Keywords

JEL Classifications

The Federal Reserve System of the United States was established on 23 December 1913, when President Woodrow Wilson signed the Federal Reserve Act. The need for a new federal banking institution became clear when a severe crisis occurred in 1907. In May 1908 the Aldrich–Vreeland Act established a bipartisan National Monetary Commission that proposed establishing a National Reserve Association with 15 locally controlled branches that would ‘provide an elastic note issue based on gold and commercial paper’ (Warburg 1930, p. 59). The proposal was not enacted, nor was a subsequent proposal for a central bank with about 20 branches that would be controlled by a centralized Federal Reserve Board, consisting largely of commercial bankers. In the debate preceding the Federal Reserve Act, banking industry domination was rejected in favour of a board that had five members appointed by the President and two ex officio members, the Secretary of the Treasury and the Comptroller of the Currency. The appointed members had staggered terms and were to represent different commercial, industrial, and geographic constituencies. A sixth appointed member representing agriculture was added in 1923. The composition of the Board and its relation to Federal Reserve banks were drastically changed in 1935. Partly because of continuing disagreements about public versus commercial bank control, the new Board’s powers were left ambiguous in the act.

The act mandated that all national banks become members of the new system and stockholders of Federal Reserve banks. Because reserves were to be concentrated in 12 Federal Reserve banks, the act substantially reduced reserve requirements at national banks. State chartered banks could join if they chose to and were judged to be financially strong. The first Board was sworn in on 10 August 1914 and the system opened for business on 16 November 1914. Federal Reserve notes that were backed 100 per cent by ‘eligible paper’ and, additionally, 40 per cent by gold began to circulate. Eligible paper was self-liquidating, short-term paper that arose in commerce and industry. The rationalization for eligible paper was the real bills doctrine, which held that credit extended for financing only the production and distribution of goods would not lead to inflation. The doctrine is invalid because of fungibility; there is no relation between paper acquired by Federal Reserve banks and loans the commercial banks are extending. In addition, all deposits at Federal Reserve banks had to be backed at least 35 per cent by gold. Subsequent amendments to the act effectively eliminated the supra-100 per cent collateralization of notes. A June 1917 amendment to the act forced all member banks to pool required reserves at Federal Reserve banks and further reduced reserve requirements to decrease the burden of membership on national banks and attract more statechartered banks to the system.

The Early Years

The early years of the Federal Reserve System were marked by struggles to define the distribution of power between Federal Reserve banks and the Board, in the context of growing US involvement in the First World War. The Board gradually assumed more powers, but was unsuccessful in controlling open-market trading, which inevitably was concentrated in New York. Benjamin Strong, the New York bank governor, managed system trading. (Until 1935 the chief executives of Federal Reserve banks were called ‘governors’. After 1935 their title was changed to ‘president’ and members of the Board were called ‘governors’.) The Federal Reserve System was made fiscal agent for the Treasury in 1920, but the Treasury dealt directly with Federal Reserve banks, not the Board. Until 1922 the Board’s statistical research office was located in New York, and arguably the Board was less informed than the New York bank about money market conditions.

Federal Reserve banks immediately sought earning assets in order to pay expenses and the six per cent required dividends on member bank capital subscriptions. As they expanded their portfolios of bills, US securities, discounted commercial paper, and acceptances, the breadth and liquidity of these markets increased. In early 1915 the New York bank was buying and selling for other Federal Reserve banks. Discount rates charged by reserve banks varied across Federal Reserve districts.

In anticipation of the US declaration of war on Germany in 1917, Federal Reserve banks became responsible for issuing and redeeming short-term Treasury debt certificates before and during Liberty Loan drives. There would be four large Liberty Loans and a Victory Loan in 1919 that required extensive Federal Reserve involvement. US bonds were sold to the public on an instalment plan by member banks; the interest rate banks charged on the unpaid balance on a bond was equal to the coupon rate on the bond. Member banks, in turn, discounted short-term US debt at Federal Reserve banks at an interest rate below the yield on the debt, which allowed them to recover their costs of instalment lending.

US government interest-bearing debt rose from $1.0 billion at the end of 1916 to $25.5 billion at the end of 1919, and would never again fall below $15 billion. This huge increase, and the fact that Federal Reserve banks offered preferentially low interest rates when member banks discounted government debt, had important lasting consequences on the money market. Before the war, Federal Reserve banks had schedules of discount rates that varied across the quality and maturity of discounted paper and the amount of borrowing by a member bank. Because of the low discount rate on government debt, member banks almost exclusively offered it as collateral when borrowing. The discount rate effectively became the rate charged on government debt. By 1922 each reserve bank effectively had a single discount rate, but rates still varied across Federal Reserve districts.

