The Growth of Financial Intermediaries as a Factor in the Effectiveness of Monetary Policy
Author: Eugene A. Birnbaum
Publication Date:
01 Jan 1958
eISBN:
9781451968651
ISBN:
9781451968651
Language:
English
Keywords:
SP; cover book; bibliography; periodical article; continuation of the bibliography; publication; Federal Reserve bank; bank asset; member banks of the Federal Reserve System; Federal Reserve member banks; savings and loan; Nonbank financial institutions; Commercial banks; Insurance companies; Government securities; Open market operations
Download PDF (2.2 MB)

THE REVIVAL of monetary policy to control the forces of inflation and deflation has focused attention on the role of nonbank financial intermediaries as a factor influencing the effectiveness of the existing instruments of monetary control. In some quarters in the United States, the need for further factual information regarding the intricate structure and development of financial institutions over the past 40 years has been emphasized. Allan Sproul, formerly President of the Federal Reserve Bank of New York, suggested the appointment of a Presidential Commission to conduct a “broad national inquiry” into this question which might serve later as a basis for appropriate legislation.1 The Committee for Economic Development (CED) subsequently announced the formation of a National Commission on Money and Credit to undertake an extensive examination of U.S. monetary and credit institutions. It is expected that the Commission will, among other things, investigate “… whether the growth of non-bank financial intermediaries, both private (such as savings and loan associations, finance companies and insurance companies) and public (such as the Federal National Mortgage Association), has impaired the effectiveness of Federal Reserve action to promote economic stability.” 2

Abstract

THE REVIVAL of monetary policy to control the forces of inflation and deflation has focused attention on the role of nonbank financial intermediaries as a factor influencing the effectiveness of the existing instruments of monetary control. In some quarters in the United States, the need for further factual information regarding the intricate structure and development of financial institutions over the past 40 years has been emphasized. Allan Sproul, formerly President of the Federal Reserve Bank of New York, suggested the appointment of a Presidential Commission to conduct a “broad national inquiry” into this question which might serve later as a basis for appropriate legislation.1 The Committee for Economic Development (CED) subsequently announced the formation of a National Commission on Money and Credit to undertake an extensive examination of U.S. monetary and credit institutions. It is expected that the Commission will, among other things, investigate “… whether the growth of non-bank financial intermediaries, both private (such as savings and loan associations, finance companies and insurance companies) and public (such as the Federal National Mortgage Association), has impaired the effectiveness of Federal Reserve action to promote economic stability.” 2

THE REVIVAL of monetary policy to control the forces of inflation and deflation has focused attention on the role of nonbank financial intermediaries as a factor influencing the effectiveness of the existing instruments of monetary control. In some quarters in the United States, the need for further factual information regarding the intricate structure and development of financial institutions over the past 40 years has been emphasized. Allan Sproul, formerly President of the Federal Reserve Bank of New York, suggested the appointment of a Presidential Commission to conduct a “broad national inquiry” into this question which might serve later as a basis for appropriate legislation.1 The Committee for Economic Development (CED) subsequently announced the formation of a National Commission on Money and Credit to undertake an extensive examination of U.S. monetary and credit institutions. It is expected that the Commission will, among other things, investigate “… whether the growth of non-bank financial intermediaries, both private (such as savings and loan associations, finance companies and insurance companies) and public (such as the Federal National Mortgage Association), has impaired the effectiveness of Federal Reserve action to promote economic stability.” 2

In other quarters, the growth of nonbank financial intermediaries is said already to have largely destroyed the effectiveness of the existing instruments of monetary control. For example, the Bulletin of April 23, 1956 of the Institute of International Finance pointed to the “great” change in the U.S. money market that has occurred over the past two decades, “sweeping changes” in the capital market in which “institutional investors play a much more important role than ever before,” and other factors which make “… evident that the methods employed by the Federal Reserve authorities during the 1920’s and the 1930’s are not suitable at the present time, when the money and capital markets and general economic conditions are so different from those formerly prevailing.” 3 It was also stated that “recent developments in the capital market indicate that the credit policies of the Reserve authorities are today less effective than in the past because of the increased importance of institutional investors, notably insurance companies.”4 In an article published in this issue of Staff Papers, Thorn stresses the threat to the effectiveness of monetary policy which is said to arise in connection with the ability of nonbank financial intermediaries to expand credit in the short run, immune from direct central bank control.

The importance of the activities of nonbank financial intermediaries in the United States was also stressed in an article by Gurley and Shaw which concluded that changes in interest rates are best explained when account is taken not only of the monetary system but also of the role of nonbank financial intermediaries.5

The primary purpose of the present study is to determine, on the basis of available statistical evidence,6 whether the growth of nonbank financial intermediaries has narrowed the scope for effective monetary control by the Federal Reserve authorities. The study is directed to the situation in the United States, since essential information on other countries is lacking. However, the final part of the paper provides a glimpse of some of the relevant financial developments in the United Kingdom, Germany, and the Netherlands.

The analysis is focused on one of the important factors that would have to be considered in determining an answer to the question whether the powers of the Federal Reserve would have to be increased to compensate for any loss presumed to have been a consequence of the growth of nonbank financial intermediaries. No attempt is made to examine whether these powers ought to be increased on other grounds. Nor is this paper concerned with the question whether it is “fair” or equitable for the member banks of the Federal Reserve System to bear more of the “burden” of monetary control than is borne by other financial intermediaries.

Financial Intermediaries in the United States

Definitions and functions

The institutions referred to as “financial intermediaries” in this study exclude some important components of the financial intermediaries sector of the U.S. economy, viz., Federal Reserve Banks, Federal Government trust funds, and Federal Government lending institutions. With these exceptions, “financial intermediaries” are defined as including banks and all kinds of insurance agency, as well as such institutions as savings and loan associations, credit unions, personal and sales finance companies, pension funds, personal trust departments, and even security brokers and dealers (where available data permit).7 Obviously, certain types of financial intermediary cannot be included, because no adequate information about them is available, e.g., unincorporated financial groups, such as a stock market investing pool of the resources of several individuals. Such omissions are probably of minor importance.

