Brad De Long
University of California at Berkeley
and National Bureau of Economic Research
December 19, 1995
In a world organized in accordance with Keynes' specifications, there would be a constant race between the printing press and the business agents of the trade unions, with the problem of unemployment largely solved if the printing press could maintain a constant lead...
Jacob Viner (1936)
I know there's the myth of the autonomous Fed... [short laugh] and when you go up for confirmation some Senator may ask you about your friendship with the President. Appearances are going to be important, so you can call Ehrlichman to get messages to me, and he'll call you.
Richard Nixon to Arthur Burns
(see Ehrlichman (1982))
I think that both of these quotes contain very important truths. For a long time after World War II, the macroeconomic world was organized--very roughly, but largely--in accordance with Keynes's specifications, and the United States macroeconomy did in the 1970s fall into a pattern that can easily be seen as a "race between the printing press and the business agents of the trade unions."
The idea of the "autonomous Federal Reserve" is a myth--not in the sense that when Richard Nixon said "frog" Arthur Burns hopped into the air, but in the sense that the Federal Reserve's freedom of action is constrained by political masters at both ends of Pennsylvania Avenue. Should all factions--the White House, and the Majority and Minority leaderships of the Congress--believe that the Federal Reserve's policies are destructive and desire them to be changed in the same direction, the Federal Reserve's policies will change--or possibly there might be a very different Federal Reserve.
The 1970s were America's only peacetime inflation: the only time when uncertainty about prices made every business decision a speculation on monetary policy. In magnitude, the total increase in the price level as a result of the sustained spurt in peacetime inflation to the five-to-ten percent per year range in the 1970s was as large as the jumps in the price level as a result of the major wars of this century.
At the surface level, the United States had a burst of inflation in the 1970s because no one--until Paul Volcker took office as Chairman of the Federal Reserve--placed a sufficiently high priority on stopping inflation. Other goals took precedence. As a result, policy makers throughout the 1970s were willing to run some risk of non-declining or increasing inflation in order to achieve other goals. After the fact, many believed that they had misjudged the risks: that they would have achieved more of their goals if they had spent more of their political capital and institutional capability trying to control inflation earlier.
At a somewhat deeper level, the United States had a burst of inflation in the 1970s because economic policy makers during the 1960s dealt their successors a very bad hand. Lyndon Johnson, Arthur Okun, and William McChesney Martin left Richard Nixon, Paul McCracken, and Arthur Burns painful dilemmas. And bad luck coupled with bad cards made the lack of success at inflation control in the 1970s worse than anyone had imagined ]ex ante.
At a still deeper level, the United States had a burst of inflation in the 1970s that was not ended until the early 1980s because no one had a mandate to do what was necessary in the 1970s to push inflation below four percent and keep it there. It took the entire decade for the Federal Reserve as an institution to gain the power and freedom of action necessary to control inflation.
And at the deepest level, the truest cause of the inflation of the 1970s was the shadow cast by the Great Depression. The Great Depression made it impossible--for a while--for almost anyone to believe that the business cycle was a fluctuation around rather than a shortfall below some sustainable level of production and employment. It took the 1970s to persuade economists, and policy makers, that "frictional" and "structural" unemployment were far more than one to two percent of the labor force (although we still lack fully satisfactory explanations for why this should be the case). It took the 1970s to convince economists and policy makers that the political costs of even high single-digit inflation were very high.
Once these two lessons of the 1970s had been learned, the center of American political opinion was willing to grant the Federal Reserve the mandate to do whatever was necessary to contain inflation. But it is hard to see how the U.S. government could have pursued an alternative, earlier, policy of sustained disinflation given a milieu where an average unemployment rate of three percent could be termed a "non-perfectionist's goal": policy was bound to sooner or later fall off the tightrope on the side of excess expansion.
And what then? The Federal Reserve gained a mandate to control inflation at the risk of unemployment during the 1970s, as a result of the discontent and anxiety produced by that decade's inflation. But it is difficult to see how the Federal Reserve could have acquired such freedom of action in the absence of an unpleasant object lesson like the inflation of the 1970s.
And the Federal Reserve certainly did not have a mandate to fight inflation by causing short-term unemployment at the end of the 1960s. William McChesney Martin defined himself as the guy who was supposed to take the punchbowl away when the party got going. But little serious punchbowl-snatching took place until Paul Volcker became chairman of the Federal Reserve. Indeed, it is hard to see the Federal Reserve existing in its current form had Arthur Burns attempted the Volcker disinflation in 1970, or even in 1977.
