A History of the Federal Reserve, Volume 1
Page 74
The Economy at the 1936 Election
The August increase had no perceptible effect on the economy in 1936. Expansion was robust as the country approached the presidential election. Industrial production increased 17 percent in the year ending in October, just before the election. Balke and Gordon’s (1986) GNP data show 9 percent growth and 1.9 percent inflation for the four quarters of 1936. Contemporary data show national income produced in 1936 rising 15 percent, with wholesale prices almost unchanged (Barber 1996, 98–99). Based on these data, income had reached 80 percent of the 1929 level, but population and economic potential had increased since 1929, so there was considerable idle capacity. Currie estimated potential output at full employment as $85 billion to $90 billion. Using those values, national income was about 65 to 70 percent of its full employment level, but the unemployment rate was 17 percent (Memo, Board of Governors File, May 18, 1936). The private sector created fewer than 30percent of the 5 million new jobs in 1936 (Barber 1996, 99). The rest were jobs in relief agencies like the Works Progress Administration (WPA).
partners with the treasury Gold inflows continued, influenced in part by fears in Europe, in part by the gold price and economic expansion. By the time the new reserve requirement ratios took effect, some of the expected decrease in excess reserves had been offset by the gold inflow. The monetary base fell (after adjusting for reserve requirement ratios). For the last six months of 1936, the base remained about 10 percent below the previous year.
179. Davis had joined the Board two weeks before, so he voted no because he lacked information. McKee wanted to postpone the decision until September.
180. Eccles claimed that in April Morgenthau agreed to the change. By August 15 the increase in required reserves was $1.79 billion, larger than the staff estimate.
Morgenthau’s tongue lashing on July 15 was followed by efforts to improve the working relationship. Eccles complained that Morgenthau was very secretive about gold operations and did not inform the Board. Even New York (as fiscal agent) was better informed. Morgenthau agreed to release weekly data to the Board on net purchases and sales by the Exchange Stabilization Fund. In return, he asked Eccles to help with bond market stabilization. The Treasury had bought heavily to keep prices of recent issues above par. He asked Eccles to participate in the purchases. The Treasury would henceforth make purchases in the open market, instead of through the New York bank. At the end of each day, the open market committee could decide to take half the amount purchased. Eccles checked with the committee and agreed to the new arrangement. This arrangement made the Federal Reserve an adjunct of Treasury or, as Morgenthau put it, the Treasury’s partner (Blum 1959, 358).
Morgenthau also urged Eccles to make an open market purchase or sale of $50 million in December. He thought the public should be accustomed to the idea that open market operations would be used, and it was best to get the market accustomed to purchases and sales after three years of inaction.
Gold Sterilization
By the end of October, excess reserves were above $2 billion, again almost equal to the size of the open market portfolio. The sustained gold inflow had three effects that worried the president and others.
First, foreigners bought United States securities, contributing to a rapid increase in stock prices. Total return to common stocks was 47.7 percent in 1935 and 33.9 percent in 1936. By the end of 1936, the total return on common stock since 1929 was again positive (Ibbotson and Sinquefeld 1989, 160–61). Second, the gold inflow added to reserves and base money, raising the price level. Inflation remained low, however. Consumer prices rose only 1 percent in 1936. Third, the United States was vulnerable to a gold outflow. A particular concern was that in a European war foreign governments would sequester private holdings of foreign securities, sell securities to finance the war, and export gold from the United States.
Similar concerns had arisen before World War I, when the Federal Reserve was unable to prevent a gold inflow, and in the early 1920s, when the Federal Reserve sterilized part of the inflow. Roosevelt wanted something done to remove speculative inflows without reducing long-term investment (Blum 1959, 359).
The November FOMC meeting came in the midst of these concerns. Eccles told Morgenthau that he proposed to sell $300 million to $400 million to offset the increase in excess reserves from August to November. Some FOMC members preferred to again increase required reserve ratios, and some preferred to wait until after the seasonal return of currency to banks in January. Others argued that, although the economy had recovered, the time for reversing policy still lay in the future. The consensus was to wait (Minutes FOMC, November 19 and 20, 1936). In the press release following the meeting, the Board alerted the country to its renewed concern about reserve growth.
Eccles soon shifted his position to favor a second 50 percent increase in required reserve ratios. The Treasury was not enthusiastic.181 Morgenthau searched for an alternative.182
The Treasury staff proposed to sterilize the gold inflow to prevent it from increasing bank reserves and the monetary base. The Treasury would continue to issue gold certificates on receipt of gold. Instead of allowing the gold certificates to increase bank reserves, the Treasury would pay for the gold by selling debt. Later, if foreigners sold securities and withdrew gold, the Treasury could reverse the operation and avoid the deflationary effect. In accounting terms, the transaction differed little from a Federal Reserve sale of debt to the public combined with a gold purchase. The difference was that responsibility for the conduct of the operation remained with the Treasury.183
181. Nor were some reserve bank presidents and their members. An increase in the reserve ratio would make member banks pay the cost of offsetting the gold inflow, discouraging membership by country banks. George Hamilton, president of the Kansas City reserve bank, made this argument in a letter to Eccles after the November FOMC meeting. Eccles’s reply did not respond to this point. Instead, Eccles pointed to the excess reserve holdings of country banks, showing that they held a higher proportion of total to required reserves than other classes of members (Hamilton to Eccles and Eccles to Hamilton, Board of Governors File, box 1450, November 24 and December 5, 1936). Hamilton warned also that “many banks are watching . . . with the idea of dumping [bonds] whenever there is a change made in our policy” (Hamilton to Eccles, Board of Governors File, box 1450, November 24, 1936, 2).