The reason for excluding the Federal Reserve Banks is that we wish to observe data which measure the impact of Federal Reserve policies on other financial intermediaries. If the Federal Reserve Banks were included in the totals for intermediaries, many of the effects of their actions on other intermediaries would not be apparent; e.g., Reserve Bank purchases of government securities from other intermediaries would not appear in the record of total holdings of government securities by the financial intermediaries sector, inclusive of Reserve Banks.

Federal Government trust funds are excluded primarily for conceptual reasons. These funds are mainly social security accounts; they come from tax receipts, a source quite different from that of other intermediaries. Also, the fact that their monies are “lent” to the Government is a legal technicality rather than an economic fact. Their operations, as both receivers and lenders of funds, are not influenced by market forces in the same way that the operations of other financial intermediaries are influenced; hence, their inclusion in the financial intermediaries sector would have a distorting effect.8 For similar reasons, government lending institutions are also excluded from the scope of the financial intermediaries sector as defined in this paper.

Relative growth

Member banks of the Federal Reserve System are subject to the control of Federal Reserve authorities more directly than other financial intermediaries, and if the operations of member banks grow relative to other financial operations, this may be a factor which strengthens the position of the Federal Reserve authorities. All financial intermediaries are, of course, subject to Federal Reserve influence in varying degrees, and the relative growth of the main types of intermediary is an indication of the way in which Federal Reserve influence may have changed through time. The relative growth of financial intermediaries, as measured by the percentage distribution of total assets held by all9 intermediaries at the end of selected years from 1900 to 1952, is indicated by Chart 1.

Chart 1.
Chart 1.

Percentage Distribution of Total Assets of Financial Intermediaries in the United States1

Citation: IMF Staff Papers 1958, 002; 10.5089/9781451968651.024.A003

1Based on data in Table 9.

The chart shows that from 1900 to the beginning of World War II the commercial banks did not maintain their position relative to other intermediaries as measured by their share in the total assets held by intermediaries as a whole. While the proportion of assets held by commercial banks declined from 53 per cent of the total in 1900 to 39 per cent at the end of 1939, the proportion for insurance companies increased from 12 per cent to 22 per cent.

This decline in the relative position of commercial banks was, as is generally known, reversed during World War II. It appears evident, however, that the reversal was only temporary, since in the postwar period the relative decline of commercial banks has been resumed. The postwar decline of the commercial banks’ position has also been accompanied by a decline in the position of personal trust departments, whose relative importance as holders of assets had increased in the earlier period (1900-1939).10 The relative gain of assets taken up by the nonbank financial intermediaries in the postwar period appears to have been shared more evenly among the various types of intermediary than the gain from 1900 to 1939, so that the expansion of insurance companies was less rapid than it had been in the earlier period. This undoubtedly is in large part attributable to the rapid expansion in this period of Federal Government insurance and social security programs, which, as previously mentioned, are omitted from these data.

Chart 2, which shows the shares of the commercial banks and of the noncommercial bank intermediaries in the changes in the total assets of intermediaries between the benchmark dates of Chart 1, illustrates more sharply the decline during this period of the commercial banks’ relative position among financial intermediaries. In each subperiod, the ratio of the change in commercial banks’ assets to the change in the assets of all financial intermediaries is less than the ratio of total commercial banks’ assets to the total assets of all intermediaries. This, of course, is merely an additional demonstration of the proportional decline shown in Chart 1 of commercial banks’ assets relative to the total assets of intermediaries. Chart 2, however, shows more clearly the rapid decline of the banks and the corresponding growth of other intermediaries; for example, in 1946-49 the noncommercial bank intermediaries were responsible for the entire net increase of assets held by intermediaries as a whole, and the assets of the commercial banks actually declined.

Chart 2.
Chart 2.

Percentage Distribution of Changes in Total Assets Held by Financial Intermediaries in the United States1

Citation: IMF Staff Papers 1958, 002; 10.5089/9781451968651.024.A003

1Based on data in Table 10.

The facts presented in Charts 1 and 2 generally substantiate the claim that in the first half of this century noncommercial bank intermediaries have grown at a more rapid rate than commercial banks. This has often been cited as evidence that the over-all influence of the Federal Reserve System has declined, since the proportion of financial intermediaries over which the Reserve authorities have direct control has declined. However, the relative decline of commercial banks (i.e., members of the Federal Reserve System) may be a necessary but certainly is not a sufficient condition for proving the case that Federal Reserve authorities are less effective than formerly. The member banks constitute only that component of the financial intermediaries sector which is most subject to the direct control of the Reserve authorities, exercised through such weapons as the discount rate, moral suasion, and variable reserve requirements. And while it is true that certain instruments of monetary control most directly affect the member banks of the Federal Reserve System, the Reserve authorities have another instrument of much broader scope, viz., open market operations, which may be used to influence the activities of other financial intermediaries as well as those of member banks. Moreover, other important aspects of the situation have to be examined before it can be determined whether the relative growth of the “uncontrolled” 11 segment of the financial intermediaries sector has diminished the effective influence of the Federal Reserve authorities.