The November 1918 armistice brought new challenges. Continuing shortages of food and other goods in Europe and large increases in the stock of money led to inflation in the United States. The rate of inflation peaked in May 1920 and was followed by a sharp deflation in the following year of about 45 per cent in wholesale prices. In that year industrial production fell by about 30 per cent and unemployment soared. Until October 1919 Federal Reserve banks were obliged to keep the low wartime discount rates in order to allow banks and the public to absorb the 1919 Victory Loan. In November, Federal Reserve banks began raising their discount rates in an effort to combat inflation. In June 1920 four banks raised the rate to seven per cent. Amplifying the effects of the interest rate increases was an outflow of gold to Europe and a sharp reduction in discount window borrowing as Federal Reserve banks cut back on subsidizing the public’s instalment purchases of US bonds.

The Boston bank lowered its rate from seven per cent to six per cent in April 1921, and was gradually followed by other reserve banks in an effort to respond to the slowdown. Deposits at all member banks reached a local maximum of $26.1 billion in the December 1919 call report and then fell to $22.8 billion in the April 1921 report. Discount window borrowings reached a year end high of $2.7 billion in December 1920 and then fell to $0.6 billion at the end of 1922 as gold flows turned positive. As gold flowed in, reserve banks lowered their discount rates to 4.5 per cent in 1923 and early 1924.

While gold inflows slackened after 1923, it became apparent that new operating guidelines were needed. Governor Strong understood that the real bills doctrine was invalid and that many countries were not acting according to the old gold-standard rules. As interest rates fell, most reserve banks were again acquiring securities to augment their income. Strong, on the other hand, had begun to sterilize the New York bank’s holdings of gold by selling its securities in the open market. The Treasury was concerned that reserve bank trading was upsetting securities markets when it was buying or selling debt. In May 1922 the reserve banks established the Governors Executive Committee consisting of the governors of the Boston, Chicago, Cleveland, New York, and Philadelphia banks to manage transactions for all 12 banks. The committee executed orders on behalf of the banks in the light of Treasury plans and made recommendations, but acted only as agents and had no executive power. In April 1923 it was renamed the Open Market Investment Committee (OMIC), which had the same membership as its predecessor but was required

to come under the general supervision of the Federal Reserve Board; and that it be the duty of this committee to devise and recommend plans for the purchase, sale and distribution of open-market purchases of the Federal Reserve Banks in accordance with…principles and such regulations as may from time to time be laid down by the Federal Reserve Board. (Chandler 1958: 227–8)

Strong dominated the OMIC and began to understand the way open-market operations worked. He noted in particular that the sum of reserve bank open-market purchases and gold inflows almost equalled negative changes in member bank borrowing. He developed a case for active monetary policy and argued that restrictive monetary policy should be initiated with open-market sales and followed by increases in the discount rate. This was the likely origin of member bank borrowings and nominal interest rates as indicators of monetary policy. Policy instruments were open-market operations and the discount rate. While proposals to change discount rates originated with Federal Reserve banks, they required Board approval, which may explain why Strong preferred to lead with open-market operations. Strong was sensitive to the effects of monetary policy on prices, but objected to any legislated targeting of prices. His analysis was seriously incomplete when banks were not net borrowers from the Federal Reserve, and in such circumstances so were his policy tactics. Tragically, beginning in 1916 Strong suffered from recurrent attacks of tuberculosis and would die in October 1928, before such circumstances arose.

The 1923 Board Annual Report advocated an activist policy, but continued to support the real bills doctrine. In response to pressure from the Treasury and the Board, Federal Reserve banks sold most of their government securities in 1923; yearend holdings fell from $436 million to $134 million between 1922 and 1923. Federal Reserve notes and member bank reserves backed by such assets were unjustifiable under the doctrine, and the Treasury objected to Federal Reserve banks profiting from such assets. However, at the end of 1924 the banks held $540 million, and the banks’ portfolio of government securities fluctuated considerably in the following years in response to changes in the volume of discounted bills and gold flows. Discount rates at Federal Reserve banks were lowered in the latter half of 1924 and 1925 before converging on four per cent at the beginning of 1926, largely following short-term interest rates in New York. Short-term market rates fell because of a sharp recession; the Federal Reserve index of industrial production (1997 = 100) fell from 7.84 in May 1923 to 6.43 in July 1924. Clearly policy was active, but not because of the real bills doctrine!

