A History of the Federal Reserve, Volume 2 (50 page)

The minutes for the 1950s show that Martin, several other members, and principal Board staff believed that to prevent long-run inflation, the Federal Reserve had to keep average money growth close to the average growth rate of real output. They recognized that short-term changes in money and velocity were erratic, so money growth could not be used for short-term money market management. And they had no theory of the demand for money or velocity to help them separate changes in the demand for money from changes in supply. Nevertheless, Federal Reserve statements gave more importance to avoiding inflation by keeping money growth close to growth of output than in earlier or later decades. The policy was generally successful, more successful by far than the alternative policies applied in the next two decades. There is good reason to doubt that the result reflected an intended policy of responding to money growth. There is not much evidence of deliberate decisions to maintain moderate money growth; Chairman Martin did not favor overt control of money.

Romer and Romer (2002b) recognize that the Federal Reserve in the 1950s succeeded in keeping inflation low. They explain this success as the result of applying a well-developed theory of inflation to which it returned in the 1980s. However, the FOMC minutes and other Federal Reserve documents show no evidence that the Federal Reserve developed or systematically applied any explicit theory or framework in the 1950s. On the contrary, there is considerable evidence showing both a lack of agreement about how monetary policy worked and disinterest in developing a more complete understanding. Later, when Governor Sherman Maisel pushed hard for more systematic procedures, the Committee on the Directive that he chaired recognized explicitly the absence of agreement on basic features of the relation between Federal Reserve actions and goals.
³⁴²

An alternative explanation seems more consistent with what the Martin Federal Reserve said and did in the 1950s and why it did not repeat its actions after 1965. Martin and many others believed that budget deficits caused inflation. During the Truman and Eisenhower administrations budget deficits did not persist. Surpluses in 1956, 1957, and 1960 meant that the Federal Reserve had fewer occasions when it felt obliged to
support Treasury operations. Base money growth remained low, and the Federal Reserve could slow inflation without raising short-term interest rates above 4 percent. The Johnson administration ran persistent budget deficits: many more periods of even keel not only reduced the time available for restricting growth of reserves and money but required additional reserve injections. Also, nominal interest rates rose with inflation, misleading Federal Reserve officials about the degree of restraint. Chapter 4 develops this alternative.

342. This comment and others like it should not suggest that effective monetary policy requires a fully specified model of the economy and its interaction with the world. Rules of thumb such as inflation targeting, a Taylor rule, a monetary base rule, or a Friedman rule may be adequate.

Although FOMC members did not agree on how monetary policy affected economic activity and inflation, the staff at times showed a sophisticated understanding of the role of anticipations. Their discussion of the transmission process recognized the effects of monetary actions on relative prices and capital values. These statements were far ahead of most academic economic analyses at the time.

Chart 2.15 suggests that trends in money and real GNP were often in the same direction and of similar magnitude, despite frequent, large, shortterm changes in monetary velocity. The chart shows that typically money growth was procyclical, rising in periods of economic expansion and falling when output or its growth rate fell. Procyclical monetary policy was a standard feature of policy under a gold standard and for much the same reason: monetary policy restricted interest rate increases during expansions, thereby permitting inflation to increase. In economic contractions, money growth declined. Under the gold standard, the delay reflected the movement of gold. In the 1950s, discretionary actions could have changed the timing of interest rate changes, for example by responding to changes
in money growth. However, the Federal Reserve ignored cyclical changes in money growth.

Several features of the Federal Reserve’s approach contributed to the neglect of money growth when implementing policy action. First, the FOMC and the staff did not have a systematic way of relating the actions they took in the money market to the longer-term objectives that they often embraced in speeches. The committee met every three weeks soon after Martin took control, and it issued a directive to the manager to guide his actions until the next meeting. There are few mentions of lags between policy actions and their effects or setting an objective to be achieved over time. Occasionally someone suggested that some of the observed changes were transitory and would reverse, but the distinction between persistent and transitory changes rarely affected actions at the time or later. Second, neither the manager nor the committee had developed a systematic way of influencing money growth or goal variables such as prices and output. Operations were more a matter of guess or judgment and trial and error. Third, the Treasury came to market frequently. The Federal Reserve adopted even keel policies from two weeks before to two weeks after Treasury sales. During even keel, it did not implement policy changes; it kept interest rates within a narrow band. Often it had to supply reserves to maintain the even keel, so policy operations yielded to Treasury operations with loss of independence. On two occasions, the FOMC bought long-term issues or new issues to support the Treasury’s operation. The reserves they supplied under even keel remained in the market to support a larger money stock. This was a smaller problem in the 1950s than after 1965, when budget deficits rose and even keel became more frequent. Fourth, average money growth remained low after 1954, as Chart 2.15 shows. The Federal Reserve explained the reason for low money growth as an adjustment of cash balances after wartime growth. Although probably correct, the staff did not offer evidence to support this claim. Fifth, money growth and free reserves were not closely related in this as in other periods. Sometimes the two moved in opposite directions as in 1953–54, sometimes in the same direction as in 1958. Control of free reserves did not imply monetary base control. Chart 2.16 shows that from 1954 to 1960, monetary base growth remained low; the compound average growth rate was less than 1 percent a year and, as the chart shows, base growth was rarely above a 2 percent annual rate and not persistently so.
³⁴³
Sixth, meeting every three weeks
focused attention on short-term changes. Members could wait for another meeting to act. Since there was no agreed framework for interpreting data, action or decisions waited for evidence of inflation or recession. Martin or the Board’s staff might predict inflation but Hayes would urge delay.