Changing Scope for Open Market Operations

Prior to the establishment of the Federal Reserve System in 1913, banks were required by law to hold gold or cash reserves equal to specified percentages of their liabilities. The inflexibility of the required reserve ratio, coupled with the absence of a central bank to act as a lender of last resort, helped to generate severe monetary crises at times when the ability of the banks to expand credit was completely exhausted. The Federal Reserve System was established to remedy this situation. At that time, it was generally believed that the gold standard mechanism could safely be permitted to be the prime regulator of the volume of money and of bank credit in accordance with “automatic” balance of payments forces.12

During the first 15 years of the Federal Reserve System, open market operations were practically unknown; member bank borrowings from the Federal Reserve Banks were substantial and constituted the main source of member bank reserves. Under these circumstances, one of the most effective instruments for credit control at the disposal of the Federal Reserve authorities was the power to vary the discount rate. Early drafts of the Federal Reserve Act “recognized open market operations in the classical British view of central banking as being a valuable means of making the bank rate effective.” 13 It was foreseen that at times the member banks would not require discounts and that it might be necessary for a Reserve Bank to sell securities for the purpose of making the rediscount rate effective.

Aside from some occasional open market operations that involved transactions in municipal and government bonds and bankers’ acceptances, there was no broad use of these operations until the 1930’s. The growth of government securities relative to total intangible assets probably contributed significantly to the gradual recognition of the potential effectiveness of open market operations. But it was not until World War I that the volume of government securities in circulation became a significant component of the total outstanding intangibles; the proportion of total intangible assets represented by government securities increased from less than 1 per cent in 1912 to 7 per cent in 1923. After 1923 this proportion declined and by the end of 1929 was down to 3 per cent.14

In the meantime, however, the experience of the 1920’s led the Federal Reserve authorities to shift more and more to the idea that “… the central bank should seek to influence business on its own initiative, instead of waiting passively for the member banks to apply for additional reserves…. By the end of the 1920’s … open market operations had become a major instrument of Federal Reserve policy.”15

The wide acceptance by the vast majority of financial intermediaries of government securities as an asset to hold is illustrated by the fact that, since the establishment of the Federal Reserve System, more than half of the total government securities held by financial intermediaries were in the hands of intermediaries other than member banks on four of the seven dates covered in Chart 3.

Chart 3.
Chart 3.

Percentage Distribution of U.S. Government Securities Held by Financial Intermediaries in the United States1

Citation: IMF Staff Papers 1958, 002; 10.5089/9781451968651.024.A003

1Based on data in Table 11.

The growth in the relative importance of government securities, as measured by their share of total assets held by financial intermediaries, is shown in Chart 4. For financial intermediaries as a whole, the ratio of claims on the Government to total assets increased from a low of 2 per cent in 1912 to a high of 49 per cent by the end of 1945. This ratio has, however, been higher for banks than for other financial intermediaries.

Chart 4.
Chart 4.

Government Obligations as Percentage of Total Assets Held by Various Types of Financial Intermediaries in the United States1

Citation: IMF Staff Papers 1958, 002; 10.5089/9781451968651.024.A003

1Based on data in Table 12. Horizontal scale indicates benchmark dates, not equal time intervals.2 Private life insurance companies, savings and loan associations, mutual savings banks, and postal savings system.

The data show that most of the “uncontrolled” intermediaries held no government securities at all in 1900 and 1912, and that 1933 is the first year for which holdings by savings and loan associations appear. However, the growth of the importance of holdings of government securities is a common characteristic of all financial intermediaries from 1929 through 1945; even after the postwar decline, these holdings have continued to be significant for nearly all intermediaries.

The ratios of government securities to the total asset portfolios of intermediaries, shown in Chart 4, reveal some interesting tendencies in the behavior of intermediaries in periods of wide economic fluctuation. From 1922 to the end of the 1929 boom, the decline in the ratio was much less for the member banks than for the nonbank intermediaries; and during the recovery from 1933 through 1939, the ratio increased much less for the banks than for the nonbank intermediaries. In the 1939-45 period of war finance, the ratio increased slightly more for the banks; from the end of the war, “uncontrolled” intermediaries, in general, shifted out of government securities much more rapidly than the banks. These facts suggest that the member banks were more responsive to Federal Reserve policy than were the “uncontrolled” intermediaries. Indeed, in all the subperiods covered in the chart, with the exception of 1929 through 1933, the member banks seem to have acted in accordance with the requirements of a countercyclical monetary policy more closely than the nonbank intermediaries.

Thus it appears that the control of the monetary authorities has generally been more effective when it has been applied to the member banks. The fact that the period from 1929 through 1933 was an exception to this general rule is very likely explained by the severe financial crisis during those years; the banking reforms of the early 1930’s have probably eliminated any possibility of a recurrence of a similar financial crisis.

The evidence that “uncontrolled” intermediaries have been less responsive than member banks to the general requirements of monetary policy does not necessarily imply that the potential effectiveness of monetary policy has been weakened by the more rapid growth of these intermediaries. Undoubtedly, the evidence is largely a result of the failure of the monetary authorities to exploit the growing power of open market operations as an instrument of control, e.g., during the extended period in which the Federal Reserve System was pegging the interest rate, and especially in the period of war finance which spilled over into the postwar period.

Contribution of “uncontrolled” intermediaries to strengthening of monetary policy

In assessing the implications of the growth of nonmember-bank financial intermediaries, it is of interest to investigate the maturity distribution of government debt holdings of financial intermediaries as a factor influencing the effectiveness of open market operations as an instrument of Federal Reserve control.

Data on the ownership on June 30, 1955 of interest-bearing government securities classified according to maturity (Table 1) show that at that time only 11 per cent of the total debt (held outside Federal Reserve Banks and Federal Government accounts) bore a maturity of 10 years or more, and that over two thirds of this fraction bore a maturity of less than 15 years. For commercial banks, only 4 per cent of their holdings of government securities had a maturity of 10 years or more; furthermore, nearly 60 per cent of their total holdings of these securities bore a maturity of less than 5 years. Although 18 per cent of the government securities held by insurance companies bore a maturity of over 10 years, the total of government securities of all maturities held by the insurance companies covered in the survey 16 comprised only 8 per cent of the total government securities outstanding (i.e., in general, the total excluding holdings of Federal Reserve Banks and Federal Government accounts).

