From the point of view of economic stability, three features of the postwar behavior of the American economy have attracted wide comment: (1) the passage of nearly 15 years without a serious depression—an unprecedentedly long period following a major war, (2) the continuation of “minor” business cycles with about the same degree of regularity as before World War II, and (3) the persistent inflationary tendencies evident in the steady climb of prices and wages.
“There has been an enormous expansion of the factual information available for analysis by research workers and for the guidance of policymakers.”The volume of writings on the related problems of economic fluctuations, growth, and inflation has expanded even more rapidly than national output; and economists have been joined by government officials, legislators, businessmen, and labor leaders in trying to understand the forces making for stability and instability in the economy. The recent extensive hearings before the Joint Economic Committee of Congress are merely one manifestation of this concern. Of theories of economic fluctuations and growth there has been no lack. At the same time, there has been an enormous expansion of the factual information available for analysis by research workers and for the guidance of policymakers.
A paradox stands out in the midst of this wealth of information, analysis, and public concern. Unquestionably, we do know much more about how to cope with instability than we did a generation ago; but our ignorance as to the causes of economic instability is still very great, and with respect to some crucial issues we do not know a great deal more than we did on the eve of World War II.
This paradox arises in the following way. Our greater ability to cope with economic instability stems largely from (1) the insights provided by Keynes and by his predecessors and followers, and (2) the tremendous improvement in the factual information available, especially that provided by the national income accounts. But as Hicks has put it, “Keynesian economics, in spite of all that it has done for our understanding of business fluctuations, has beyond all doubt left at least one major thing quite unexplained; and that thing is nothing less than the business cycle itself.”1J. R. Hicks, A Contribution to the Theory of the Trade Cycle (London: Oxford University Press. 1950), 1. See also R. A. Gordon. “Stabilization Policy and the Study of Business Cycles,” American Economic Review, May 1957, especially p. 122. This is because the Keynesian system is essentially static. It has given us a set of tools and has led to the delineation of a branch of economics called “macro-” or “aggregative” economics. From this analysis have emerged some widely accepted conclusions as to the policy measures needed to increase employment or to maintain full employment, but we still do not have a generally accepted body of dynamic analysis which satisfactorily explains why the economy experiences the kinds of fluctuations that have in fact occurred. There is still no general agreement, for example, as to why the Great Depression occurred, why we have avoided a serious depression since World War II, why minor cycles continue to occur, whether there will be another recession next year or the year after, and, if so, how deep and long it might be. The proliferation of business-cycle theories, often in elegant mathematical form, has not lessened this ignorance. Most builders of dynamic economic models disavow any attempt to explain either the kinds of minor cycles that the American economy has regularly experienced or catastrophic declines of the sort that occurred after 1929. If we exclude these and the war periods, we are forced to conclude that much of formal business-cycle theory does not help us very much in explaining most of the economic fluctuations that have occurred in the last few decades.
The tools of aggregative economics have helped economists to identify some of the more crucial variables requiring study, have suggested types of relationships the investigation of which may help our understanding of economic behavior, and have provided the policymaker with some measures of control that are clearly helpful. But the area of ignorance as to the causes of instability is still distressingly large.
Conference held in June 1959
This helps to explain why, in accordance with a suggestion made by the economists who were directors of the Council in 1957, a conference on research on economic instability was held under the Council’s auspices at the University of Michigan on June 17–19, 1959. About 20 economists, all of whom have been concerned with some aspect of the related problems of economic instability and growth, attended.2Present were: Moses Abramovitz, Stanford University; Gardner Ackley, University of Michigan; Sidney Alexander, Massachusetts Institute of Technology; Joseph W. Conard, Swarthmore College; Edward F. Denison, Committee for Economic Development; James S. Duesenberry, Harvard University; Otto Eckstein, Joint Economic Committee, US Congress; Rendigs T. Fels, Vanderbilt University; Irwin Friend, University of Pennsylvania; Gary Fromm, Harvard University; R. A. Gordon, University of California, Berkeley; Bert G. Hickman, Brookings Institution; David W. Lusher, Council of Economic Advisers; Geoffrey H. Moore, National Bureau of Economic Research; Kenneth D. Roose, Oberlin College; Edward S. Shaw, Stanford University; Henry C. Wallich, Council of Economic Advisers; Paul Webbink, Social Science Research Council; Louis Weiner, Board of Governors of the Federal Reserve System. Participants included not only university research workers but also representatives of private research organizations and of various branches of the federal government.
The purpose of the conference was threefold: (1) to take stock of what economists think they now know about the inherent instability of the American economy, with particular reference to the economy’s assumed greater stability since World War II, (2) to suggest lines along which further research is needed, and (3) if thought desirable, to suggest ways in which the Council might help to stimulate research in this field.
James S. Duesenberry had been requested to prepare a working paper which might provide the starting point for the conference discussion. This request led to a monographic study by Duesenberry, Otto Eckstein, and Gary Fromm, which was, for the most part, an econometric analysis of the way in which the American economy behaves in recessions. It will be published in Econometrica. In addition, discussion papers were presented by Bert G. Hickman, Geoffrey H. Moore, and R. A. Gordon on phases of the overall problem not dealt with at length in the study by Duesenberry, Eckstein, and Fromm.
