A History of the Federal Reserve, Volume 2 (55 page)

“Built-in stabilizers” could be counted upon to increase spending and reduce tax collections when income declined or growth slowed, and conversely. This fiscal response was helpful but insufficient. Tax collections began to rise and the budget deficit declined as soon as the economy began to recover. “For these reasons, the task of economic stabilization cannot be left entirely to built-in stabilizers. Discretionary budget policy, e.g., changes in tax rates or expenditure programs is indispensable—sometimes to reinforce, sometimes to offset, the effects of the stabilizers” (ibid., 71).
¹⁹

Monetary policy had to adapt flexibly to prevailing conditions and to fiscal policy. “There is, in principle, a variety of mixtures of fiscal and monetary policies which can accomplish a given stabilization objective.
²⁰
Choice among them depends upon other objectives and constraints” (ibid., 85). As an example, the Council’s report cited the balance of payments problem as a restriction on short-term interest rates and money growth during the 1960–61 recession. The reasoning was not obvious. The United States had more than $17.5 billion in gold. Gold that flowed out as interest rates fell relative to foreign rates in recession would return when interest rose in the subsequent expansion, if policy avoided inflation.

Coordination of fiscal and monetary actions under administration guidance diminished Federal Reserve independence. In general, the report treated discretionary choice of monetary policy as similar to fiscal policy. Interest rates declined in recession and rose in expansions without policy action. “Being automatic stabilizers, they can only moderate
unfavorable developments; they cannot prevent or reverse them. . . . Discretionary p
olicy is essential, sometimes to reinforce, sometimes to mitigate or overcome,
the monetary consequences of short-run fluctuations” (ibid., 85).
²¹
However, the balance of payments problem, and the administration’s commitment to slow or stop the gold outflow, restricted discretionary action.
²²

19. Heller (1966, 68–69) lists three significant policy changes. “It became more activist and bolder . . . , [policy] has to rely less on the automatic stabilizers and more on discretionary action . . . less on rules and more on men . . . , [and] not only monetary policy but fiscal policy has to be put on constant, rather than intermittent alert.”

20. Mundell (1962) formalized the “assignment problem” that fit well with the new Keynesian idea. Soon the Council assigned fiscal policy to domestic expansion and monetary policy to control of the international payments deficit under fixed exchange rates.

The 1962 report expressed concern about price and wage changes in less competitive industries and found a role for government in moderating the increases by offering guidelines relating wages and prices to changes in productivity (ibid., 185). A major aim was to shift the Phillips curve to get more employment at lower inflation. President Kennedy never endorsed the guidelines explicitly, though he used them to pressure the steel industry to cancel a price increase and at other times. The Council also emphasized the need to improve the functioning of the labor market by increasing the flow of information to shorten the time workers spent unemployed. The report mentioned the need to balance the goals of inflation and unemployment (ibid., 44). It did not mention the Phillips curve by name, but the idea of a short-term tradeoff is present, as is the possibility of shifting the Phillips curve by improving the functioning of labor markets.
²³

21. Heller emphasized many times that he did not neglect discretionary monetary policy. “The economic policy of the 1960s, the ‘new economics’ if you will, assigns an important role to
both
fiscal and monetary policy. Indeed, the appropriate mix of policies has often been the cornerstone of the argument” (Friedman and Heller, 1969, 16; italics in the original.). Here and elsewhere, Heller criticized Friedman’s proposal, calling for constant money growth; but unlike many earlier Keynesian economists, he did not dismiss monetary policy or suggest that the short-term nominal interest rate should be low and constant. Examples of the more extreme view are the survey of monetary theory written for American Economic Association’s
Readings
in
Business
Cycles
(Gordon and Klein, 1965) or the association’s
Survey
of
Contemporary
Economics
(Villard, 1948). These ideas survived into the 1980s in the writings of Kaldor (1982) and Robinson and Wilkinson (1985). The latter wrote (ibid., 90), “The notion that inflation is a monetary phenomenon and that it can be prevented by refusing to allow the quantity of money to increase is to mistake a symptom for a cause.” Gardner Ackley succeeded Heller as chairman of the Council. His textbook, Ackley (1961), is in the older Keynesian tradition, denying any significant influence to money. The Radcliffe Committee in Britain also denied a role to money in inflation (Committee on the Working of the Monetary System, 1959). See Brunner and Meltzer (1993). Friedman challenged Heller’s statement about the role of money in the “new economics.” He claimed that Heller had changed his opinion based on experience. “I reread the reports of the Council of Economic Advisers that were published when he was chairman . . . I do not believe that anybody can read those reports and come out with the conclusion that they say that money matters significantly” (Friedman and Heller, 1969, 49).