The discount rate was four per cent in June, when Federal Reserve banks began to cut the rate to 3.5 per cent and to make open-market purchases. At the beginning of 1928 discount rates were increased because of developing speculation in the stock market and continued to rise to as much as six per cent in October 1929, when the stock market crashed. In part, Federal Reserve discount rates were again responding to changes in industrial production, which had been quite sluggish until the end of 1927 and then began to grow rapidly until July 1929. In part, the 1927 rate cut reflected Federal Reserve efforts to help the United Kingdom maintain sales of gold at the pre-war sterling price, which had been restored in 1925. Governor Strong and Montagu Norman, the Governor of the Bank of England, were working to reestablish a gold standard that could restore order to international finance. To help the United Kingdom in 1925, the New York bank extended the Bank of England a $200 million gold credit and attempted to keep interest rates low in New York relative to those in London. By reopening gold sales at the pre-war price, Britain had effectively revalued the pound upward in 1925 by about ten per cent, with devastating consequences for its economy.

As Strong’s health failed in 1928, a leadership vacuum developed. In an attempt to coordinate policy among all 12 reserve banks and the Board, the Board proposed in August 1928 that the five member OMIC be replaced by a new Open Market Policy Committee (OMPC) that included all 12 reserve bank governors and was chaired by the Governor of the Federal Reserve Board. This proposal was rejected by bank governors, but a modified form was adopted in January 1930. Strong had been aware of growing stock market speculation and did not object to Federal Reserve open-market sales and the increase in the discount rate. These actions were reinforced by outflows of gold. In mid-1928 gold flows reversed, apparently attracted by high and rising short-term interest rates. Federal Reserve banks continued to sell bills and government debt, forcing member banks into the discount window to the extent of about $1 billion in the second half of 1928 and in the middle of 1929. At the end, Strong was aware of the danger of restrictive monetary policy actions over an extended period on the real economy, but remained reasonably optimistic that the situation could be controlled (Chandler 1958: 460–3). After his death the struggle for control continued between his successor at the New York bank, George L. Harrison, and the Board; the latter argued that the real bills doctrine was not dead and that reserve banks should take direct action to penalize member banks making loans that supported security speculation. The Federal Reserve index of industrial production peaked in July 1929, Bureau of Labor Statistics (BLS) wholesale and consumer price indices had been slowly falling since 1926, and in October the stock market collapsed.

The Great Depression

Led by the New York bank, the Federal Reserve flooded the money market with cash by aggressively buying government securities. Discount window borrowing by member banks fell from $1037 million in June 1929 to $632 million in December and to $271 million in June 1930. Further, discount rates at reserve banks were rapidly reduced; at the New York bank the rate was lowered from six per cent in October to 2.5 per cent in June 1930. The monthly average Standard and Poor common stock index (1935–1939 = 100) began to stabilize; it was 195.6 in January 1929, 237.8 in September, 159.6 in November, and 191.1 in April 1930. However, the index of industrial production continued to fall after the open-market purchases, and the BLS index of wholesale prices was ten per cent lower in 1930 than in 1929.

In mid-1930 reserve banks sharply reduced their purchases of government securities in the belief that monetary policy was adequately expansionary. The OMPC seems to have been guided by what Meltzer (2003: 164) calls the Riefler–Burgess Doctrine: ‘If [discount window] borrowing and interest rates were low, policy was easy; if the two were high policy was tight.’ An interpretation is that if member banks wanted to lend they could have inexpensive and relatively easy access to funds; if not, there was little more that the Federal Reserve could do. While total member bank discount window borrowing was positive, many banks were holding excess reserves. Conventional wisdom has it that the reserve banks should have continued buying securities. However, it is unclear even today whether continued large open-market purchases by the Federal Reserve would have had much of an impact on real economic activity in late 1930; the experiment was never tried. Rapid expansion of reserves and member bank deposits did occur in the late 1930s, with little effect on real economic activity.

On average about 600 bank failures a year occurred between 1920 and 1930; most failing banks were small and not members of the Federal Reserve System. The number of failing banks doubled in 1930 and increased by another 70 per cent in 1931. The total deposits of failing banks between 1920 and 1930 averaged less than $200 million a year, but more than quadrupled in 1930 and doubled again in 1931. Total deposits and currency had begun to fall after December 1928 and continued to fall after the stock market crash. Currency in circulation began to rise in November 1930, as bank failures increased. Industrial production and wholesale prices were falling at an accelerating rate. The directors of the New York bank counselled Governor Harrison to continue open-market purchases in 1930, but he encountered opposition in the OMPC and little was done. Net gold inflows were offset by open-market sales because the OMPC collectively believed monetary policy was expansionary. Reserve bank discount rates and money market interest rates trended down until 21 September 1931, when the United Kingdom suspended gold payments.