Thus the memory of the Great Depression meant that the U.S. was highly likely to suffer an inflationary episode like the 1970s in the post-World War II period--maybe not as long, and maybe not exactly when it occurred, but nevertheless a similar episode.
All measures of price inflation show three cycles, the first peaking in 1969, the second in 1973-74, and the third in 1980-81 at the highest level of inflation.
Measures of wage inflation suggest a somewhat different picture.
Wage inflation increases sharply in the mid- to late-1960s, from
around 3 percent per year on average to some seven percent per year.
Thereafter it drifts upward, reaching a peak of some nine percent per
year by the beginning of the 1980s. The large cycles in price
inflation are absent from movements in nominal wages: "special
factors" like oil shocks, food price shocks, shortages, and so forth
are hard to detect in the pattern of nominal wage movements.
There were, however, political obstacles. The first of them was Richard Nixon, extremely wary of economic policies that promised to fight inflation by increasing unemployment. He attributed his defeat in the 1960 Presidential election to the unwillingness of Eisenhower and his economic advisors to stimulate production and employment at the risk of triggering increasing inflation.
Thus Herbert Stein (1984) describes how he and his colleagues at the Nixon-era Council of Economic Advisers were "surprised and unhappy" when they learned that President Nixon had authorized Labor Secretary George Shultz to tell the AFL-CIO that the Nixon administration would "control inflation without a rise of unemployment." In retrospect, Stein would think he should have placed more weight on his first meeting with Nixon, in December 1968:
He asked me what I thought would be our main economic problems, and I started, tritely, with inflation. He agreed but immediately warned me that we must not raise unemployment. I didn't at the time realize how deep this feeling was or how serious its implications would be...
How were economic advisors to deal with a situation in which they found the Phelps-Friedman argument--that reducing unemployment would require a period during which inflation would have to be above its natural rate--convincing, and yet in which their political superiors did not authorize such a policy? By minimizing cognitive dissonance:
The inflation rate was about 5 percent at the beginning of 1969. It did not have to be reduced very far. Unemployment was only 3.3 percent. There seemed considerable room for an increase of unemployment without reaching a level that anyone could consider unusually high...
and hoping that parameter values would turn out to be favorable.
Another political obstacle to a policy of disinflation in the early 1970s was that Arthur Burns did not believe that he could use monetary policy to control inflation. In 1959 Arthur Burns had given his presidential address to the American Economic Association--"Progress toward Economic Stability." Toward the end of his speech, Burns spoke of how:
During the postwar recessions the average level of prices in wholesale and consumer markets has declined little or not at all. The advances in prices that customarily occur during periods of business expansion have therefore become cumulative. It is true that in the last few years the federal government has made some progress in dealing with inflation. Nevertheless, wages and prices rose appreciably even during the recent recession, the general public has been speculating on a larger scale in common stocks, long-term interest rates have risen very sharply since mid-1958, and the yield on stocks relative to bonds has become abnormally low. All these appear to be symptoms of a continuation of inflationary expectations or pressures...
Before World War II such inflationary expectations and pressures would have been erased by a severe recession, and by the pressure put on workers' wages and manufacturers' prices by falling aggregate demand. But Burns could see no way in which such pressures could be generated in an environment in which workers and firms rationally expected demand to remain high and recessions to be short.
We know that Arthur Burns placed little weight on being what Nixon called "a team player." He began contradicting administration policy almost from the day he moved into the Chairman's office. As a critic of Kennedy-Johnson policy and as a Counselor to the President in the first year of the Nixon administration, Burns had been opposed to wage-price guideposts. But things looked different from the Federal Reserve: on May 18, 1970, Burns called for Nixon to adopt an "incomes policy" to "shorten the period between suppression of excess demand and restoration of price stability."
Paul McCracken, especially, was irritated because he thought that Burns had "proposed [an incomes policy] without anything in mind but the phrase" (Wells (1994)), but such a proposal is consistent with Burns's vision.If the President who appointed you does not want a deep recession, and if you do not believe that even a deep recession would generate significant downward pressure on prices--for in post-World War II circumstances who would believe ex ante that a recession would be deep or ex post that it would be long?--then you need some kind of incomes policy. That President Nixon is opposed to an incomes policy and is upset with your advocacy of it would be irrelevant, because the alternatives to an incomes policy are things that the President would dislike even more.