Table 1.

U.S. Government Interest-Bearing Debt and Guaranteed Obligations, as of June 30, 1955, Classified According to Ownership and Maturity1

article image

Based on data from Annual Report of the Secretary of the Treasury on the State of the Finances for the Fiscal Year Ended June 30, 1955 (Washington, 1956), pp. 514-15.

Holdings of Federal Reserve Banks and of U.S. Government investment accounts are excluded.

Some scattered information on the maturity distribution of the total asset portfolios of some representative groups of financial intermediaries is presented in Table 2. This table appears to indicate not only that the asset position of the banks has, historically, been more liquid than that of other intermediaries, but that there also has been a tendency in recent years for the banks to increase their liquidity through investing a larger portion of their funds in short-term assets. At the same time, the insurance companies (and other nonbank intermediaries) seem to be channeling relatively more of their funds into long- term assets. However, the impossibility of further subclassifying the large securities category by maturity makes it difficult to draw any firm conclusions on this matter. Furthermore, the increasing application of government guarantees—to private mortgage indebtedness, for example—has undoubtedly changed the liquidity character of long-term asset portfolios.

Table 2.

Asset Portfolio, Classified According to Maturity, of Certain Financial Intermediaries in the United States1

(In per cent of total portfolio for each group)

article image

Based on data from Raymond W. Goldsmith, Financial Intermediaries in the American Economy Since 1900; see text footnote 6. Cash, tangibles, and miscellaneous assets are excluded.

The growth of “uncontrolled” financial intermediaries has been a factor increasing the effectiveness of open market operations to the extent that this growth has increased the average maturity of government debt. The longer the maturity of a debt, the larger will be the capital gain or loss associated with any given change in the interest rate. The longer the average maturity of debt holdings, the larger will be the fall in the capital value of government security holdings produced by any increase in the interest rate, and, therefore, the larger the reduction of the total potential amount that might be disposed of by selling government securities in order to finance credit extension to the private sector, and the greater the effectiveness of Federal Reserve policy in using a higher interest rate as an instrument for discouraging financial intermediaries from shifting out of government securities into claims on the private sector. Figures showing how the hypothetical destruction of capital would vary in accordance with the maturity distribution of government debt holdings when the interest rate is increased from 3 per cent to 4 per cent per annum illustrate this point, the loss being recorded as a percentage of the total principal value which existed when the interest rate was 3 per cent:

article image

In the limit (i.e., for a security issued in perpetuity), the proportion of capital destroyed through raising the interest rate by one third (e.g., from 3 per cent to 4 per cent per annum) would be 25 per cent. But the U.S. Government has never issued a perpetual security. Furthermore, the data in Table 1 have already shown that only 11 per cent of the government securities outstanding (as defined) on June 30, 1955 bore a maturity of 10 years or more and that only a relatively small amount of these had a maturity longer than 15 years.

The hypothetical destruction of capital which bears a 10-year life is much less than that for a perpetuity, viz., only 8.18 per cent against 25 per cent. In view of the fact that a large (albeit unknown) proportion of the government long-term securities held by financial intermediaries are doubtless, in accordance with individual diversification requirements, considered to be a “permanent” part of their asset portfolios, and hence would not be generally considered as potentially eligible for sale prior to maturity, it follows that the “capital destruction effect” of interest rate increases has probably been a negligible factor with respect to the effectiveness of Federal Reserve open market operations. However, to the extent that the existence of “uncontrolled” intermediaries may have promoted a lengthening of the average maturity of debt, these institutions have helped to strengthen the Federal Reserve System.

It has also been claimed that the anti-inflationary operations of the Open Market Committee are strengthened by the so-called “locking-in” effect on the asset portfolios of financial intermediaries. This effect is based on the notion that financial intermediaries are often reluctant to unload a previously acquired asset at a price below that of the original purchase, irrespective of whether more attractive assets are currently available, in order to avoid the necessity of showing a realized capital loss on the firm’s books. Such a capital loss need never be realized on any asset whose original purchase price is equal to or less than its principal value at redemption, provided that the firm continues to hold the asset until maturity, or until such time as its market price might return to that of the original purchase. This factor, also, would be more important for those lending institutions which hold a large volume of long-term government securities than for those which hold predominantly short-term securities.

Irrespective of whether they are held to maturity or sold, the capital loss associated with a price decline of assets held in portfolios can never be recouped. This is not intended to imply that under all conceivable circumstances a firm’s best interest, and perhaps even the best interest of its clientele, would not be served by holding to maturity, just in order to avoid showing a loss on the books, any assets whose prices have fallen. (Undoubtedly, if anyone benefits it is the firm’s management; with the growing separation of management and ownership functions, this point might have considerable importance.) However, if the “psychological impact” of revealing a loss is neglected, any firm which operates at all times in such a way as to maximize profits will decide whether to shift out of government securities into other claims solely on the basis of which claim is relatively the more attractive or profitable, i.e., without regard to the effect of current price changes on the book valuation of previously acquired assets.17

In addition, the current provisions of the tax law permit banks to deduct from their ordinary taxable income the full amount of a realized net capital loss on bonds and other instruments of indebtedness. Hence, by selling bonds when bond prices are declining, the loss can be applied in full against taxable income from other operations, thereby reducing the current tax liability by 52 per cent of the capital loss. If the proceeds of the tax savings are then added to sales proceeds and together are reinvested in the same 18 or similar securities, the bank can actually increase its holdings of securities.19

Thus, an investment manager of a financial intermediary who acts strictly in accordance with the indications given by the so-called “locking-in” effect disregards the question whether the firm should shift out of government into private securities on the basis of which might be the more profitable asset to hold; furthermore, if he be a banker he also disregards the additional profits to which he would be legally entitled by taking advantage of the current provisions of the corporate income tax law as applied to banks. However, to the extent that the “locking-in” effect actually operates (and it apparently has some validity), it is a factor which increases the effectiveness of Federal Reserve policy.20

Inasmuch as the previously discussed effects may tend to strengthen the impact of open market policy and are particularly applicable to the operations of “uncontrolled” financial intermediaries, since they are inclined to hold debts of longer maturities, we are led to the conclusion that the growth of these intermediaries has been a factor which has increased the effectiveness of open market policy as an instrument of control.