The British abandonment of gold led to very large withdrawals of gold and currency from the United States that were initially partially offset by open-market purchases of bills and increased discount window borrowing, which occurred at sharply higher interest rates as recommended by Bagehot (1873). However, Federal Reserve bank credit fell from $2.2 billion in October 1931 to $1.6 billion in March 1932. During this period of rising bank failures, rapidly declining economic activity, and falling prices, Harrison argued against open-market purchases for a number of reasons, but primarily because of the possibility of a shortage of ‘free gold’, that is, gold that was not required as collateral for Federal Reserve notes and reserves. The Glass–Steagall Act of 1932 authorized the Federal Reserve banks temporarily to use US government securities as collateral for Federal Reserve notes and thus largely solved the problem of a lack of free gold. In February 1932 Federal Reserve banks began aggressive open-market purchases of government securities that more than offset continuing gold losses and allowed member bank borrowings to fall about 50 per cent by August 1932. Discount rates at the New York and Chicago banks were lowered to 2.5 per cent in June 1932, but all other banks kept their rates at 3.5 per cent until the national banking ‘holiday’ that began on 5 March 1933 when President Roosevelt closed all US banks. Net free reserves (excess reserves minus discount window borrowing) had turned positive in September and thus signalled excessive ease to some individuals on the OMPC.

Restructuring the Federal Reserve System

It was obvious that the Federal Reserve had been ineffective in combating the collapse of the banking system and responding to the Great Depression. The banking system and the Federal Reserve needed to be restructured and strengthened. The Emergency Banking Act of 9 March 1933 authorized the Treasury to license and reopen national banks that were judged to be sound; state chartered banks that were sound would receive licences from state banking commissioners. Many reopening banks received capital injections by selling preferred stock to the Reconstruction Finance Corporation. At year end 1929 there were 24,026 commercial banks of which 8522 were members of the Federal Reserve System; at year end 1933 there were 14,440 commercial banks of which 6011 were member banks. For a period of one year all banks, whether members or not, could borrow on acceptable collateral from Federal Reserve banks.

Many of the reforms that were adopted would survive at least until late in the 20th century. Because of a belief that the collapse lay in undisciplined stock market trading, the Glass–Steagall Act of 1933 required that commercial banks divest themselves of investment banking activities. This act introduced deposit insurance that became effective in January 1934. It also banned interest payments on demand deposits and allowed the Board to impose ceilings on interest rates that banks could pay on time and savings deposits. Finally, the act renamed the OMPC the ‘Federal Open Market Committee’ (FOMC), but as in earlier incarnations its executive committee remained the same. The Securities Exchange Act of 1934 authorized the Board to impose margin requirements on stock market trades. Federal Reserve banks were authorized to make commercial and industrial loans to non-financial firms.

Having failed to expand reserve bank credit between July 1932 and February 1933, the Board found itself under extraordinary political pressure to expand resources to the banking system. As Meltzer (2003: 435–41) explains, President Roosevelt threatened to have the Treasury issue currency in the form of greenbacks if the FOMC failed to expand sufficiently. Net free reserves turned positive in May 1933 and rose to more than $3.0 billion by January 1936. The revaluation of gold in February 1934 together with subsequent large gold inflows from Europe and hesitancy to lend by member banks contributed to this surge in excess reserves.

The reconstruction of the Federal Reserve System continued with Roosevelt’s nomination of Marriner Eccles to become Governor of the Federal Reserve Board in November 1934. Eccles had argued that system power should be concentrated in the Board and that reserve banks be prevented from undertaking open-market operations on their own accounts. Eccles’s initiatives were opposed by Senator Carter Glass, many reserve bank governors, and the banking industry, but he largely succeeded in achieving his goals. The reforms were in the Banking Act of 1935, which restructured the Board to consist of seven appointed governors, each with a staggered 14-year term. The FOMC was restructured to consist of the seven governors and five reserve bank presidents. Two of the governors were to be appointed for four year terms as chairman and vice-chairman of the Board by the president, with the advice and consent of the Senate. Eligible paper was no longer restricted to being short-term paper that originated in commerce and industry. The Board was empowered to vary reserve requirements; the upper limit was twice the percentages that were specified in the 1917 amendments to the Federal Reserve Act.