But the most important political obstacle to a policy of inflation-reduction in the early 1970s was the complete lack of support anywhere in the White House or in Congress for such a policy. Nixon had no taste for fighting inflation by causing unemployment. Congressional Republicans' calls for more aggressive fights against inflation were muted. And Democrats in Congress, especially, were eager to blame any rise in unemployment as due to Nixon's unwillingness to take "another, less painful, route to price stability [than gradualism and recession], which Mr. Nixon was too ideological to follow." If Nixon and Burns used the price controls program as a substitute for, not a complement to monetary restraint, they had the wholehearted approval of Congressional Democrats. Herbert Stein (1984) cites Walter Heller, testifying before Joint Economic Committee on July 27, 1972 on how Nixon administration policy was too contractionary:
As I say, now that we are again on the [economic] move the voice of overcautious conservatism is raised again at the other [White House] end of Pennsylvania Avenue. Reach for the [monetary] brakes, slash the [fiscal] budget, seek an end to wage-price restraints.
Moreover, there was apparently sufficient reason to believe that "gradualism" and the Nixon price controls program were working. One of the most curious things about the inflation of the 1970s is that in every year inflation was expected to fall. Anyone seeking to be reassured about the future course of inflation had to do nothing more than glance at the consensus of private-sector economic forecasters to be told that the economy was on the right track, and that inflation next year would be lower than it had been this year. Mistakes in judgment made by economists and government policy makers were also shared by private-sector forecasters--and by those who paid to receive their forecasts.
Blinder (1982) argued that double-digit inflation in the 1970s had a single cause: supply shocks that sharply increased the nominal prices of a few categories of goods, principally energy and secondarily food, mortgage rates, and the "bounce-back" of prices upon elimination of the Nixon controls program. Such shocks were arithmetically responsible for
the dramatic acceleration of inflation between 1972 and 1974....The equally dramatic deceleration of inflation between 1974 and 1976....[And] while the rate of inflation.... rose about eight percentage points between 1977 and early 1980, the `baseline'... rate may have risen by as litle as three.
Arithmetic decompositions of the rise in inflation into upward jumps in the prices of special commodities were never convincing to those working in the monetarist tradition. As Milton Friedman asked:
The special conditions that drove up the price of oil and food required purchasers to spend more on them, leaving them less to spend on other items. Did that not force other prices to go down, or to rise less rapidly than otherwise? Why should the average [emphasis in original] level of prices be affected significantly by changes in the price of some things relative to others? (Friedman (1975), cited in Ball and Mankiw (1995))
Ball and Mankiw (1995) have recently argued that the missing link in Blinder's argument can be provided by menu-cost models, and that their indices of the asymmetry of relative price changes are better indices of supply shocks than are the standard direct measures of the supply shocks themselves. Certainly the swings in prices relative to measures of "core" inflation like the average rate of nominal wage growth are substantial, and match the dates of the OPEC price increase announcements and of the acceleration of food price inflation in 1972-1973. The bursts of inflation in 1972-1974 and 1978-1980 are invisible in the track of average nonfarm wage growth: yet it is surprising for a phenomenon that accelerates the rate of growth of average prices in the economy by five percentage points within a year to have next to no effect on the price of labor--unless something like Ball and Mankiw's (1995) menu cost-based explanation is at work.
Perhaps the supply shocks of the 1970s had so little apparent effect on the rate of growth of nominal wages because they were not fully accomodated, but were instead accompanied by serious recessions. Perhaps an alternative world in which the Federal Reserve sought to fully accomodate the increases in nominal spending and avoid a supply-shock recession entirely would have generated significant acceleration in wage increases. But the combination of supply-shock inflation and supply-shock recession, taken together, appears to have had little permanent impact on the nominal wage dynamics of the U.S. economy in either the mid or the late 1970s: before the supply shocks hit wage inflation was slowly trending upward, and after the supply shocks had passed price inflation quickly returned to levels consistent with wage and productivity growth, and wage inflation was slowly trending upward.