182. Toma (1982) explains the 1936–37 increases, and the recession that followed, as an effort by the Federal Reserve to increase seigniorage. There is no mention of a seigniorage or revenue motive, and as noted, the Treasury was displeased and in 1938 forced a reduction in reserve requirement ratios.
183. The accounts showed a Treasury purchase of gold paid for by drawing on its deposit account at the Federal Reserve and a sale of debt to the public to replenish its deposit. The net effect on the Treasury’s balance sheet is a larger gold stock offset by increased debt outstanding. The Treasury issued gold certificates but held the gold in the “general fund in gold,” part of Treasury cash. These operations neutralized the effect on the monetary base; no additional reserves were created.
Eccles could not make up his mind. He alternated between seeing the proposal as a way to avoid increasing reserves and concern about the shift in responsibility for monetary policy from the Federal Reserve to the Treasury.184 He argued also that the timing was bad; reserve growth and the stock market had slowed. The Board could raise reserve requirement ratios, at no cost to the government, instead of selling short-term debt to sterilize gold inflows. In a letter to Morgenthau he demanded that the policy, if adopted, should be automatic, not left to the discretion of the Treasury to operate monetary policy.
The letter annoyed Morgenthau and led to another in the series of disputes that frequently disturbed their relationship. Eccles withdrew from the agreement to share in the Treasury’s bond market support program. Morgenthau threatened to take control of monetary policy: “I think there is one more issue to be settled . . . that i
s whether the Government through the Treasury should control. . . monetary policy . . . or whether control should be exercised through the Federal Reserve Banks who are privately owned and dominated by individuals who are banker minded” (quoted in Blum 1959, 363).
Eccles then sent a conciliatory letter but followed it on December 10 with a more formal letter explaining that he would endorse the sterilization policy if the Treasury would agree not to run its own discretionary monetary policy (Hyman 1976, 221–22). As usual, Roosevelt listened to the two disputants. Instead of making his own case, Eccles changed sides, endorsed Morgenthau’s case for sterilization, and declared his own preference for sterilization over the Board’s proposal to use open market sales or higher required reserve ratios to neutralize the gold inflow (Hyman 1976, 223; Blum 1959, 364–65).
Roosevelt ordered the sterilization program to begin. On December 23 the Treasury began sterilizing gold inflows and newly mined United States gold. Between December 1936 and July 1937, when gold sterilization ended, gold certificates outstanding increased $1.3 billion and Treasury cash increased by a like amount. Bank reserves rose only $180 million in this period.
The FOMC’s executive committee met on December 21 to discuss the System’s role in smoothing the government securities market. There was general agreement that with short-term rates near zero, much of the market activity was in longer-term securities. Hence the long-term market was now “a huge part of the money market” (Minutes, FOMC Executive Committee, December 21, 1936, 4). The committee agreed that it was responsible for smoothing the market, either alone or in partnership with the Treasury.
184. At the New York bank, Burgess wrote a strong objection to gold sterilization as putting additional monetary control in Treasury (political) hands (Sproul Papers, Excess Reserves, December 9, 1936).
The committee then met with Secretary Morgenthau. They agreed to renew the partnership operation. The Treasury ended its own purchase operations, restoring the role of the manager of the System Open Market Account acting on orders from the Treasury. The Federal Reserve agreed to share in the purchases and sales up to the authority granted by the FOMC, $50 million at the time. If more purchases were needed, the FOMC would meet to discuss what action should be taken.
The Second and Third Increases in Reserve Requirements
By late 1936, short- and long-term interest rates were at the lowest levels experienced to that time. The economy and the stock market continued to recover, and gold stocks were at record levels. Many bankers believed that low rates would not persist in that environment. Strengthening that belief was the almost continuous discussion of policy actions to reduce excess reserves by open market operations or a change in reserve requirement ratios.
Perhaps typical of prevailing attitudes is the letter from a Missouri banker who wrote to the Kansas City reserve bank urging open market sales instead of a higher reserve requirement ratio for country banks.
We are vitally interested in protecting our capital funds from depreciation when the ultimate increase in interest rates comes and brings along a depreciation in longer term securities. This being true the only chance we have to maintain earnings at all is through an increase in volume. Our deposits show a substantial increase but if reserve requirements were again substantially raised, it would limit our resort to this procedure in what seems to me a very serious way. (J. E. Garm to Hamilton, Board of Governors File, box 1450, November 23, 1936; emphasis added)
Morgenthau reported a similar view in his diary. Discussion of future inflation and proposals to increase reserve requirement ratios, he believed, convinced many bondholders that interest rates would rise (Blum 1959, 367). Morgenthau was concerned that higher interest rates would raise Treasury borrowing costs, increasing the deficit, and hurt the economy by reducing investment. He urged the Board to reach a decision before February 1. To help him plan the March 15 bond issue, he wanted the increase to be effective by March 1.