The liquidity argument

Perhaps the most frequently offered argument regarding the significance of the maturity distribution of debt as an influence on the effectiveness of counterinflationary monetary policy is that which may be called the “liquidity” argument. Short-term government securities are close substitutes for cash and might therefore, for practical purposes, be regarded as the same thing as cash. Since cash forms the basis of credit expansion, the larger the holdings of money, including “near money,” the more difficult it is to prevent an excessive credit expansion by applying the instruments of monetary control. The vast accumulation of short-term government securities in the wake of war finance is therefore regarded as a unique source of potential inflation in the postwar period in the face of pent-up private demand for loanable funds.

The importance of this argument has been exaggerated, and it can be misleading. The general problem facing monetary authorities in the aftermath of an extended period of inflationary war finance is basically the same, regardless of whether the assets which have accumulated in the hands of financial intermediaries are predominantly short-term or predominantly long-term. The potential inflation stems from the existence of too many low-yielding government securities in circulation, be they long-term or short-term. Furthermore, the accumulation of assets in the form of claims on the Government is evidence of the wartime squeezing-out of private enterprise. With the removal of wartime restraints, the business community quickly responds in its attempt to profit from continuing shortages, and financial intermediaries are, of course, eager to provide the necessary credit, both because business obligations (and residential mortgages, etc.) provide a higher rate of return than government securities, and also, to some extent, because the intermediaries wish to diversify their asset holdings, i.e., to rid themselves of the excessive volume of government securities that they have accumulated.

If it is assumed, in the interests of simplicity, that the Government is neutral in an inflationary (postwar) period, the monetary authorities are confronted with a constant supply of government securities on which to operate, and it should make no fundamental difference whether the supply is primarily short-term or long-term. The problem would be, in any event, the same, viz., to slow the expansion of credit to the private sector. One important way of accomplishing this is by making government securities competitive with private obligations in terms of their rate of return. Thus, the Open Market Committee might reduce the price of government securities, be they short-term or long-term, to the extent required, to make them more attractive than other securities so that the shift from government to other securities would occur only at the rate which the monetary authorities desired.

If it so happens that financial intermediaries have a large portfolio of long-term government securities, this could have serious repercussions on their profits. If, on the other hand, they had correctly foreseen what was coming, and had therefore decided to hold a quantity of short-term securities large enough to protect them against future declining prices, they would be in a much more fortunate position during the ensuing period of greater returns on loanable funds. To the manager of an individual financial firm, the question whether to buy short-term securities (or hold cash) now, and thereby to take a smaller return on investments than is currently available from long-term securities, in order to invest in long terms later when it is expected that interest rates will be higher, is a serious and real dilemma.21

This dilemma is, however, basically irrelevant to the situation con-fronting the monetary authorities (neglecting welfare implications). The problem remains that of raising to the desired degree yields on government securities relative to other securities. If the authorities succeed (through open market operations, for example) in inducing the holders of government securities, either short-term or long-term, to continue to hold them, the inflationary potential arising from the danger of shifting from government to other securities will be effectively controlled.

In brief, the introduction of open market operations has provided the Federal Reserve authorities with a powerful instrument of control. The growth of “uncontrolled” financial intermediaries relative to member banks does not appear to have materially weakened the potential ef-fectiveness of Federal Reserve open market operations; on the contrary, the growth of “uncontrolled” intermediaries has helped promote the development of this instrument of control and probably even enhanced its effectiveness.

Growth of Financial Intermediaries Relative to National Finance, Income, and Assets

It has been shown that the importance of financial intermediaries other than commercial banks, as measured by the relative growth of their share of the total assets held by all22 financial intermediaries, has increased at the expense of the commercial banks. Open market operations have emerged as a potent instrument of control, the effectiveness of which has not been diminished by the growth of “uncontrolled” financial intermediaries. To some extent, indeed, the growth of “uncontrolled” intermediaries in certain directions has actually increased the effectiveness of open market operations. The final part of the analysis for the United States will investigate structural changes in the financial intermediaries sector relative to national income and national assets, since, despite the more rapid growth of “uncontrolled” intermediaries, the operations of “controlled” (member bank) financial intermediaries may have increased enough to be no less significant in relation to total economic activity than in the past.

As a first step, we may investigate the role of financial intermediaries as a source of finance. The share of financial intermediaries from 1901 through 1949 23 in the aggregate sources of funds utilized by all other sectors of the economy is shown in Table 3. The aggregate sources of funds data are estimates of the total (combined, not consolidated) net borrowings of the nonfinancial intermediaries sectors minus their net loans; a net decrease in cash balances is counted as a source of funds and a net increase as a use of funds.24

Table 3.

Share of Financial Intermediaries in Aggregate Sources of Funds of All Other Sectors of the U.S. Economy

article image
article image
Source: Raymond W. Goldsmith, Financial Intermediaries in the American Economy Since 1900 (see text footnote 6), Tables 89 and 90, pp. 298–99 and 303.

Both numerator and denominator are negative.