Members of the renamed Board of Governors of the Federal Reserve System took office in February 1936, with Eccles as chairman. For some time the FOMC had expressed concern about the inflationary potential of large excess reserves. In particular, because excess reserves exceeded reserve bank credit, the FOMC would not be able to absorb them without an increase in reserve requirements. Employing its new policy instrument, on 14 July 1936 the Board announced an increase in reserve requirements on August 15 of 50 per cent on all deposits at member banks. The increase was expected to absorb less than half of system excess reserves and was not expected to impinge on member bank lending or the economic recovery. In part because of continuing gold inflows, excess reserves were $3.0 billion at the end of July 1936, and averaged about $2.0 billion through the end of February 1937. Because excess reserves continued to be large, the Treasury began to sterilize gold inflows in December 1936, but not to the extent desired by the Board. At the end of January the Board announced a further two-step increase in reserve requirements of one-third to take place in March and May 1937. These actions took reserve requirements to their legal maxima and reduced excess reserves to below $800 million in summer months. In August and September reserve banks reduced their discount rates to one per cent or 1.5 per cent, levels that would last until December 1941. Coinciding with the May increase, the industrial production index.

(1997 = 100) reached a high of 10.4 and then decreased to 7.0 in May 1938. Continuing gold inflows and the Treasury’s February 1938 abandonment of gold sterilization allowed excess reserves to increase to $1.5 billion in March 1938. Beginning after the Board’s reduction in reserve requirements of more than ten per cent in April 1938, excess reserves began a rise to nearly $7 billion in late 1940; however, industrial production did not pass its 1937 peak until October 1939, after the Second World War had begun in Europe.

Second World War and Recovery

As the war approached gold flowed into the United States, and the FOMC allowed its security holdings to fall and their maturity to lengthen. In response to inflationary pressures, the Board introduced consumer credit controls in September 1941 and again raised reserve requirements to their legal maxima in November. After the United States declared war, monetary policy was constrained to facilitate war finance. In April 1942 the FOMC set interest rate ceilings on treasury bills at 0.375 per cent and on long-term bonds at 2.5 per cent. The yield curve was upward-sloping and effectively ‘pegged’ by these two boundary conditions into the post-war period. Because capital gains could be earned by buying high coupon securities and selling as they approached maturity, the cost of intermediate term debt was higher than rates shown on the yield curve. Discount rates were lowered to one per cent by all reserve banks and were not raised again until 1948. A preferential discount rate of 0.5 per cent was charged for loans collateralized by short-term US debt. Reserve requirements for central reserve city member banks were lowered in 1942, causing interest-free reserves to disappear into interest-bearing US securities. Finally, a variety of selective credit controls were imposed during and after the war, which ended in August 1945.

Yearend deposits and government securities of member banks had risen from $61.7 billion and $19.5 billion in 1941 to $129.7 billion and $78.3 billion respectively in 1945. Because of the pegging of the yield curve, Federal Reserve bank yearend ownership of US securities rose from $2.3 billion in 1941 to $24.3 billion in 1945; treasury bills were $10 million in 1941 and $14.4 billion in 1946.

The preferential discount rate was eliminated in the spring of 1946. In July 1947 the FOMC relaxed the rate ceiling on treasury bills and the rate rose to about one per cent by yearend. Reserve banks raised the discount rate to 1.25 per cent in early 1948. Eccles’s long term as chairman ended in February 1948, but he continued as a member of the Board. Reserve requirements were increased in 1948 as the Board sought to control inflation, although prices were actually falling at yearend when a recession occurred. Indeed, the reserve requirement policy instrument was used many times between April 1948 and February 1951 because it was perceived not to have a direct effect on treasury interest rates. A continuing struggle between the Board and the Treasury for an independent monetary policy would not be resolved until a spurt of inflation after the start of the Korean War led to an accord signed on 4 March 1951. It effectively freed the Board from pegging interest rates. Partly because of frictions leading to the accord, a new chairman, William McChesney Martin, Jr., was appointed in April.

Resumption of Discretionary Monetary Policy

In the Martin era of discretionary monetary policy, new operating techniques were needed. In 1953 the FOMC settled on a policy of ‘bills only’, which meant that open-market operations would be largely confined to the market for treasury bills, because it was recognized that large policy actions in thin markets could impair market efficiency. Indicators of monetary policy continued to be net free reserves and market interest rates. Because evidence was lacking that interest rates had much effect on private sector investment, a new paradigm, the ‘availability of credit’ doctrine, was used to rationalize the transmission of policy actions to the real economy. It argued that banks rationed credit to marginal borrowers when restrictive policy led to rising interest rates or indebtedness at the discount window. With these adjustments the FOMC vigorously and unsuccessfully pursued goals of lowering inflation and combating unemployment in the turbulent decade of the 1950s. In that decade there were three business cycles, which were marked by successively rising peaks of interest rates, inflation, and unemployment. The reason for this failure was thought to be inflation-induced rising marginal rates of taxation, which were addressed by large tax cuts in the following decade.