Thus it is hard to sustain the argument that the root of the U.S. inflation problem in the 1970s was the interaction of one-shot upward supply shocks with a backward-looking wage-price mechanism. The baseline inflation rate was some five percent per year in the early 1970s before there were any supply shocks; the baseline inflation rate was pushed up by perhaps two percentage points as a result of the collapse in productivity growth; the baseline inflation rate appeared to be eight or nine percent per year by 1980. There is very little to be accounted for by any feedback of supply shock-induced price increases into the wage setting process--unless you hold a strong belief that nominal wage growth would have significantly decelerated in the 1970s in the absence of supply shocks.
The recession of 1974-1975 pushed the political nation in the direction of calling for reductions in unemployment--no matter what the inflation cost. Near the trough of the recession, Hubert Humphrey and Augustus Hawkins sought to require that the government reduce unemployment to 3 percent within four years after passage, that it offer employment to all who wished at the same "prevailing wage" that Davis-Bacon mandated be paid on government construction projects, and (in its House version) that individuals have the right to sue in federal court for their Humphrey-Hawkins jobs if the federal government had not provided them. In early 1976 the National Journal assessed its chances of passage as quite good--though principally as veto bait to create an issue for Democrats to campaign against Gerald Ford, rather than as a desirable policy.
Arthur Burns tried to avoid getting sucked into what he saw as a no-win situation:
Humphrey-Hawkins... continues the old game of setting a target for the unemployment rate. You set one figure. I set another figure. If your figure is low, you are a friend of mankind; if mine is high, I am a servant of Wall Street.... I think that is not a profitable game... (Wells (1994))
The bill that finally passed and was signed in 1977:
In other words, the bill that was signed did nothing at all--save commit the Federal Reserve Chairman to a twice-a-year round of Congressional testimony. Nevertheless the existence of Humphrey-Hawkins, and the concommitant belief that returning the economy had the highest priority had unpleasant consequences. To some degree it was a matter of bad luck: the productivity slowdown had obvious effects on inflation. To some degree it was the result of the lack of interest and focus in the Carter White House on inflation, in spite of efforts by economists like Charles Schultze to warn that inflation was likely to suddenly become a severe surprise problem in 1979 and 1980 unless a strategy for dealing with it was evolved earlier. But to a larger degree it was that--as long as something like Humphrey-Hawkins was the policy of the Democratic Party--neither the White House nor the Congressional Majority Leadership dared worry about inflation.
Inflation did become a surprise severe political problem in 1979. And this generated the only episode in history in which a Council of Economic Advisers Chairman and a Treasury Secretary waged a campaign of leak and innuendo to try to get the Federal Reserve Chairman to tighten monetary policy--and in which the Federal Reserve Chairman answered that monetary policy had nothing to do with the price of wheat or the price of oil. I have found no one willing to say a good word about G. William Miller's tenure as Chairman of the Federal Reserve. He lasted sixteen months, and then replaced Michael Blumenthal as Secretary of the Treasury.
Stuart Eizenstat--President Carter's Assistant for Domestic Policy--always claimed that Miller's departure from the Federal Reserve was an accident:
The President "accepts" the resignation of [Treasury Secretary] Blumenthal. Blumenthal is known as a voice against inflation, and this adds to the confusion. So we were without a Treasury Secretary. So the President makes calls. Reg Jones of General Electric, Irv Shapiro of Du Pont, David Rockefeller of Chase Manhattan--all are asked and turn it down. This becomes a grave situation. The idea surfaces--I'm not sure where--that Bill Miller take the job. Bill takes it. That then creates a hole at the Fed. And that makes the financial markets even more nervous...
Could the Volcker disinflation have been undertaken earlier?
Had Gerald Ford won reelection in 1976 and reappointed Arthur Burns, would we now speak of the Burns disinflation? Or would the same political pressures that had driven Nixon into wage and price controls have driven a second Ford administration into an overestimation of the available room for economic expansion?
My guess is that the Federal Reserve did not as of the mid-1970s have a mandate to do whatever turned out to be necessary to curb inflation.
Indeed, at the end of 1979 the FOMC decided to shift from targeting interest rates to targeting the money supply. Paul Volcker was asked in the subsequent press conference "if the likely result would be slower growth and higher unemployment." And Volcker dodged the question:
Well, you get varying opinions about that.... I don't think it will have important effects in that connection. I would be optimistic about the results of these actions.... I think the best indications that I have now, in an uncertain world, is that this can be accomplished reasonably smoothly...
His hesitancy makes me think that Paul Volcker was not sure at the end of 1979 that he had a mandate to do whatever was necessary to contain inflation.
Professor of Economics J. Bradford De Long, 601