Regional President Speech

FORECASTING AND POLICYMAKING; SOME LESSONS FROM EXPERIENCE by DAVID P. EASTBURN President, Federal Reserve Bank of Philadelphia NATIONAL ASSOCIATION OF BUSINESS ECONOMISTS Seminar on A New Look at Short-Range Forecasting Techniques Thursday, April 8, 1971 Marriott Motor Hotel Chicago, Illinois Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis FORECASTING AND POLICYMAKING: SOME LESSONS FROM EXPERIENCE It is greatly tempting for me to assume the role of a policy­ maker here today and exhort you all as economists to produce better fore­ casts. Certainly, policymakers realize now more than ever before how dependent they are on good forecasts. And, certainly, policy in recent years would have been much sounder had economists supplied policymakers with better forecasts. But having participated in both kinds of activities, I’m con­ vinced that as much can be gained by considering also what might be accomplished by better policymaking. Not that policymakers should neces­ sarily try to make the forecaster’s life any easier, but better policies can make for better forecasts, which can make for better policies. For better or worse, all of us— forecasters and policymakers— are in the same boat; our futuresrise and fall together. With this in mind, I should like to make a few comments on policymaking over the past decade or so, especially as it relates to monetary policy. The point will be to see if we can find anything in this experience that can help forecasters and policymakers to be of greater use to each other in the future. Three elements of policymaking have been especially important in recent years. These are: 1. The value judgments which policymakers have brought to bear in trading off social costs of unemployment against social costs of inflation. 2. The role which monetary policy has played vis-a-vis other ways of stabilizing the economy, especially fiscal policy. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 2 - 3. The degree of precision— or fine tuning— which policymakers have tried to achieve in their results. I propose to touch quickly on the first two and spend most of my time on the third, inasmuch as that one poses perhaps the greatest challenge for forecasters. Then, in conclusion, I'll try to indicate what actions policymakers might take with respect to each of these three points and some implications of these actions for forecasters. Value judgments Despite their very real worries about inflation, I believe it is fair to say that policymakers in the past decade or so have been in­ creasingly concerned about the social impact, especially on disadvan­ taged groups, of unemployment. This concern has caused them to move more gradually— and probably less effectively— to deal with inflation than otherwise would have been the case. No one can say this value judgment is "right" or "wrong," but it has tended to lead forecasters to some conclusions that have greatly complicated problems for policymakers. For the typical forecaster has interpreted this judgment as imparting an inflationary bias to the economy for the indefinite future. To the extent his principals act on the basis of this forecast, they make it self-fulfilling. And to the extent it is fulfilled, it is reinforced in the next round of forecasts. The policymaker then finds himself facing the very difficult dilemma of permitting this spiral of expectation and actuality to accelerate or of taking extraordinary measures to bring it to a halt. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis Role of monetary policy A second broad lesson from policymaking over the past decade or so is that fiscal policy is still an uncertain and unreliable tool. By this I mean we can't be certain that Government will use fiscal policy in a sufficiently timely and flexible way to be of much help in stabilizing the economy. As a result, monetary policy has been made more difficult and has assumed a larger role than it should have. This development, too, has had ramifications for forecasters. For if the Fed delays taking effective steps in hopes that fiscal action may be forthcoming (as happened in recent years), forecasters may have to calculate the probabilities that such action will not, in fact, be forth­ coming. And if monetary policy bears a larger burden of restraint than it should, forecasters— who, after all, are not so much concerned with what happens to total GNP as the outlook for particular markets— will have to take special pains to calculate likely effects on certain parts of the economy— like financial markets, housing, and state and local governments. "Fine tuning" A third element of policymaking has been the search for greater precision in results. This is perhaps most popularly summed up in the term "fine tuning." More broadly, it reflects the ever-rising standards of performance which the public demands of the economy and of policymakers. In my view, this is an irreversible trend, and much as we might long for the days when margins of error were larger and standards lower, forces are moving us inexorably in the other direction. This search for precision poses a severe problem for both the policymaker and the forecaster. As I look back over the last 15 years or Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 4 - so, I believe the gap between what