As interest rates rose, the opportunity cost of holding excess reserves rose, which led to the reappearance of a federal funds market in which banks traded reserves. Because banks paid no interest on demand deposits, there was also rapid expansion of the market for commercial paper in which large firms with good credit ratings traded idle funds without the direct intervention of banks. Both markets had atrophied after the 1920s because of low interest rates, and served to change the relation between open-market operations and real economic activity. They were precursors of a wave of innovations that would have similar effects in the coming decade. These included large-denomination negotiable certificates of deposit, one-bank holding companies, offshore ‘shell’ branches, the Eurodollar market, and bankrelated commercial paper.

Beginning in 1961, the Kennedy administration attempted to coordinate fiscal and monetary policy by proposing large tax cuts to encourage investment and economic expansion. A new problem was that the United States was experiencing large gold outflows as the world continued to recover from the world war. To cope with this new approach and problem, the FOMC was encouraged to abandon its bills-only policy and to attempt to twist the yield curve by buying long-term bonds and selling bills. As short-term rates rose the Board repeatedly raised the ceiling on interest rates that banks could pay on time and savings deposits. It was argued that lower long-term interest rates would encourage capital formation and that higher short rates would discourage foreign interests from converting dollars into gold, as they were entitled to under the Bretton Woods agreements. These efforts were not successful in discouraging gold outflows, but investment and the economy expanded strongly. In 1965 the Board introduced a Voluntary Foreign Credit Restraint programme, which discouraged banks from overseas lending that was not financing US exports. Nevertheless, gold continued to flow out and the requirement that Federal Reserve notes and reserves be backed by gold was cancelled in 1968. Large open-market purchases had been needed to offset gold losses.

Policy coordination between the Board and the new Johnson administration effectively ended in December 1965, when the Board approved an increase in the discount rate because of inflation arising from mobilizing for the Vietnamese War. Net free reserves had turned negative in 1965 and were increasingly so until late 1966. Short-term interest rates rose until October. Higher rates increased the cost of the mobilization and had devastating effects on residential construction and the savings and loan associations and mutual savings banks (hereafter thrifts) that financed it, because in September Congress passed legislation limiting interest rates that thrifts could pay on time and savings accounts. These limits meant thrifts would experience withdrawals of funds or ‘disintermediation’ because depositors switched funds to government securities, which had no limits. This policy transmission channel would soon disappear because Congress and the administration could not withstand the resulting political pressures. In 1968 the Federal National Mortgage Association was privatized and in 1970 the Federal Home Loan Mortgage Corporation was created. Both bypassed depository institutions by securitizing mortgage loans. Banks also responded to Board policies and restrictions on innovations by opening overseas offices that were not subject to them. A ten per cent income tax surcharge in 1967 was insufficient to stop inflation, and short-term interest rates rose to new highs in January 1970, when Chairman Martin’s term ended. Net free reserves averaged about a negative $1 billion between May 1969 and July 1970. A decrease in short-term interest rates followed the then largest-ever US bankruptcy of the Penn Central Transportation Company in June 1970, but led to large new capital outflows in 1971 that pressured the dollar. The FOMC responded by forcing short-term rates and net borrowed reserves up again.

Towards Flexible Exchange Rates

The amplitude of changes in interest rates increased between 1965 and 1971, and the United States experienced a recession in 1970. As in the 1950s the Federal Reserve was unable simultaneously to achieve satisfactory unemployment, inflation, and exchange rate outcomes. Many of the Board’s policy instruments, such as the discount rate, reserve requirement changes, and many regulations had effectively been disabled by innovations, so that only open-market operations were available to achieve multiple targets. For example, an increase in reserve requirements induced banks to resign from the system or to conduct more of their business overseas. One exception to this loss of powers was the 1970 amendments to the Bank Holding Company Act, which finally gave the Board regulatory authority over one-bank holding companies. In August 1971 the Nixon administration, with new Board Chairman Arthur F. Burns as an advisor, announced a 90-day freeze on prices and wages, suspension of gold sales, and several other major changes in the United States. The suspension of gold sales led to a floating exchange rate system, devaluation of the dollar, and sharp rises in dollar-denominated prices in international markets. The shift from a fixed to a floating exchange rate system is likely to have increased the potency of monetary policy, as was predicted by Mundell (1961). The FOMC responded to consequent high inflation by driving nominal short-term interest rates to very high levels in 1973 and 1974, which helped to induce a severe recession beginning in August 1973, but were inadequate because on average the real federal funds interest rate (calculated with the GDP deflator) was negative between the end of 1973 and 1978. Real estate and other durable goods prices rose relative to the GDP deflator, and the international value of the dollar fell. After the resignation of President Nixon in 1974, Congress required the Chairman to explain policy in semi-annual public hearings and report the FOMC’s targets for two money stock measures: M1, a measure of transactions balances, and M2, a measure of liquid assets. Friedman and Schwartz (1963) had recommended using money as an indicator of monetary policy instead of interest rates or net free reserves.