22. “Persistent payments deficits and gold losses have made it necessary for the U.S. government to give greater attention to the net financial outcome of its transactions, and those of its citizens, with the rest of the world. . . . [T]he balance must be under control. . . . Action to safeguard the international position of the dollar is . . . essential” (Council of Economic Advisers, 1962, 144–45). This reflected President Kennedy’s concern. His assistant Arthur Schlesinger (1965, 601) reported Kennedy’s statement that the “two things that scared him most were nuclear weapons and the payments deficit.”

23. At one point, the report distinguished between real and nominal interest rates, but it
ignored the distinction when it presented and discussed interest rates in three recent business cycles (Council of Economic Advisers, 1962, 86–87, 90). The president appointed a subcommittee on wage and price policies headed by George Shultz with members from industry and labor. The report did not support the Council’s proposal. It said: “Private parties will not act against their own interests over any extended period of time, nor would it be desirable for them to do so” (Heller papers, wage price guideposts, August 4, 1961, 8).

The Kennedy administration soon put many of these ideas into practice. Instead of general policies that permitted private decision makers to respond, it offered targeted tax reduction for new investment and more rapid depreciation of capital assets and pressured the Federal Reserve to abandon the bills-only policy and buy longer-term securities. After a few years, it proposed general tax reduction to reduce fiscal drag and stimulate growth. Perhaps most important of all here, it accepted the position that low or moderate inflation would permanently increase employment and that inflation would start to increase before the economy reached full employment. To avoid that outcome, government had to intervene in labor and product markets. This framework dominated policy action and was used to support greater concern for unemployment than for inflation until 1979.

There were four main errors or weaknesses in the policy framework. First, it presumed that economists could forecast with enough accuracy to guide active, discretionary policy changes. There was no sound basis for this optimism at the time or later (Meltzer, 1987). Separating growth of nominal income into real growth and inflation proved inaccurate, in large part because of the inaccuracy of initial estimates of output growth and difficulties of estimating the permanent growth rate (Orphanides et al. 2000; Orphanides, 2001).
²⁴
Implementing the type of activist monetary policy that the “new economics” required for success becomes difficult when forecasts and data are inaccurate. Long after the action, data revisions may show that the action was too small, too large, or even unnecessary. Kennedy’s advisers did not address this problem. They acted as if today’s errors or excesses could be corrected later. Although they recognized that Congress might be slow to act, they did not connect their desire to give the president (and themselves) authority to change tax rates to their belief in
activism. The difficulty encountered in raising tax rates to reduce deficits from 1966 to 1968 must have been an unpleasant surprise.

24. Arthur Okun, who served as a member (1964–68) and chairman (1968–69) of the Council of Economic Advisers, disliked using the term “fine tuning” or “the new economics” to describe the Council’s position. “The switches in policy in recent years have not reflected an effort by economists to keep the economy on some precise, narrow course. We were smart enough to know that we weren’t that smart!” (Okun, 1970, 111). Using more technical jargon, James Tobin described the concern about employment as recognition that it takes many Harberger triangles (social loss from inflation) to fill an Okun gap (social loss from unemployment.) See Tobin 1980. This ignores duration. Postwar, cyclical unemployment lasted a few weeks for individuals on average; inflation was far more persistent.

Second, part of the forecaster’s problem is to recognize whether changes are permanent or temporary. Nowhere was this more important than in separating inflation and real growth. The problem arose particularly when productivity or its growth rate changed after 1966 (Chart 3.1), but it was important also in judging whether increases in the price level were onetime changes in level or persistent changes in the rate of change. After several cycles in which the Federal Reserve acted to reduce inflation only until unemployment rose, more of the public became convinced that inflation was permanent and might increase secularly. Once the public perceived increased inflation as a permanent change, long-term interest rates and wage increases did not decline in recessions, reflecting the public’s anticipations.
²⁵

Third, inflation proved unpopular and a political liability. Unemployment was unpopular also. Instead of the stable policy that encouraged investment and growth, policy after 1965 shifted emphasis frequently between these two goals. And since government now accepted responsibility for economic management, advocates of other goals and objectives made their case for assistance. By the late 1960s, the alleged “selective effects” of monetary policy on housing, income distribution, and other concerns made it difficult for the government to pursue, or the public to expect, consistent policies. The result was called “stagflation.” Inflation remained positive, and the unemployment rate drifted upward from the 1960s to the 1970s. On average, inflation and the unemployment rate rose together and later declined together, contrary to the Phillips curve reasoning that the Council advocated. The public recognized that the commitment to reduce unemployment dominated concerns about inflation, so any decline in inflation was treated as temporary.

