A History of the Federal Reserve, Volume 2

by Allan H. Meltzer

  Alan Greenspan’s interpretation of wage and profit data to infer a change in the productivity growth rate is a now famous example of data-based judgment improving on and replacing model-based forecasts. Model forecasts at the time called for higher interest rates. Greenspan resisted. Several years passed before the models confirmed the productivity increase.

  Further, the Woodford model is not the first to dismiss the role of money. In the history of economics, money has been dismissed many times, only to return. The early Keynesian era and Chairman Martin’s tenure are examples; the nineteenth-century Banking School is another. The high variability of short-period money growth makes money an unreliable indicator at this frequency. Long ago, the Federal Reserve learned that, using the St. Louis equation, quarterly changes in money did not promptly affect the rate of inflation, but persistent changes did. The European Central Bank ignores short-period changes but pays attention to maintained growth rates. The many charts in the text showing the relation of real base growth to real GDP growth before each recession is evidence of a reliable influence but not a constant lead. Researchers at the Bank of England wrote, “Understanding the role of money in the economy has always been an important issue for policymakers. . . . Monetary data can potentially provide important corroborative or incremental information about the outlook for inflation” (Berry et al., 2007, abstract). Earlier, Nelson (2002) showed that the monetary base growth had a significant effect on inflation in the United Kingdom.

  A common result of research on the term structure of interest rates rejects the theory that explains long-term rates as dependent only on current short rates and expectations.10 Rudebusch, Sack, and Swanson (2007) studied changes in the term premium. They found that the premium changes by amounts too large to ignore (Federal Reserve Bank of San Francisco, 2004).11 Litterman and Scheinkman (1991) concluded that changes in longer-term bond yields reflect three factors: changes in the level of all interest rates, changes in the slope of the yield curve that raise or lower longer-term rates relative to short-term rates, and changes in the curvature of the yield curve. For example, an increase in energy prices that is expected to persist but not increase further raises short-term rates relative to long rates. An increase in anticipated inflation that is expected to persist raises all rates in a parallel shift. To interpret changes in level slope and curvature require more than a single short-term rate.

  10. A simple regression of current and lagged changes in the short-term rate on the change in the long-term rate using monthly data explains about 25 percent of the long-rate change in the period 1951–66 and 45 percent for 1967–81.

  11. Many studies show a significant relation between recessions and the term structure of interest rates. These studies have two deficiencies. They relate two endogenous variables, usually an unreliable procedure. And they seek to explain real output using differences in nominal interest rates.

  Feldstein (2003, 376) concluded that “the models are far too limited. . . . to be an operational guide to policy. The models inevitably give inadequate attention to financial conditions, to changing institutions, to international markets, and to many other things.”

  Lucas (2007, 168) concluded in his usual succinct way that neglect of money was a mistake. “Events since 1999 have not tested the importance of the [ECB’s] second monetary pillar. . . . I am concerned that this encouraging but brief period of success will foster the opinion, already widely held, that the monetary pillar is superfluous and lead monetary policy analysis back to the kind of muddled eclecticisms that brought us the 1970s inflation.”

  This rejects the use of models with only one interest rate and no role for money. It suggests a role for asset markets including credit and money, and it supports the distinction between price level changes and maintained inflation (Brunner and Meltzer 1989, 1993). Svensson (2002–3, 9) accepts part of this view. “Any realistic model of the economy requires more variables than just inflation and the output gap to describe the state of the economy.”

  Mistaken models of the economy contributed to policy mistakes.12 It is an unfortunate fact, nonetheless a fact, that the Federal Reserve produced better results for the economy in the atheoretical 1950s and the eclectic 1980s and 1990s. I believe that part of the problem in the 1960s and 1970s was that there was too much emphasis on quarterly forecasts and too little attention to medium- and long-term policy implications. Economics is not the science that gives reliable quarterly forecasts. Currently, there is no such science. Federal Reserve policy will do better at achieving stable growth and low inflation if it directs more attention to medium- or longerterm results.

  12. A recent costly error based on faulty judgment was the concern with deflation in 2003. An economy with a very large budget deficit and a depreciating currency is extremely unlikely to experience deflation. The Federal Reserve did not accept this argument. Its concern for deflation delayed the increase in the federal funds rate, contributing to the subsequent housing price increase and substantial expansion of housing credit at low introductory rates. Federal Reserve history shows that concerns about a zero bound for a single short-term rate are misplaced. There are many non-zero interest rates and relative prices. Sellon (2003) supports the analyses in Brunner and Meltzer (1968) rejecting the standard liquidity trap and zero bound.

  International monetary policy varied from the classical gold standard at the start in 1913 to fluctuating exchange rates after 1973. Efforts to coordinate international policy in the 1920s under the gold exchange standard reflected the strong public belief in the gold standard at that time. In the 1930s, Britain, France, and the United States attempted limited coordination under the Tripartite Agreement. In the postwar period from 1945 to 1973, most countries adopted the Bretton Woods system. Later, the 1978 and 1985–86 agreements to coordinate broke down. All such efforts failed.