Part of the explanation for the policy failure was continuing financial market innovation. Foreign banks operating in the United States grew rapidly and were unregulated until the 1978 International Banking Act, which placed them under Board supervision. The introductions of money market mutual funds (MMMFs) and negotiable order of withdrawal (NOW) accounts in 1972, the Chicago Board Options Exchange in 1973, and financial futures markets in 1975 again began changing the relation between financial and real markets. A more important change was the rapid expansion of repurchase agreements after 1970. In a repurchase agreement, a client’s deposits are borrowed to finance a bank’s or dealer’s inventory of government securities, often only overnight. Large bank holdings of government securities often represented transactions balances of large corporations and state governments that could not easily be controlled.

The real federal funds rate turned distinctly positive in the third quarter of 1979 when Paul A. Volcker became chairman. In early October he announced that the FOMC would no longer limit fluctuations in short-term interest rates and would use open-market operations to control bank reserves. This was a major policy change from practices dating from the 1951 accord. Further, he imposed eight per cent marginal reserve requirements on non-deposit liabilities, that is, Eurodollar borrowing, federal funds purchased from non-member banks, and funds acquired through repurchase agreements. These vigorous actions together with large income tax cuts by the Reagan administration between 1981 and 1983 drove real short-term interest rates to levels not seen since the early 1930s and caused MMMFs to grow rapidly. In only two quarters between 1979 and 1986 was the average real federal funds less than five per cent. These high rates caused the trade-weighted value of the US dollar to appreciate by 87 per cent between July 1980 and February 1985, which savaged US exports and attracted imports with adverse consequences for US manufacturing.

Financial Deregulation

The landmark Depository Institutions Deregulation and Monetary Control Act was signed by President Carter at the end of March 1980. It radically changed the Federal Reserve System by eliminating the significance of membership in the system. After an 8 year phase-in period, all depository institutions would be subject to uniform reserve requirements on demand and time deposits, although the requirement on the first $25 million of transactions deposits was less than that on other transactions deposits. The Board could vary reserve requirements. All depository institutions had access to reserve bank discount windows. This strengthened the system because banks could no longer threaten to leave it in order to get the lower requirements that many states imposed. Further, Federal Reserve banks were required to charge banks for the cost of services they provided. Before this act they had been giving away services as an inducement for banks to stay in the system. This pricing requirement in turn forced depository institutions to begin to charge their clients for services, which changed the way banking services were used. The act mandated that interest rate ceilings on time and savings accounts be eliminated after six years, increased deposit insurance, and had other important provisions that are beyond the scope of this discussion.

In late 1980 the Board announced that transfers from overseas branches to the United States could be treated as collected funds on the day they were transferred. Before then, transfers in a day were not ‘good funds’ until the following day. The expansionary effects of this change, rapidly growing repurchase agreements, and other innovations are evident in demand deposit turnover statistics that the Board reported from 1919 until August 1996. Turnover is the annualized value of all withdrawals from deposit accounts divided by aggregate deposit balances.

High interest rates were savaging thrift institutions, which had negative gaps (more fixed-rate assets than fixed-rate liabilities on most future dates), and allowed MMMFs to expand rapidly. Congress intervened in September 1982 by passing the Garn–St Germain Act, which provided temporary emergency assistance and among other changes introduced money market deposit accounts and super NOW accounts, which paid market interest rates. MMMF growth was slowed by this act, but the weakening condition of banks and thrift institutions would result in large numbers of failures as the decade wore on. Large banks also experienced large losses because the appreciating dollar had resulted in failures of sovereign states, especially in Latin America, to meet their loan obligations. Chairman Volcker was heavily involved in negotiating solutions for these defaults.

The restrictive monetary policy resulted in the deepest recession since the Depression; the unemployment rate was 10.8 per cent at the end of 1982. At the end of Volcker’s term in August 1987 the unemployment rate had fallen to six per cent and the consumer inflation rate was less than two per cent. Real interest rates had fallen from 10.5 per cent in mid-1981 to four per cent, and the trade-weighted value of the dollar fell correspondingly. Volcker’s February 1987 statement of monetary policy objectives to the Congress reported that M1 was not a reliable indicator of monetary policy and would be de-emphasized.

