Popular in Course
verified elite notetaker
Popular in Business
This 7 page Document was uploaded by an elite notetaker on Monday December 21, 2015. The Document belongs to a course at a university taught by a professor in Fall. Since its upload, it has received 10 views.
Reviews for Modern-Business-Cycle-Analysis--A-Guide-to-the-Prescott-Summers-Debate
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 12/21/15
Federal Reserve Bank of Minneapolis Modern Business Cycle Analysis: A Guide to the Prescott-Summers Debate (p. 3) Rodolfo E. Manuelli Theory Ahead of Business Cycle Measurement (p. 9) Edward C. Prescott Some Skeptical Observations on Real Business Cycle Theory (p. 23) Lawrence H. Summers Response to a Skeptic (p. 28) Edward C. Prescott Federal Reserve Bank of Minneapolis Quarterly Review Vol. 10, NO. 4 ISSN 0271-5287 This publication primarily presents economic research aimed at improving policymaking by the Federal Reserve System and other governmental authorities. Produced in the Research Department. Edited by Preston J. Miller and Kathleen S. Rolte. Graphic design by Phil Swenson and typesetting by Barb Cahlander and Terri Desormey, Graphic Services Department. Address questions to the Research Department, Federal Reserve Bank, Minneapolis, Minnesota 55480 (telephone 612-340-2341). Articles may be reprinted it the source is credited and the Research Department is provided with copies of reprints. The views expressed herein are those of the authors and not necessarily those of the Federal Reserve Bank of Minneapolis or the Federal Reserve System. Federal Reserve Bank of Minneapolis Quarterly Review Fall 1986 Some Skeptical Observations on Real Business Cycle Theory* Lawrence H. Summers Professor of Economics Harvard University and Research Associate National Bureau of Economic Research The increasing ascendancy of real business cycle business cycle theory and to consider its prospects as a theories of various stripes, with their common view that foundation for macroeconomic analysis. Prescott's pa- the economy is best modeled as a floating Walrasian per is brilliant in highlighting the appeal of real business equilibrium, buffetedby productivity shocks, is indica- cycle theories and making clear the assumptions they tive of the depths of the divisions separating academic require. But he does not make much effortat caution in macroeconomists. These theories deny propositions judging the potential of the real business cycle para- thought self-evident by many academic macroecono- digm. He writes that "if the economy did not display the mists and all of those involved in forecasting and con- business cycle phenomena, there would be a puzzle," trolling the economy on a day-to-day basis. They assert characterizes without qualification economic fluctua- that monetary policies have no effect on real activity, tions as "optimal responses to uncertainty in the rate of that fiscal policies influence the economy only through technological change," and offers the policy advice that their incentive effects, and that economic fluctuations "costly effortsat stabilization are likely to be counter- are caused entirely by supply rather than demand productive." shocks. Prescott's interpretation of his title isrevealing of his If these theories are correct, they imply that the mac- commitment to his theory. He does not interpret the roeconomics developed in the wake of the Keynes- phrase theory ahead of measurement to mean that we ian Revolution is well confined to the ashbin of history. lack the data or measurements necessary to test his And they suggest that most of the work of contempo- theory. Rather, he means that measurement techniques rary macroeconomists is worth little more than that of have not yet progressed to the point where they fully those pursuing astrological science. According to the corroborate his theory. Thus, Prescott speaks of the key views espoused by enthusiastic proponents of real busi- deviation of observation from theory as follows: "An nesscycle theories, astrology and Keynesian economics important part of this deviation could very well disap- are in many ways similar: both lack scientific support, pear if the economic variables were measured more in both are premised on the relevance of variables that are conformity with theory. That is why I argue that theory in fact irrelevant, both are built on a superstructure of is now ahead of business cycle measurement " nonoperational and ill-defined concepts, and both are The claims of real business cycle theorists deserve harmless only when they are ineffectual. The appearance of Ed Prescott's stimulating paper, *An earlier version of these remarks was presented at the July 25, 1986, "Theory Ahead of Business Cycle Measurement," meeting of the National Bureau of Economic Research Economic Fluctuations affords an opportunity to assess the current state of real Group. 23 serious assessment, especially given their source and interest rate to be 4 percent. Over the 30-year period he their increasing influence within the economics profes- studies, it in fact averaged only about 1 percent. This list sion. Let me follow Prescott in being blunt. My view is of model parameters chosen somewhat arbitrarily that real business cycle models of the type urged on us could be easily extended. by Prescott have nothing to do with the business cycle A more fundamental problem lies in Prescott's phenomena observed in the United States or other capi- assumption about the intertemporal elasticity of substi- talist economies. Nothing in Prescott's papers or those tution in labor supply. He cites no direct microeco- he references is convincing evidence to the contrary. nomic evidence on this parameter, which is central to Before turning to the argument Prescott presents, let his model of cyclical fluctuations. Nor does he refer to me offer one lesson from the history of science. any aggregate evidence on it. Rather, he relies on a Extremely bad theories can predict remarkably well. rather selective reading of the evidence on the inter- Ptolemaic astronomy guided ships and scheduled har- temporal elasticity of substitution in consumption in vests for two centuries. It provided extremely accurate evaluating the labor supply elasticity. My own reading predictions regarding a host of celestial phenomena. is that essentially all the available evidence suggests And to those who developed it, the idea that the earth only a minimal response of labor to transitory wage was at the center seemed an absolutely natural starting changes. Many studies (including Altonji 1982; place for a theory. So,too, Lamarckian biology, with its Mankiw, Rotemberg, and Summers 1985; and Eichen- emphasis on the inheritance of acquired characteristics, baum, Hansen, and Singleton 1986) suggest that the intertemporal substitution model cannot account at successfully predicted much of what was observed in studies of animals and plants. Many theories can