policymakers require from forecasters to make decisions and what forecasters are able to deliver has expanded rather than diminished. I further believe that increased cognizance of this growing gap has had a major impact on the course of monetary policy over the past year. So, a close look at this gap between what policy­ makers need and what forecasters provide may give us some important clues about the future direction of Federal Reserve policy actions. Consider for a moment the policymaking climate that prevailed in the 1950fs. Then we were content to avoid depressions but were will­ ing to accept recessions as necessary to control inflation. This posi­ tion required little explicit forecasting on the part of the Federal Reserve. The Fed evaluated current data about aggregate demand, employ­ ment, and prices to determine if the economy was operating satisfactorily. Much more direct attention was paid to the actual course of economic events than to the projected path of the economy. If policymakers became convinced that a turning point in the economy had occurred, then they changed policy. In this kind of environment, what did policymakers require of forecasters? Because routine instability was taken for granted, fore­ casters did not really have to prognosticate turning points. It was good enough— for policymakers and forecasters alike— to identify turning points as they occurred or not long thereafter. And a r'eading of the record indicates that forecasters were pretty successful at supplying this kind of information. But as the 50fs wore on and we headed into the 1960Ts, the public began to question routine instability as an acceptable way to oper­ ate an economy. Tolerance, especially for recessions and unemployment, Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 5 - began to diminish rapidly. A recession for the nation as a whole still meant depression levels of unemployment for some groups in society— particularly blacks and teenagers. Hence, the effort to fine tune. And it did look as though the effort was paying off. After two recessions in four years (1957-61), the economy purred along for much of the early 60's. Monetary and fiscal policies seemed to be providing just the right amount of stimulus to spur economic growth and reduce unemploy­ ment, but without igniting inflation. The business cycle, as far as the public was concerned, was about to be consigned to the museum of things past and economists were riding high on this new wave of accomplishment. It is now clear, of course, that the early 1960's was not a true test of whether forecasters could actually bear the new burdens which fine tuning placed upon them. Policymakers really didn't have to worry about out-wobbling the wiggles between 1961-65. The basic problem was one of reducing a very large amount of excess capacity in the economy. But all of this changed in the latter part of the 1960's. The economy was now bumping along the ceiling of prosperity, inflation was a real threat, and unemployment was increasingly unacceptable. The public had come to accept the idea that the economy could be adjusted without the pain and bloodletting of the 50's. Policymakers were under the gun, and so turned to forecasters for the necessary information. What kinds of information did the policymakers need to fine tune successfully? Unlike the 1950's, when they needed basically to recognize only changes in direction of the economy, they now needed, in addition, precise information about the magnitude of change and when the change would take place. Moreover, they needed to know how much counter­ cyclical impact policy changes would have on the economy and precisely when the impact of these changes would occur. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis o Now this is no small order. If we have learned anything about the economy in recent years, we have learned that it is much more compli­ cated than we had thought. Internally, we have found that its inter­ connections are quite loose. Externally, we have found that it isn’t well insulated from unpredictable shocks such as wars and strikes. In addition, we have become much more sensitive to lags. We do not have a push button economy in which more money instantly brings more jobs. In fact, lags between policy change and impact on the world of sales, jobs, and prices have proved variable and unpredictable. Fine tuning required an ability to predict these lags and to be able to fore­ cast far enough ahead to compensate for them. I think it’s safe to say that, despite improved data and more sophisticated techniques, forecasters, so far at least, have not been able to deliver the kind of information policymakers need to fine tune the economy. Conclusions We find ourselves, as a result of the lessons of the past decade or so, in this position: 1. Forecasters interpret value judgments of policymakers as giving the economy an inflationary bias. In doing so, they create further problems for policymakers. 2. The unreliability of fiscal action and over-reliance on monetary policy create problems in gauging the timing and sectoral impacts of changes in policy. 