“While the discussion at the conference (as at most conferences) was fairly discursive, it did concentrate on a number of substantive issues which had been set out in an initial agenda.”While the discussion at the conference (as at most conferences) was fairly discursive, it did concentrate on a number of substantive issues which had been set out in an initial agenda. These were: ways in which the cyclical response mechanism is different than it was before the war (including assumed “structural changes,” the effect of the “automatic stabilizers,” the behavior of consumers’ expenditures, etc.); the recurrence of “minor cycles” about as frequently as before the war; how to secure a better understanding of inventory behavior (which seems to be fully as unstable as it was before the war); the determinants of noninventory investment (about which our ignorance is greatest, and which are critical to the question whether we shall always be able to avoid another major depression); other problems particularly calling for further research; and questions of research methodology.
The conference sensitively reflected our present state of knowledge regarding the causes of economic fluctuations and growth and regarding our ability to keep the economy reasonably stable at a high level of employment. The conference, as well as the substantial literature since the Great Depression, suggests that: (1) we know more than we did 25 years ago about the causes of economic instability and particularly about what needs to be done to prevent serious depressions, but (2) there is almost no important topic concerning which our knowledge is yet firmly enough grounded to generate something close to universal agreement. Our ignorance is especially acute in some areas, particularly that concerned with the behavior of private investment. In addition, as the conference brought out, there are a number of research methodologies available to students in this field; there is some disagreement as to the relative usefulness of each; and not enough has yet been done to combine these methodologies so that they most profitably supplement each other.
The conference indicated a large number of topics for future research. It also led to some helpful suggestions for coordinating some of the lines of research now being carried on and for improving the flow of data, chiefly from the federal government, essential for empirical work in this field. In all these respects, the conference participants believed that a Council committee could make useful contributions.
Proposal for a new Council committee
At its last session, therefore, the conference unanimously recommended that the Council be requested to establish and support a committee on economic stability, the functions of which would be more or less as follows:
1. To facilitate coordination of research activity now going on, where such coordination can play a helpful role. As a first step, it was suggested, leading scholars concerned with econometric models should be assisted to compile an inventory of work already done and particularly of studies already made of particular sectors of the economy. It is generally agreed that econometric models require further disaggregation. Progress toward such disaggregation would be greatly facilitated if the results of available partial studies could be brought together, evaluated, and where appropriate fitted into models of the whole economy now being developed. In this way econometric business-cycle research could have much more of a cumulative effect than has been true in the past, when each investigator has started largely from scratch. It might also be possible to secure agreement on the main features which need to be built into these econometric models. Probably one of the first assignments of a committee on economic stability would be to investigate the possibility of a small working conference of those doing econometric research in this area.
2. To help integrate current research methodologies. More specifically, there was a strong opinion at the conference that historical studies and statistical work of the sort done by the National Bureau of Economic Research can usefully supplement econometric dynamic models (and vice versa). The proposed committee could encourage research that would appraise the results of econometric model-building in the light of quantitative and nonquantitative historical evidence that cannot be fitted into a set of equations. In this way new light might be thrown on those aspects of past cyclical behavior that result from recognizable and continuing dynamic properties of the economic system, and it might be possible to get a better idea as to the extent to which past behavior is attributable to combinations of circumstances which may be unique to particular historical periods.
3. To facilitate the collection and publication of needed data, particularly by the federal government. Scholars in this field need some way of keeping informed regarding the data-collection plans and activities of government and private agencies. While the proposed committee could not devote a great deal of time to this effort, the existence of the committee might well lead data-collecting agencies to seek its advice, and individual scholars could seek its help in inducing particular agencies to compile needed new data or to continue publication of particular statistical series whose discontinuance has been proposed.
4. To serve as a channel of communication and a facilitating agency in the field of research on problems of economic instability. Various types of activity could contribute to this end:
- The committee could try to induce individuals and research groups to undertake research projects that meetings of the committee or special conferences indicated were desirable. This might include providing lists of suggested dissertation topics which could be widely distributed.
- The committee could help to stimulate existing research agencies to undertake lines of investigation that are needed, and it might be called on by such agencies for advice on research plans.
- Conferences might be organized on special topics when it appeared that progress in a particular area could be furthered by such meetings.
- The committee might help to stimulate comparative studies covering a number of countries. While economic fluctuations are assumed to be characteristic of all private-enterprise economies, there are relatively few studies that compare the kinds and causes of fluctuations in different countries. It would be desirable in particular to arrange for a set of empirical studies that would follow a common analytical framework and be addressed to a common set of questions for various countries.
The recommendation made by the conference was approved by the Council’s Committee on Problems and Policy at its September meeting. The initial members of the new Committee on Economic Stability are R. A. Gordon, University of California, Berkeley (chairman); James S. Duesenberry, Harvard University; Bert G. Hickman, Brookings Institution; Lawrence R. Klein, University of Pennsylvania; David W. Lusher, Council of Economic Advisers; and Geoffrey H. Moore, National Bureau of Economic Research. It is expected that the first meeting of the committee will be held in Washington in December, in connection with the annual meeting of the American Economic Association.
The author, professor of economics at the University of California, Berkeley, has been a member of the board of directors of the Social Science Research Council since 1955. He presented the substance of this report informally at the annual meeting of the board in September 1959.
R. A. Gordon (1908–1978) was an American economist and he taught in the Department of Economics at University of California, Berkeley, from 1938 until his retirement in 1976. Two of his most important works are Business Leadership in the Large Corporation (1945) and Business Fluctuations (1952), the latter of which addressed business cycles. He served as chairman of President John F. Kennedy’s Committee to Appraise Employment and Unemployment Statistics. Gordon served as chairman of the SSRC’s Committee on Economic Stability from 1959 until 1978, and he also served on the Council’s board of directors (1955–1960).
This essay originally appeared in Items Vol. 15, No. 4 in December 1959. Visit our archives to view the original as it first appeared in the print editions of Items.