In view of the importance of the Federal Government sector, Table 3 presents two sets of data, one including, and the other excluding, the Government, Since Goldsmith includes in his aggregate sources of funds data the finance provided to the Federal Government from such sources as Government trust funds and Federal Reserve Banks, the data on financial intermediaries used in the table also include the operations of these agencies and, therefore, have a wider coverage than is required by the earlier definition of financial intermediaries. The distortion thus introduced is, however, negligible for that part of Table 3 which excludes the Federal Government, inasmuch as nearly all of the credit activities of these agencies is channeled to the Federal Government.

The highly irregular nature of financial operations during the great depression of the 1930’s and the rapid wartime inflation of the 1940’s, and the unavoidable errors of estimation and the absence of data more up to date than those of 1949, place important limitations on the significance of the data presented in Table 3; nevertheless some interesting tendencies can be observed. From 1901 to 1930, external financing fairly consistently provided some 25 per cent of the total sources of funds of the economy exclusive of the Federal Government as a user; for the 1946-49 period, the proportion fell slightly, to 22 per cent. Whether this decline is only temporary or indicative of a trend cannot be reliably ascertained in the absence of up-to-date data.

The data also show that the relationship between external finance and total sources of funds is more irregular when the Federal Government is included. This probably results primarily from the fact that the Government’s requirements for finance have been more erratic than the requirements of the private sector; in addition, the Government’s financial requirements are, of course, always potentially capable of being satisfied practically at its own dictates, i.e., with little regard to existing market conditions. Because of such factors, the comparatively large postwar decline of the share of external finance in the total sources of funds of the economy including the Federal Government as a user cannot be regarded as necessarily indicative of any basic or structural change.

These observations seem generally applicable also to the relationship between external financing and total net financing,25 except, of course, that the proportion of external financing to total net financing is larger. It appears, therefore, that external financing has continued to provide approximately the same proportion of the economy’s requirements as formerly, although in periods of extreme economic fluctuation, such as the great depression and World War II, the significance of external financing has also varied widely.

Although the importance of external financing in recent years appears to have been about the same as it was earlier, the data clearly indicate that the share of external financing provided by the financial intermediaries has increased significantly. The data relating financial intermediaries to the external financing of the economy excluding the Government (line 7) show that intermediaries furnished an average of 45 per cent of external finance prior to 1930, but that in the period 1946-49 their share had grown to 62 per cent. Although their share of private external financing appears to have dropped precipitously in 1934-45, a glance at the data which include the Government (line 15) shows that this merely reflects a shift from financing the private sector to financing the government sector.

The share of financial intermediaries in private external financing in 1946–49 was greater than in the pre-1930 period by 38 per cent. At the same time, however, the “controlled” intermediaries’ share of the total assets of all financial intermediaries declined by only 20 per cent.26Hence, relative to total (private) external financing, the member banks of the Federal Reserve System had grown in importance by 1949 in spite of their decline relative to other financial intermediaries. Since external financing appears to have maintained the same general relationship to economic activity as formerly, the tentative conclusion is suggested that during the early postwar period there was, on balance, an increase in the importance of the member banks of the Federal Reserve System in relation to economic activity. This implies a further strengthening of Federal Reserve influence.

For additional evidence concerning the changing scope of “controlled” and “uncontrolled” financial intermediaries, we may compare data since 1900 on the ratios of their assets to measures of general economic activity (Table 4 and Chart 5). Since data for total noncommercial-bank intermediaries are not available after 1952, a series covering a sample of noncommercial-bank intermediaries has been included. Inasmuch as the sample series has closely paralleled the series for total noncommercial-bank intermediaries since 1933, it appears reasonable to assume that movements in the sample series have continued to be representative since 1952.

Table 4.

Ratios of Assets Held by Financial Intermediaries in the United States to U.S. National Income1

(In per cent)

article image

Assets data are from Table 9. National income data for 1900, 1912, and 1922 are estimates of the National Industrial Conference Board, Inc., published in National Income of the United States, 1799-1938 (New York, 1939); for 1929-57, Department of Commerce data, published in Survey of Current Business.

Preliminary.

Based on unrounded data.

Chart 5.
Chart 5.

Ratio of Assets Held by Financial Intermediaries in the United States to U.S. National Income1

(1922 —100; logarithmic vertical scale)

Citation: IMF Staff Papers 1958, 002; 10.5089/9781451968651.024.A003

1Based on data in Table 4. Horizontal scale indicates benchmark dates, not equal time intervals.2 Private life insurance companies, savings and loan associations, mutual savings banks, and postal savings system.

In our earlier analysis of the relationship of intermediaries to external financing, it was shown that commercial banks had gained in importance in the early postwar period in comparison with the years prior to 1930. The data in Table 4 and Chart 5 appear to confirm these findings. However, as the period to be averaged since World War II is further extended, the difference between the positions of the banks in relation to income for the pre-1930 and the post-1945 periods tends to disappear. If the decline in bank assets relative to income should continue, it might eventually indicate a lessening in the importance of commercial banks to economic activity; however, the data so far give no clear indication that this has yet occurred.

On the other hand, Chart 5 clearly shows the relative gain in the growth of noncommercial-bank financial intermediaries since 1900. However, unlike their growth in relation to the financial intermediaries taken as a whole (as shown in Chart 1 and Table 9), their growth in relation to national income has been interrupted by periods of significant decline, especially in 1933-45. Furthermore, their growth rate relative to income since 1945 has been nowhere near so rapid as it was from 1900 to 1934.27

A further comparison of the movements of the data presented in Chart 5 shows that in five of the ten periods between the benchmark dates, ranging from December 31, 1900 to December 31, 1957, the ratio of assets held to national income for commercial banks and for noncommercial-bank financial intermediaries moved in the same direction. This merely demonstrates that the growth of “uncontrolled” intermediaries need not occur at the expense of the banks, i.e., some fraction of savings channeled to the noncommercial-bank intermediaries undoubtedly consists of basically “nonmonetary” savings. Or, put in another way, some of the savings channeled to such intermediaries as insurance companies, for example, undoubtedly would have been saved without recourse to the banking system even if the insurance companies did not exist. Thus, the growth of nonbank financial intermediaries may occur at the expense of such alternative means of savings as the direct purchase of marketable stocks and bonds, or direct investment in unincorporated enterprise, rather than at the expense of the banking system. Obviously, to the extent that savings channeled to “uncontrolled” intermediaries are basically of such a character, the growth of nonbank intermediaries would not in any sense reduce the scope of the commercial banks, and would thus not be a factor reducing the effectiveness of Federal Reserve control.

