismanagement and defalcation of the sort that still produce an occasional failure among insured banks, and some were due to hangover condi

tions from the depression period of 1920-21, similar to some of the failures among insured banks during the first few years of Federal deposit insurance. But for the most part the failures of 1922-29 represented an inability of banks in the agricultural regions of the Nation to adjust themselves to the impact of a set of economic circumstances having an adverse effect on agriculture and on the trading centers of agricultural areas, even though business throughout the nation was generally prosperous.

The argument may be made that the situation that existed during the prosperous period of the 1920's will not recur, on the grounds that the "overbanked" condition of that period no longer exists and that improvements in bank management and bank supervision have made the banks less vulnerable to adverse economic conditions. Yet the fact is that the banks are probably more vulnerable today, rather than less, to adverse economic conditions, because of their weaker capital positions, and because their capital positions, as they are measured today, are based on the assumption that the Government obligations owned by the banks can be held to maturity and consequently exaggerate the margin of safey provided by capital funds. Economic maladjustments as serious as those of 1922-29, affecting either agriculture or another segment of the economy, may 'occur again, with as great an impact on the solvency of banks.

Perhaps the best method of visualizing how the Federal Deposit Insurance Corporation would be affected by a recurrence of a situation like that of the 1922-29 period is to set up a hypothetical deposit insurance fund for the period from 1900 to 1930. If established at the beginning of 1900 such a fund would have operated prior to 1922 nearly as long as the Federal Deposit Insurance Corporation has been in existence. Tables 9, 10, 11, and 12 show how such a hypothetical fund would have fared, had it been set up with a capital fund, assessment rate, insurance coverage, and assured borrowing power comparable with those of the Federal Deposit Insurance Corporation, and with various assumptions about the portion of the net assessment income retained by the fund. The first of these four tables shows the losses of that period on depositors' balances of $5,000 and under, which may be taken as the losses met by the fund, the income of the fund, and its size at the end of the year, assuming that all the net assessment income was retained by the fund. The second table shows the income of the fund, and its size at the end of each year, with 40 percent of the net assessment income retained by the fund; and also if the entire net assessment income had been credited to the banks.

The third table shows the disbursements of the fund each year on account of failed banks, on the assumption that such disbursements would have been equal to one-half of the deposits of the failed banks, and also the estimated portion of the fund that would have been tied up in assets acquired from failed banks. The fourth table deals with the adequacy of the fund's assured borrowing power to meet the situation of the middle and late 1920's.

These tables indicate that a deposit insurance fund established at the beginning of 1900 would have survived the banking difficulties of 1907 and 1908 without recourse to borrowing had the fund retained all of its net assessment income, though the assets acquired from failed banks would have absorbed a larger portion of the fund than was the case with the fund of the Federal Deposit Insurance Corporation in 1940. Had a fund established in 1900 credited the entire net assessment income to the banks borrowing would have been necessary in the 1907-08 situation, and it might have been needed with retention of 40 percent of net assessment income.

By 1920 the hypothetical fund started in 1900 would have been more than four times as large as its initial size had all assessment income been retained, but less than twice as large as its initial size if the entire net income had been credited to the insured banks. However, in 1920 the deposits in operating banks were more than five times as large as at the beginning of the century, having increased at an average rate of more than 8 percent per year. Consequently, the fund would have declined relative to deposits. Had the fund started with an amount equal to three-fourths of 1 percent of the deposits in operating banks and retained all the net assessment income, by 1920, it would have declined to about three-fifths of 1 percent of the deposits in operating banks. With retention of 40 percent of net assessment income the fund would have been reduced to two-fifths of 1 percent of the deposits in operating banks. With the entire net assessment income credited to the banks, the fund would have been reduced nearly to one-fifth of 1 percent of the deposits in operating banks.

The hypothetical fund would also have successfully weathered the depression in 1921. The wave of failures beginning in the latter part of 1920 and continuing through 1921 would not have required the fund to make use of its borrow

ing power; and the proportion of the fund tied up at the end of 1921 in assets acquired from failed banks would have have been considerably less than in 1908. In fact, a fund retaining all of its net assessment income would have had only one-fourth of the fund in such assets at the end of 1921-not much greater than was the case with the Federal Deposit Insurance Corporation in 1940.