  Two related reasons dominate. First, the public and policy typically give greater weight to avoiding recession and perhaps inflation than to maintaining fixed exchange rates. Second, governments take responsibility for social problems, including but not limited to unemployment, and use budget expenditure or tax changes that produce budget deficits. Third, governments have little interest in adjusting policies for the benefit of other countries. Given these conditions, international coordination or fixed exchange rates are unlikely to return to the United States.

  In nearly a century of experience with financial failures, the Federal Reserve never developed and announced a lender-of-last-resort policy. Sometimes it lets the institution fail; sometimes it lends to keep it solvent. Failure to announce and follow an explicit strategy increases uncertainty and encourages troubled institutions to press for bailouts at taxpayers’ expense. The credit crisis after 2007 is the latest example.

  Many of the faults and errors arise for three principal reasons. First, the FOMC gives excessive weight to current and near-term events over which it has little control. Longer-term consequences that could be prevented or achieved receive little attention. Much of the discussion at FOMC meetings is about whether to change the federal funds rate by 0.25 percentage points. Forecasts that look ahead a year or more have little observable effect on current or future actions. The Federal Reserve has not developed procedures that adjust its actions to achieve its medium- or longer-term objectives.

  Second, economic data are subject to permanent and temporary changes. Oil price changes are examples of permanent or persistent changes that are usually unforeseen. Bank failures, especially failure of a large bank or financial firm, can occur without prior warning. Most changes in productivity growth are of this kind. Wars, major political events, and currency devaluations or revaluations are other examples. Permanent or persistent changes alter wealth and income. Adjustment to persistent changes in the environment can cause temporary changes in consumption investment, employment, and measured rates of price change. Most of these changes are real, not monetary. The Federal Reserve has to learn to make the distinction.

  Thir
d, the Federal Reserve frequently confuses private and public interest. This occurs especially in its response to banking and financial problems. Preventing financial failures protects stockholders often at the expense of taxpayers. The Federal Reserve’s public responsibility requires maintenance of the payments system. Depositors are insured.

  The Federal Reserve makes mistakes. No readers of these volumes should have any doubt about that. Many of the mistaken actions result from errors, but not all. During the Great Inflation, Board members and the FOMC often failed to distinguish between nominal and real values. Some errors reflect uncertainty about the future. Also, the Federal Reserve is the agent of Congress. Too many of its actions respond to actual or perceived congressional pressure. To protect its independence from congressional legislation, it accedes to political pressure.

  Not all chairmen are intimidated. In August 1982 Paul Volcker refused to commit to reduce interest rates if Congress voted to reduce the budget. His decision came despite his efforts to reduce interest rates at the time. This was unusual.

  Offsetting some of the errors are some notable achievements. These include developing the payments system, maintaining a reputation for integrity and the absence of corruption, abetting development of an efficient financial system, managing a complex domestic and international monetary system under changing external conditions, and recognizing that it is lender of last resort to the entire financial system.

  A strong, independent chairman can overcome the emphasis on the short term. The Federal Reserve, and its then chairman, deserve praise and respect for reducing inflation after fifteen years of increasing inflation amid widespread skepticism. This remarkable period in Federal Reserve history brought in the period known as “the great moderation.” The next chairman maintained low inflation and relatively stable growth. Success of this kind increases public confidence and reduces congressional and administration pressures. The country experienced three of the longest periods of growth punctuated with mild recessions.

  TOWARD BETTER RESULTS

  In more than twenty countries, central banks have now (2009) adopted some type of inflation target as the policy goal. Usually the goal is a dual mandate—stable growth and low inflation. The European Central Bank has a legally stated objective to maintain price stability. But it does not ignore unemployment. A principal benefit of a stated objective for inflation is that it directs attention toward a medium-term objective and away from the excessive concentration on current and near-term changes. The Federal Reserve greatly needs to give more attention to the medium-term consequences of its action.

  The earlier chapters show that some FOMC members recognized at times the need to direct more attention to the medium term. Proposals of this kind had no effect on policy action or decisions. Although it has not adopted an explicit inflation target, the FOMC has encouraged the belief that it has a 1 to 2 percent target. There is, as yet, little evidence that the possible target consistently affects FOMC decisions.

  Four reasons bolster the case for directing more attention to the medium term. First, quarterly values of several of the variables FOMC members watch can be described as near random walks. Output or real GDP is one of those variables. Short-term movements are dominated by random movements that cannot be predicted reliably or controlled effectively.

  Second, Federal Reserve quarterly forecasts have smaller errors in the recent more stable years than the errors that I summarized in my presidential address to the Western Economic Association (Meltzer, 1987). Forecast errors are still large. McNees (1995, table 7) reported mean absolute forecast errors by private and FOMC forecasters for the years 1983 to 1994. Table 10.1 shows these errors.13 The table shows that most FOMC forecasts are about as accurate as the average of private forecasts. Inflation forecasts for eighteen months ahead were better than other forecasts.