The total assets of financial intermediaries can be compared not only with data on national income, but also with figures provided by Goldsmith on total national assets and the components of this series: national intangible assets and national wealth (tangible assets, including land). Goldsmith’s concept of national assets is based on a combined national balance sheet “which is simply the sum of the balance sheets of all units domiciled within the nation’s boundaries, using consolidated balance sheets only for groups of parents and subsidiaries joined by majority ownership.”28 The combined balance sheet obviously duplicates some items, and in some instances, which, however, are probably unimportant, certain assets are counted even more than twice.29

Ratios of the total assets of financial intermediaries to total national assets and to their intangible and tangible (wealth) components, and also ratios of national income to national assets, are presented in Table 5. In Chart 6, the ratios for commercial banks and total other financial intermediaries are shown; and in section D of the chart, the ratios of national intangible assets to national wealth are shown along with a second series, derived from the data for Chart 1, which represents the ratios of the total assets of financial intermediaries other than commercial banks to the total assets of commercial banks.

Table 5.

Total Assets of Financial Intermediaries (Adjusted) in the United States as Per Cent of Total National Assets and Main Components1

article image
article image

Based on data from Raymond W. Goldsmith, A Study of Savings in the United States (see text footnote 10) and Financial Intermediaries in the American Economy Since 1900 (see text footnote 6).

The adjustment refers to the exclusion of Federal Reserve Banks, Federal Government trust funds, and Federal Government lending institutions from Goldsmith’s definition of intermediaries; see above, pp. 386-87. In the comparisons of Goldsmith’s totals (column 7) to national assets, etc., however, the assets of these institutions are included in both numerator and denominator.

Column 1 = columns 2 + 4 = columns 3 + 5. For statement on adjustment, see footnote 1

See footnote 1.

Including land.

Chart 6.
Chart 6.

Ratio of Assets Held by Financial Intermediaries in the United States to Total National Assets and Main Components1

(Logarithmic vertical scale)

Citation: IMF Staff Papers 1958, 002; 10.5089/9781451968651.024.A003

1Based on data in Table 5. Horizontal scale indicates benchmark dates, not equal time intervals.

Goldsmith considers the ratio of total assets of financial intermediaries to total national assets to be “the most comprehensive, and for many purposes the most significant, indicator of the position of financial intermediaries in a country’s economy….”30 The data show that for the United States this ratio has tended to increase since 1900. The upward trend is shown most markedly in section C of Chart 6, which presents the ratio of assets of intermediaries to national wealth, although the tendency to increase can also be observed in the comparisons with total intangible assets (section B) and total national assets (section A). Hence, the evidence again suggests that the role of financial intermediaries in the economy on the whole increased during the period following World War II (up to 1953), in comparison with the years prior to 1930. The postwar gains over the pre-1930 period for both commercial bank and noncommercial-bank intermediaries are summarized in Table 6. According to these data, the commercial banks maintained, at least until 1952, a position of greater importance in the economy as a whole than they had held in 1900-1929; the fact that bank assets have declined relative to the assets of other financial intermediaries cannot therefore necessarily be taken as proof of any decline in the importance of banks in relation to economic activity.

Table 6.

Percentage Increase Between Average for 1900-1929 and Average for 1949-1952 of Ratio of Total Assets of Financial Intermediaries in the United States to (1) Total National Assets, (2) Total Intangible Assets, and (3) Total Tangible Assets1

article image

Based on data in Table 5.

It is of further interest to note that the data of Table 5 and Chart 6 indicate that the ratios of financial intermediaries’ assets to measures of general economic activity have sometimes fluctuated widely. This instability is immediately observable from inspection of the chart, in spite of the fact that use of a logarithmic vertical scale has the visual effect of reducing the apparent variability.

Section D of Chart 6, when compared with the other sections, seems to indicate the basic forces which underlie the observed fluctuations. The large area between the curves in section D can be associated with similar areas in the other sections of the chart. During the two periods covering World Wars I and II, the section D series show a rise in the ratio of intangible assets to national wealth and reflect the fact that during war periods financing grows faster than wealth and the commercial banks are more active than other intermediaries as a source of finance.31 At the conclusion of the war periods, however, the situation has been abruptly reversed, and the nonbank intermediaries have gained assets more rapidly than the banks, until some point has eventually been reached when the banks have begun to recover. Just prior to World War I, the banks were gaining slightly relative to other intermediaries; this occurred again after 1933. These data, however, show no repetition of this tendency since World War II.

The wide variability in the relationship of financial intermediaries as a whole to measures of general economic activity can thus be associated with the wide variability in the stock of financial (intangible) assets relative to tangible assets; and the wide variability in the relationships of banks and other intermediaries to economic activity has in large part been associated with contrasts between wartime and peacetime conditions. Such “ad hoc” variation, nevertheless, does not detract from the strong evidence of a secular growth of financial intermediaries relative to economic activity which, although most clearly evident in the growth of the nonbank institutions, is also observable in the growth of the banks. As long as the banking system continues to grow roughly in step with economic activity, the power of the monetary authorities need not diminish, even if other financial institutions happen to be expanding somewhat more rapidly. Moreover, the fact that Federal Reserve authorities can implement their policies by means of open market operations, an instrument of broad scope, implies that the growth of nonmember-bank financial intermediaries by no means checks the power of the monetary authorities.