Fourth, perhaps the most serious flaw in the economic analysis underlying policy was the belief that policymakers could maximize economic welfare by choosing the optimal mix of monetary and
fiscal stimulus or restraint to achieve the optimal combination of inflation and unemployment.
²⁶
Influenced by early versions of the Phillips curve, economists in the Kennedy and Johnson administrations believed they could perma
nently reduce unemployment by choosing to accept higher inflation. Milton Friedman (1968b) pointed out that a permanent or persistent tradeoff of this kind was inconsistent with rational behavior; the long-run Phillips curve must be vertical because inflation is a nominal variable and unemployment is a real variable. Rational behavior required that any influence of nominal variables on real variables last only as long as it takes markets to learn and adjust.
²⁷
Okun (1970, 108) recognized that Friedman’s point had “some relevance to the real world.” But it should not prevent policymakers from choosing how much inflation and unemployment to accept. “The policy maker must face up to the near-term tradeoff and he must be guided by his perception of public attitudes. It is clear that 4 and 5 percent rates of price increase are intolerable to the American public—whether or not they are likely to speed up to 8 and 10 percent” (ibid., 109).
²⁸

25. Many critics accepted the importance of counter-cyclical fiscal policy but claimed that long-term structural reform was the more appropriate action. Brunner (1986, 109–10) makes this case forcefully.

26. “The economics profession and the nation need better analytical criteria by which to balance the costs of a little more upcreep in prices with those of a little more unemployment” (Okun, 1970, 106).

Later research showed that a discretionary policy reacting to current conditions was sub-optimal. The weight on current concerns, and the discretionary response to those concerns, led policymakers to stimulate the economy when unemployment rose and promise to reduce the resulting inflation later. With discretionary policy, keeping the commitment might not be the desirable option if considered only on the basis of prevailing conditions at the time the commitment came due. Kydland and Prescott (1977) showed that optimal policy required commitment to a rule or known procedure. Failure to follow a rule made policy inconsistent with the promise. The public, on the margin, learned about the inconsistency and disregarded the promise. This very general result applied to promises to maintain low inflation and balanced budgets, main elements in the Council’s strategy.

For the Federal Reserve the new approach required closer coordination with other parts of the government followed by increased pressure to allow more inflation with the intent of reducing unemployment. Later it also meant new ideas that put an end to reliance on free reserves as a principal policy indicator and increased attention to a short-term interest rate, the
overnight rate on federal funds.
²⁹
Policy coordination also meant direct pressures from the president and his staff to change or modify policy actions to accommodate these pressures. Martin had worked for a decade in the 1950s to restore the System’s independence. In the next decade, he sometimes displayed his independence, but he also delayed or avoided actions that he believed necessary or took expansive actions to offset predicted deflationary fiscal actions.
³⁰

27. Okun’s book appeared after Friedman’s paper. Okun (1970, 109) accepted Friedman’s logic but denied its general applicability. “We could not escape our current task of searching for a compromise [tradeoff]. The long run is apparently a matter of decades rather than years.” This empirical judgment soon proved wrong and costly. President Nixon announced that he would reduce inflation without increasing unemployment (Hargrove and Morley, 1984, 328). Paul McCracken, chairman of Nixon’s Council of Economic Advisers, told him that he was mistaken, but the message had gone out and he did not retract it (ibid.).

28. Okun was not alone in rejecting the relevance of Friedman’s analysis. Paul Samuelson’s popular textbook said: “Even if this pessimistic view were to have an element of truth [sic], no doubt many would argue that, in an uncertain world it is better to grasp the lower unemployment that can be had at hand than to wait for the lower unemployment that . . . can be found only in some future bush” (Samuelson, 1973, 835).