3. Efforts to fine tune, although responsive to pressures in society, pose an extremely difficult challenge to both fore­ casters and policymakers, a challenge which neither is able Digitized for FRASER to meet given the current state of the art. http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 7 - What are the implications of all this for policymaking? The policymaker will have to do something convincing about inflation before the forecaster can remove the inflation factor from his predictions. He will have to achieve better balance between monetary and fiscal policy and move more decisively to deal with structural problems in society if problems of timing and sectoral impacts of policy are to be reduced. And he will have to reconsider how best to meet the public’s demand for top performance without engaging in fine tuning operations that actually may destabilize the economy. These, of course, are all very difficult things to do. For­ tunately, they all move in the same direction and reinforce each other. The only way to persuade forecasters to remove the inflation component from their predictions is for policymakers to demonstrate that value judgments about inflation and unemployment look to the long run as well as the short run. The social costs of unemployment cannot be avoided in the long run by permitting inflation to go unrestrained now. The Federal Reserve has already provided ample liquidity. In my view, the Fed should avoid moving rapidly to stimulate the economy further; inflation is still too deeply rooted. At the same time, policymakers should not give up on fiscal policy. Fiscal policy is still a potent tool and, despite all our dis- illusionments and disappointments, policymakers should keep trying to master it. The period immediately ahead might, in fact, prove an ideal time to restore fiscal policy as a flexible tool. Monetary policy has supplied abundant funds; if the economy lags behind expectations, it may well be appropriate to take fiscal action to reinforce that already taken on the monetary front. Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis - 8 - In the longer run, both monetary and fiscal policy can be made much more effective if steps are begun now to deal with structural problems in our society. Income maintenance and more comprehensive unem­ ployment compensation are needed to protect disadvantaged groups against economic slowdowns which policymakers may have to produce in fighting inflation. Training and education can better prepare those groups for the shocks of economic change. Finally, policymakers will have to figure out how best to meet the objectives of fine tuning without attempting more than they are capable of. It is tempting, of course, to give up on discretionary monetary policy altogether. If fine tuning aggravates rather than re­ duces economic fluctuations, one argument goes, then we ought to forget it and adopt some kind of rule, such as a fixed rate of growth in the money stock. I am not prepared to take this step, but I do believe experience now suggests a more restrained approach to discretionary monetary policy. The kind of extremes that prevailed between 1965-69 is clearly not warranted given the shortfall in forecasting information. Current Fed policy reflects a workable compromise between too much discretionary action and too little. For more than a year now, the Fed has pursued a path of moderation. It has sought to provide enough funds to sustain the recovery without refanning the flames of inflation. The Fed has resisted the temptation of deviating too far from a rate of monetary expansion that is not sustainable over long periods of time; yet, it has not hesitated to deviate from the course for short periods of time when this seems advisable, as, for example, during the Penn Central episode. An overall policy of moderation may not yield precisely Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis 9 the pace of recovery in 171 that some might prefer or that, theoretically, fine tuning could deliver if we had the knowledge to fine tune, but moder- tion will go a long way towards avoiding actions this year that we would regret in T72, ’73, and f74. Where does all this leave the forecaster? If policymakers do a better job, his task will be made easier in some respects. He should be able to make better forecasts. But policymakers will need still better forecasts if they are to meet the rising standards expected of them. The policymaker and the forecaster will be even more dependent on each other in the 70fs than they were in the 60’s. But the relation­ ship should be more comfortable and realistic because of the experience we have both been through. With policymakers more realistic about the kinds of information forecasters can reasonably provide, perhaps we can do a better job of stabilizing the economy in the next decade than we did in the past five years. DPE-4/7/71 (2) Digitized for FRASER http://fraser.stlouisfed.org/ Federal Reserve Bank of St. Louis

David P. Eastburn (1971, April 7). Regional President Speech. Speeches, Federal Reserve. https://whenthefedspeaks.com/doc/regional_speeche_19710408_david_p_eastburn

@misc{wtfs_regional_speeche_19710408_david_p_eastburn,
  author = {David P. Eastburn},
  title = {Regional President Speech},
  year = {1971},
  month = {Apr},
  howpublished = {Speeches, Federal Reserve},
  url = {https://whenthefedspeaks.com/doc/regional_speeche_19710408_david_p_eastburn},
  note = {Retrieved via When the Fed Speaks corpus}
}