Growth of Financial Intermediaries in Other Countries

Few data of the kind discussed above for the United States are available for other countries. However, in an article, “Financial Structure and Economic Growth in Advanced Countries: An Experiment in Comparative Financial Morphology,” 32 Goldsmith has presented some selected statistical series on the United Kingdom, the Netherlands, and Germany. Some of these series were originally prepared by other authors within somewhat different frameworks, but they nevertheless shed some light on the development of the financial structures of these countries. The data in Table 7 have been derived from the figures in this article.

Table 7.

Selected Characteristics of Financial Structure: United Kingdom, United States, Netherlands and Germany1

article image

Derived from data in Raymond W. Goldsmith, “Financial Structure and Economic Growth in Advanced Countries: An Experiment in Comparative Financial Morphology,” Capital Formation and Economic Growth (papers presented at Conference of the Universities—National Bureau Committee for Economic Research, Princeton University Press, Princeton, N.J., 1955), pp. 113-60.

In the absence of detailed statistics, these data cannot be adjusted to conform with the definition of the scope of financial intermediaries used above for the United States. The data for financial intermediaries include government institutions, such as social insurance funds, and the data for banks include the central bank as well as other noncommercial banks (but presumably not investment banks). For comparative purposes, Table 7 also provides data for the United States that are based on a definition similar to that used for the other countries and are therefore inconsistent with the statistics for the United States used in the first part of this study. Owing to the general lack of statistical details and other pertinent information, the analysis must be limited to a discussion of some of the more interesting highlights of the financial developments in the countries covered in Table 7.

Perhaps the most interesting point that is relevant to the general interests of this study is that the banking system appears to have developed in different ways in the United States and in the other countries covered in Table 7 (see line 2). In the United Kingdom, bank assets relative to total intangible assets increased by more than 75 per cent from 1913 to 1947-49; in Germany, the ratio of bank assets to intangibles increased by about 80 per cent from 1913 to 1948. In contrast, the banks in the United States in 1949 maintained the same proportion of total intangibles as in 1912, viz., 20 per cent. Similarly, from 1939 to 1947 the banking system in the Netherlands increased its holdings of assets relative to total intangibles by 120 per cent, while in approximately the same period (1939-49) in the United States there was a slight reduction in this ratio.

An indication of the growth of banks in these countries, relative to the growth of other financial intermediaries, can be derived from Table 7 as the ratio of line 2 to line 1 (Table 8). From these data the growth of banks relative to other financial intermediaries in the United Kingdom, the Netherlands, and Germany appears to have been primarily a phenomenon of World War II. For the United States, the war period was shown above to have been one in which the long-term downward trend in the relative position of commercial banks was abruptly, though temporarily, reversed (see Chart 1). Because of the difference in coverage of the financial intermediaries sector (e.g., Government trust funds were not eliminated), the U.S. data in Table 8 do not show the reversal of the trend.

Table 8.

Ratio of Bank Assets to the Total Assets of Financial Intermediaries: United Kingdom, United States, Netherlands, and Germany1

(In per cent)

article image

Based on data in Table 7 (line 2 divided by line 1).

Large gaps in the figures for the Netherlands and the United Kingdom and the absence of up-to-date data leave little basis for conclusions about observable trends. The two postwar periods of monetary reform in Germany also destroy much of the comparability of the figures; however, there is fairly strong evidence that the German banks have tended to maintain, and perhaps even to strengthen, their position relative to other financial intermediaries. On the other hand, German banks have been noted for carrying on functions which would elsewhere be regarded as outside the scope of commercial bank operations ; the extent to which these secondary functions may be responsible for the continued growth of German banks relative to other intermediaries is not clear.

The more rapid growth of banks in the United Kingdom and the Netherlands can be directly associated with the more rapid expansion of government debt in those countries. From 1912-13 to 1947-49, the government debt component of total intangibles in the United Kingdom rose nearly twice as fast as in the United States; from 1939 to 1947-49, the government debt component of total intangibles in the Netherlands also increased approximately twice as fast as in the United States. Again, the monetary reforms, under which much of the outstanding government debt was repudiated, make impossible any useful comparison with Germany.

The record suggests two other interesting points. First, insurance companies have increased their assets relative to total intangibles in the United Kingdom, the United States, and the Netherlands, and up to 1948, also in Germany. The fact that the increase was wiped out in Germany by 1951 indicates that more up-to-date figures are required before reaching firm conclusions regarding the Netherlands and the United Kingdom. Second, the reduction of foreign assets in the United Kingdom and the Netherlands is of interest; the reduction relative to tangible assets is the more meaningful series (line 9 rather than line 7 of Table 7). The reduction of foreign assets in the Netherlands is clearly seen only in the comparison with intangible assets, which, however, is practically meaningless.

Because of the inadequacy of the information, little can be said about the implications of these data in relation to the effectiveness of monetary policy. The fact that the banking systems of these countries seem to have fared better than that of the United States suggests that the potential effectiveness of monetary policy may have increased more in these countries than in the United States. On the other hand, the growth of their banking systems may be largely a reflection of the inflationary aftermath of the war. Not only much more up-to-date statistics but also a careful analysis of other additional information would be necessary before any firm conclusion could be justified.

Table 9.

Distribution of Total Assets of Financial Intermediaries in the United States1

article image
article image

Based on data in Raymond W. Goldsmith, Financial Intermediaries in the American Economy Since 1900 (see text footnote 6), Table 10, pp. 73-74, and Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, various issues.

Preliminary.

End-of-year data.

Table 10.

Distribution of Changes in Total Assets of Financial Intermediaries in the United States1

article image
article image

Based on data in Table 9.

As derived in Table 9.