  One striking conclusion is that short-term forecasts are not substantially more accurate than forecasts one-and-a-half years ahead. Also, forecast errors for near-term real GDP growth of about 1 percent are a large fraction of average growth—2.5 to 3 percent. This suggests that aiming at a longer term would not greatly reduce accuracy. And by concentrating on longer-term results, accuracy might improve.

  Chart 10.2 shows the difference between forecast and reported values of real output from 1971 through 1999.14 There seems little benefit to be gained from continued reliance on short-term forecasts. The Federal Reserve should not ignore current data. It should reduce the weight given to these data using the method explained in Muth (1960) that extracts the persistent component in the data. Muth showed that the weights attached to transitory and permanent changes should reflect the relative variances of the two types of change.

  13. Blue chip forecasts are the median value of a large number of professional forecasters. FOMC forecasts were made semiannually by members of FOMC. The data show the range and central tendency for these forecasts. Many other summaries of forecast errors yield similar results. In 2008, the FOMC members started to publish quarterly forecasts.

  14. SPF forecasts are similar to Board of Governors staff forecasts.

  Third, data revisions are a second source of error. Although data revisions are large at times and capable of misleading policymakers, revisions are less troublesome than forecast errors. Chart 10.3 shows the size of data revisions. In the relatively stable 1990s, data revisions became less troublesome.15 However, the July 2005 revision of national income accounts (not shown on Chart 10.3) raised the 2004 inflation rate from 1.6 to 2.2 percent.

  Fourth, central banks make their judgments under considerable uncertainty about the evolution of the economy, the public’s behavior, financial innovation, and a host of other events, including errors in reported data. One source of uncertainty is the persistence of observed changes and failure to separate persistent and temporary changes. Some of the latter reverse promptly.

  To reduce the influence of temporary or one-time changes the Federal Reserve relies on a measure of inflation that excludes volatile food and energy prices. This moves in the right direction. As noted above, Muth (1960) proposed a better alternative. If the transitory variance is relatively large, the reported change is considered transitory, so the best response is to do nothing. If the permanent variance is relatively large, the change is likely to persist, so policy should respond. In intermediate cases, the proper size of the response depends on the degree to which it appears to be persistent.

  15. Here is an example: The Bureau of Labor Statistics (BLS) reports payroll employment and unemployment on the first Friday of the next month. Employment is close to 140 million. BLS tries to hold its estimate within 1 percent of total employment. It fails frequently. Small changes in the seasonal factor can produce large errors (Furchgott-Roth, 2007). The Federal Reserve pays considerable attention to the announced unemployment rate despite frequent large revisions.

  Adhering to an explicit inflation target can overcome the problem of giving excessive attention to recent events and reports that are frequently revised. Some FOMC members have recognized these problems, but until recently they did not offer an alternative. Flexible inflation targeting that seeks to control inflation and minimize loss of output is an alternative. But it should not be adopted without congressional acceptance of the proposal. Congress could require a target for the unemployment rate as in Humphrey-Hawkins legislation. Congressional authority over money stems from the Constitution. The Federal Reserve is its agent. If principal members of Congress concerned with monetary policy do not accept or acquiesce in the proposal, it is unlikely to succeed or remain.

  Inflation targeting introduces rule-like behavior, the type of behavior Otmar Issing suggested in the epigraph to this chapter. It provides a solution that is neither fully discretionary nor a rigid rule. Under the dual mandate, the public knows that the central bank aims to achieve both a low rate of inflation and sustained growth, so it can develop its plans with less fear of unanticipated disturbances. Of course unforeseen events do occu
r. Financial distress is one of the most important: The Federal Reserve is the lender of last resort. To carry out this function, it departs temporarily from its rule-like stance for monetary policy. Former Chief of Staff Stephen Axilrod concluded after years of experience that the best course for monetary policy is somewhere between discretion and a rule.

  Since the 1970s, the Federal Reserve has improved its operation as lender of last resort. It recognized that this function requires response to the financial system, not just banks. But in its more than ninety years, it has never announced a strategy or rule, and it has not been consistent. It helped First Pennsylvania and Franklin National to survive. It allowed Drexel, Burnham to fail, but it helped in the rescue of Long-Term Capital Management. These and many other examples create uncertainty that contributes to the virulence of the market’s reaction to crises and to pressures for assistance and bailouts. Recent credit problems show much evidence of changing actions and inconsistent responses.

  Significant changes in policymaking occurred in the 1980s and thereafter. The Federal Reserve now distinguishes real and nominal interest rates and exchange rates. It recognizes that policy should not be made one meeting at a time, although it is not always willing to give longer-term objectives greater weight than short-term, possibly transitory or random, events. It gave up regulation Q interest rate ceilings. In 1994 and 2001, it successfully implemented counter-cyclical policy actions. In 1994 it began to announce the current interest rate target and some clues to what it planned for the next meeting.