Nevertheless, the bank failures of the prosperous period of 1922-29 would have been disastrous to the hypothetical fund started in 1900, particularly if the fund had retained only 40 percent or none of its net assessment income. Had it retained the entire net assessment income it could have survived, with the use of only a portion of its assured borrowing power, until 1930. But with retention of only 40 percent of net assessment income both the fund and assured borrowing power would have been exhausted by the end of 1929. With the entire net assessment income credited to the banks both would have been exhausted by the end of 1926. In these computations it is assumed that the fund would have borrowed from the United States Treasury without payment of interest; if interest were charged the exhaustion of the borrowing power would have come earlier.

Taking the year 1922-29 by themselves, the cumulative deficit of the fund incurred from the losses charged off would have been about one-half of 1 percent of the average annual aggregate deposits in operating banks. In addition, the assets acquired from failed banks in excess of the losses, and held to the end of 1929 on the assumption used in the computation that all liquidations are completed by the end of the fourth year after the year of failure, would have amounted to about three-tenths of 1 percent of the deposits in operating banks. Under the circumstances of such a period, liquidation of the acquired assets would probably proceed less rapidly than is implied by this assumption, so that a more reasonable allowance for holdings of assets acquired from failed banks would be one-half of 1 percent. This leads to the conclusion that to meet a situation like that of 1922-29, the Federal Deposit Insurance Corporation would need an accumulated fund of 1 percent of the deposits in insured banks.

The claim may be made that since the present deposit insurance fund plus the assured borrowing power of $3 billion now equals more than 2 percent of the total deposits of insured banks, though the fund itself is only 0.8 percent of such deposits, this is sufficient to prepared for a contingency such as we have been discussing here. However, as deposits in insured banks increase with the passage of time, the assured borrowing power will become relatively smaller. In addition, dependence on the Corporation's borrowing power to meet such a contingency overlooks the fact that the banking difficulties of such a period, if they occur, will appear during a stable or prosperous period for the economy as a whole. For the Corporation to resort to the borrowing power in such a period—which action is tantamount to an admission that the deposit insurance fund is insolvent-would be detrimental to public confidence.

The greatest contribution the Corporation has made, and should continue to make, is maintenance of the confidence of the depositing public. While the standby arrangement with the Treasury is a source of comfort to the Corporation, to be used in the event of a serious emergency, the minute the Corporation is forced to show bills payable in its financial statement the public confidence might be shattered.

Summary and conclusions.-Proposals to reduce the deposit insurance assessment usually compare the Corporation's loss experience since 1933 with the present assessment income and include an assumption that banking difficulties of the kind which were common prior to Federal deposit insurance will probably not recur. Such proposals state or implicitly assume that the present deposit insurance fund of 0.80 percent of the deposits in insured banks is more than adequate to meet the need of the Corporation for a reserve fund.

When all the evidence is considered, and not merely the Corporation's loss experience, there is no reason to conclude that the present deposit insurance fund is adequate to meet the disbursements which may be involved in the contingencies for which the Corporaton should be prepared. The data that the Corporation has been able to collect and analyze-relating both to historical experience and to the current distribution of the Corporation's potential liabilitylead us to the conclusion that a reserve fund amounting to 1 percent of the deposits in insured banks is the smallest figure that can possibly be considered reasonably adequate in view of the Corporation's responsibilities. Under the present assessment provisions, continuance of the remarkably low losses of the Corporation to date, and continuance of a 4 percent per year rate of increase in the deposits in insured banks, it will take about 15 years for a fund of this size to be accumulated.

84444-57-pt. 232

TABLE 1.-Deposits in insured banks and the deposit insurance fund, 1934-56

1 Estimated by applying to the deposits in the various types of account at the regular call dates the percentages insured as determined from special reports secured from insured banks.

2 All figures for Dec. 31, 1956, except the amount of the deposit insurance fund, are estimated.