While his successor, Alan Greenspan, inherited a much improved economy, many problems remained from a rising wave of bank failures and the collapse of thrift institutions. Real estate markets were especially disorderly when the thrift crisis was resolved beginning in 1989 and were further distorted by provisions in the Tax Reform Act of 1986, which disallowed many interest tax deductions. After 1990 interest on home loans was effectively the only deductible interest on individual income tax returns. In addition, a collapse of stock prices in October 1987, strong foreign demand for US currency associated with the collapse of the Soviet Union, and a recession at the end of 1990 presented further challenges. The FOMC responded to these challenges by varying the real federal funds rate, defined using the contemporaneous GDP price deflator inflation rate. This rate fell sharply for two quarters after the stock market crash, rose before falling for two quarters after a second stock market dip in October 1989, and then began to fall in the fourth quarter of 1990. In July 1993 testimony before Congress, Greenspan disclosed that the FOMC was downgrading M2 as an indicator of monetary policy and, as could have been surmised from its actions, that an important guidepost was now real interest rates. The real federal funds rate averaged less than one per cent in 1993. In early 1995 it had risen to four per cent and held that value as an average until the collapse of a large hedge fund in September 1998. After the fallout from the hedge fund collapse had been resolved, the real federal funds rate was restored to an average of about four per cent in 2000. When a new recession appeared in 2001 together with a sustained large collapse in stock market prices, the real federal funds rate was lowered to near zero in the fourth quarter; the rate had averaged zero for 13 consecutive quarters as of March 2005.

Between December 1990 and April 1992 reserve requirements on time and demand deposits were reduced, which helped banks to increase net income. In January 1994 ‘retail sweep programmes’ were introduced. In these programmes, a bank shifts funds from a depositor’s transactions account to a synthetic time deposit account in the depositor’s name in order avoid reserve requirements, usually without the depositor’s knowledge. The Board does not measure the amount of funds swept, except at the time the programme was established. The Board estimated that as of August 1997 required reserves fell by one-third because of these programmes.

In November 1999 President Clinton signed the Financial Services Modernization (Gramm–Leach–Bliley) Act, which reversed the 1933 Glass–Steagall Act’s ban on combining commercial and investment banking. The ban had been eroding since 1987, when some large bank holding companies were authorized by the Board to establish subsidiaries that could underwrite state and local government revenue bonds. The new act authorized the establishment of financial holding companies, which were to be regulated by the Board and could engage in an approved list of activities that included commercial banking, insurance, securities underwriting, merchant banking, and complementary financial undertakings. In 2003 there were more than 600 financial holding companies, which resemble the universal banks that exist in other countries.

In December 2002 the Federal Reserve discarded the discount rate as a policy instrument by replacing it with an interest rate on primary credit extended by the discount window that is one per cent above the FOMC target federal funds rate. Primary credits are collateralized loans to banks in sound financial condition.

As the foregoing dramatic institutional changes suggest, the Federal Reserve System is a work in progress. Its set of policy instruments and its dimensions have radically changed. Because of offshore banking facilities and retail sweep accounts, reserve requirement changes are no longer an effective policy instrument. As noted in the preceding paragraph, the discount rate has been discarded as an instrument; it is simply a penalty rate that is related to a bank rate, as is often the practice in other countries. Regulations on the interest rates banks pay on time and savings deposits have been discarded. Open-market operations are almost the sole policy instrument that can be used to achieve the Board’s target nominal and real federal funds interest rates. While the FOMC has been able to control the overnight federal funds rate, the linkage between it and real economic activity is changing. First, the combined holdings of US government securities by foreign central banks have recently exceeded those of Federal Reserve banks. Foreign central bank holdings are partly a result of their efforts to manipulate exchange rates; their holdings are likely to change when FOMC policies change. Second, repurchase agreements and offshore transactions vary considerably over time and their volumes appear to be sensitive to US economic activity. Third, the outstanding stock of securitized mortgage and other debt has been growing rapidly; such debt is a close substitute for US government debt and its amount has real economic effects. Fourth, because of decreasing required reserves and growing offshore holdings of US currency, 89 per cent of Federal Reserve liabilities were in the form of Federal Reserve notes in December 2003; the corresponding share was 34 per cent in 1941, 57 per cent in 1970, and 79 per cent in 1989. In part, the Federal Reserve recently has become an institution for collecting seigniorage from the rest of the world. Finally, over the decade ending in 2003, the share of all credit market assets held by depository institutions in the Federal Reserve’s flow of funds accounts fell. In the context of the most recent 13 quarters of a zero real federal funds interest rate, more changes could be expected.

See Also