At Vice Chairman Ransom’s suggestion, the Board met with Morgenthau to hear his opinion directly (Board Minutes, January 19, 1937, 2). Morgenthau expressed reservations about a second increase in reserve requirement ratios, but he gave his approval (Blum 1959, 368). A memo from Goldenweiser predicting only small increases in short- and long-term interest rates reassured him (Memo, Goldenweiser, Board of Governors File, box 418, January 12, 1937).185
Eccles and some of the Board’s staff hesitated. On January 25, Currie prepared two memos. One warned that the proposed increase in the reserve requirement ratio for time deposits was too large. The second argued the opposite side at greater length (Currie to Eccles, Board Files, January 25, 1937.186 He concluded that the current stock of money was sufficient to support full employment.
The reserve bank presidents received a briefing from Goldenweiser on the day of Currie’s memos. There is no mention of Currie’s estimates.187 Goldenweiser urged the increase. He expected short-term rates to increase: “Short-term rates had been abnormally low in relation to long-term rates and some stiffening of the former would be desirable” (Board Minutes, January 26, 1937, 3). The Board or the FOMC would have to reduce excess reserves at some time in the future, and he believed that the “most effective time for action to prevent the development of unsound and speculative situations is in the early stages of such a movement when the situation is still susceptible of control . . . [S]uch a time had arrived” (3).188
Goldenweiser added that aggregate excess reserves of $2.1 billion could absorb the $1.5 billion increase in required reserves. However, 2,435 banks would have to draw on correspondent balances, and 197 would have a reserve deficiency that would require borrowing or asset sales (ibid., 4). He also dismissed concerns about loss of membership. John H. Williams reinforced Goldenweiser’s arguments and urged prompt action. The longer the Board delayed, the greater the likelihood that future action would force liquidation of loans.
A majority of the presidents spoke in favor.189 The following day, Goldenweiser assured the FOMC that the increase in required reserve ratios would not reverse the easy money policy but would place the System in a position to influence the market by open market operations when needed. Three days later, Eccles reported that Morgenthau had again not opposed the change, provided it was effective no later than the close of business on February 27 so that the market could adjust before the March 15 financing. Eccles and Morgenthau then discussed the issue with the president. Roosevelt left the decision to the Board but did not object to the increase (Board Minutes, January 28, 1937, 4).
185. Goldenweiser argued that rates on Treasury bills would be held down by rates of 0.5 percent on banker’s acceptances and that rates on long-term bonds would remain low until short-term rates equaled or exceeded long-term rates (Board of Governors File, box 418, January 12, 1937, 3, 5). The prediction proved to be wrong.
186. Currie also computed the estimated nominal value of national income three years ahead, based on estimates of velocity and his belief that the price level would rise by 10 percent as the economy returned to full employment in 1939.
187. However, Goldenweiser dismissed the argument that time deposits be exempt from the increase.
188. This is probably a reference to a revised view of the 1927–29 stock market speculation.
189. Harrison again proposed that reserve banks be given emergency powers to purchase and sell securities in amounts up to $50 million without prior approval. The FOMC postponed discussion until January 26. Eccles opposed the motion, but it passed six to five, with Governor Broderick voting with the five presidents. Broderick then changed his vote to abstain on grounds that motions of this character should not be carried by such a narrow margin (Minutes, FOMC, January 26, 1937, 15–16).
Governor McKee proposed that the increase be made in two steps, half at the end of February and half in April or May. Eccles later asked Morgenthau and Burgess about this suggestion. Both found it acceptable. The following Saturday, January 30, 1937, the Board increased reserve requirement ratio
s by 331/3 percent of prevailing levels. The vote was five in favor, one (McKee) not voting. Deferring a bit to Treasury concerns, only half the increase became effective on March 6. The rest was scheduled for May 1. Table 6.4 shows the changes.
The Board’s press release emphasized that policy had not changed and affirmed its view that the $1.5 billion of excess reserves was superfluous: “Member banks will have excess reserves of approximately $500 million, an amount ample to finance further recovery and to maintain easy conditions” (Press Release, Board of Governors File, box 291, January 30, 1937, 2). The release cited the earlier experience, warned about the risks of inaction, and repeated its earlier conclusion: “It is far better to sterilize a part of these superfluous reserves while they are still unused than to permit a credit structure to be erected upon them and then to withdraw the foundation of the structure” (4).
The Board had now used all of its new authority to raise reserve requirements. With gold sterilization limiting increases in reserves and an open market portfolio five times the estimated volume of excess reserves, the Board believed it had the power to control future inflation.
Burgess met with Morgenthau and the Treasury soon after the announcement. There were no complaints. The main discussion concerned Treasury issues in March and June.190