343. Base growth is based on the Rasche-Anderson (St. Louis) data including adjustment for changes in reserve requirement ratios. Note that base growth declined before each of the recessions.

One of the principal arguments against monetary control was that monetary or base velocity was highly volatile in the short-run. Chart 2.17 shows growth of average cash balance per unit of income, the reciprocal of base velocity growth, at three frequencies. Monthly and annual data show considerable variability. By 1955, the three-year moving average of base velocity growth became more stable although fluctuations continued.

The Federal Reserve under Martin, and Martin himself, wanted to maintain stable prices. In part he believed that low inflation contributed to growth. This view did not gain wide acceptance at the time, but President Eisenhower shared it. By the standards of the 1960s and 1970s, the System did well. Martin often said that the Federal Reserve was the only institution willing to act against inflation. His concern was the budget deficit and the pressure on the Federal Reserve to finance it.

His concern was often overstated. The Eisenhower administration, especially the president and Treasury Secretaries Humphrey and Anderson, also believed that deficit spending was inflationary. The government ran a large deficit in 1959, a result of the 1958 recession, but it closed the deficit quickly at a time when monetary policy tightened, bringing on the
1960–61 recession. And the Eisenhower administration achieved a budget surplus in three of its eight years in office. Almost forty years would pass before the Clinton administration would improve on the budget performance. Inflation rose and later fell in periods with sizeable deficits relative to GDP.

The presence of a large outstanding debt was a major difference between the 1920s and the 1950s. At first, many economists claimed that the debt reduced the role of monetary policy. Any change in interest rates would have a magnified effect on the price of long-term debt and, thus, on the wealth of debt owners. This argument can be found as late as 1959 in modified form in the
Staff
Report
on
Employment,
Growth,
and
Price
Levels,
where the authors argue that “the growth of the public debt, increased efficiency of the Government securities market, and the expanded role of sophisticated financial institutions have reduced . . . the effectiveness of monetary policy” (Joint Economic Committee, 1959a, 345)
³⁴⁴

344. A representative Federal Reserve statement describing the views held in the 1940s and 1950s is in Clarke (1970, 5). Higher interest rates “may have brought about grave disturbances in the market for government securities, with damaging repercussions upon our entire financial mechanism as well as serious economic effects upon public confidence in the Government’s credit.” Total gross debt rose from $270 to $285 billion between 1946 and 1959, but publicly held marketable debt declined from $161 to $148 billion. As a share of GNP, publicly held debt fell from 87 percent to 44 percent. The claim about reduced effectiveness of monetary action was repeated in almost every expansion and contraction for the next forty years. The reason changed. Among the reasons cited have been intermediation, euro-dollars, sophisticated financial instruments, declining relative size of banks, etc.

At the Federal Reserve the opposite argument gained adherents (Roosa, 1951). The existence of a large debt freed the financial system from reliance on the call money market.
³⁴⁵
There was now a market in which the Federal Reserve could affect bank reserves without concern about the size of its operations or not having securities to sell in the event of a gold inflow or a surge in member bank borrowing. Also, the large debt held by the reserve banks removed concerns about reserve bank earnings. All reserve banks shared in the earnings on the open market account and together returned 90 percent of their earnings to the Treasury.

Soon after Martin became chairman, he proposed that, except in the unusual circumstance of a disorderly market, all open market purchases and sales would take place at the short end of the maturity structure. The reasons for the policy included reduced discretion for the manager and the New York bank, but the main purposes were to develop an active private market for long-term debt and end support of a rigid interest rate structure. Martin and Riefler argued that, without bills-only policy, market participants would depend on Federal Reserve intervention to stabilize the market and would expect the Federal Reserve to support new issues if their prices declined. It would be difficult to prevent a return to pegged or controlled long-term rates, with a resulting loss of independence and nullification of the Accord. Their solution was to deal in bills only.

New York opposed the proposal when it first appeared in 1953 and forever after. Most economists who discussed the issue sided with New York, and many members of Congress, including Senator Paul Douglas and Congressman Wright Patman, blamed bills-only for the rise in interest rates during the 1950s.

Martin and the Board’s staff had the better economic arguments, but they could not meet the political arguments, especially because long-term rates increased with economic growth and modest inflation. The Federal Reserve had claimed initially that bills-only would strengthen the long-term market, but they had difficulty producing evidence to support the claim, and they did not offer an alternative explanation of the rise in interest rates. Most critics dismissed the fact that, in eight years, the Federal Reserve had to intervene only twice to prevent disorderly long-term markets.

Bills-only was part of Martin’s effort to reestablish the independence that the Federal Reserve had lost in the 1930s and 1940s. He succeeded
in this aspect during the 1950s. The market for long-term debt absorbed the debt that the Treasury sold without direct support by the Federal Reserve.
³⁴⁶
The political cost was high. Influential members of Congress blamed the bills-only policy for the rise in long-term rates. The 1960 election brought in an administration staffed with many opponents of the policy. The Federal Reserve ended bills-only shortly after that election. Like most central banks, it continued to operate mainly in the short-term market.

345. The call money market post at the New York Stock Exchange closed.

346. Direct support meant Federal Reserve purchases of the new issue or “when issued”
securities. As noted, the Federal Reserve gave indirect support through its even keel policy. Under this policy, the Federal Reserve could, and often did, supply additions to reserves equal to the size of the issue times the average reserve requirement ratio (0.18) times the banks’ share of bond purchases.