either the micro or the macro level for fluctuations in la- approximately mimic any given set of facts; that one bor supply. theory can does not mean that it is even close to right. Prescott is fond of parameterizing models based on long-run information. Japan has for 30 years enjoyed Prescott's argument takes the form of the construc- tion of an artificial economy which mimics many of the real wage growth at a rate four times the U.S. rate, close properties of actual economies. The close coincidence to 8 percent. His utility function would predict that such of his model economy and the actual economy leads rapid real wage growth would lead to a much lower him to conclude that the model economy is a reasonable level of labor supply by the representative consumer. I if abstract representation of the actual economy. This am not aware that this pattern is observed in the data. claim is bolstered by the argument that the model econo- Nor am I aware of data suggesting that age/hours pro- my is not constructed to fit cyclical factsbut is parame- files are steeper in professions like medicine or law, terized on the basis of microeconomic information and where salaries rise rapidly with age. the economy's long-run properties. Prescott's argument Prescott's growth model is not an inconceivable is unpersuasive at four levels. representation of reality. But to claim that its parame- ters are securely tied down by growth and micro obser- Are the Parameters Right? vations seems to me a gross overstatement. The image First, Prescott's claim to have parameterized the model of a big loose tent flapping in the wind comes to mind. on the basis of well-established microeconomic and long-run information is not sustainable. As one exam- Where Are the Shocks? ple, consider a parameter which Prescott identifies as My second fundamental objection to Prescott's model being important in determining the properties of the is the absence of any independent corroborating evi- model, the share of household time devoted to market dence for the existence of what he calls technological activities. He claims that is one-third. Data on its aver- shocks. This point is obviously crucial since Prescott age value over the last century indicate, as Martin treats technological shocks as the only driving force Eichenbaum, Lars Hansen, and Kenneth Singleton behind cyclical fluctuations. Prescott interprets all (1986) have noted, an average value of one-sixth over movements in measured total factor productivity as the past 30 years. This seems right—a little more than being the result of technology shocks or to a small half the adult population works, and those who work extent measurement error. He provides no discussion of work about a quarter of the time. I am unable to find the source or nature of these shocks, nor does he cite evidence supporting Prescott's one-third figure in the any microeconomic evidence for their importance. I cited book by Gilbert Ghez and Gary Becker (1975). To suspect that the vast majority of what Prescott labels take another example, Prescott takes the average real technology shocks are in fact the observable concomi- 24 Lawrence H. Summers Skeptical Observations tants of labor hoarding and other behavior which blue-collar hours paid for were hoarded. Similar con- Prescott does not allow in his model. clusions have been reached in every other examination Two observations support this judgment. First, it's of microeconomic data on productivity that I am aware hard to find direct evidence of the existence of large of. technological shocks. Consider the oil shocks, certainly In Prescott's model, the central driving force behind the most widely noted and commented on shocks of the cyclical fluctuations istechnological shocks. The propa- postwar period. How much might they have been gation mechanism is intertemporal substitution in em- expected to reduce total factor productivity? In one of ployment. As I have argued so far, there is no inde- the most careful studies of this issue, Ernst Berndt pendent evidence from any source for either of these (1980, p. 85) concludes that "energy price or quantity phenomena. variations since 1973 do not appear to have had a significant direct role in the slowdown of aggregate What About Prices? . . . labor productivity in U.S. manufacturing, 1973-77." My third fundamental objection to Prescott's argument This is not to deny that energy shocks have important is that he does price-free economic analysis. Imagine an effects. But they have not accounted for large move- analyst confronting the market for ketchup. Suppose ments in measured total factor productivity. she or he decided to ignore data on the price of ketchup. Prescott assumes that technological changes are This would considerably increase the analyst's freedom in accounting for fluctuations in the quantity of ketchup irregular, but is unable to suggest any specific techno- logical shocks which presage the downturns that have purchased. Indeed, without looking at the price of ketch- actually taken place. A reasonable challenge to his up, it would be impossible to distinguish supply shocks model is to ask how it accounts for the 1982 recession, from demand shocks. It is difficult to believe that any the most serious downturn of the postwar period. More explanation of fluctuations in ketchup sales that did not generally, it seems to me that the finding that measured confront price data would be taken seriously, at least by productivity frequently declines is difficult to account hard-headed economists. for technologically. What are the sources of technical Yet Prescott offers us an exercise in price-free eco- regress? Between 1973 and 1977, for example, both nomics. While real wages, interest rates, and returns to mining and construction displayed negative rates of capital are central variables in his model, he never looks productivity growth. For smaller sectors of the econo- at any data on them except for his misconstrual of the my, negative productivity growth is commonly ob- average real interest rate over the postwar period. Oth- served. ers have confronted models like Prescott's to data on prices with what I think can fairly be labeled dismal A second observation casting doubt on Prescott's assumed driving force is that while technological results. There is simply no evidence to support any of shocks leading to changes in total factor productivity the price effects predicted by the model. Prescott's work are hard to find,other explanations are easy to support. does not resolve—or even mention—the empirical real- Jon Fay and James Medoff (1985) surveyed some 170 ity emphasized by Robert Barro and Robert King firms on their response to downturns in the demand for (1982) that consumption and leisure move in opposite their output. The questions asked were phrased to make directions over the business cycle with no apparent clear that it was exogenous downturns in their output procyclicality of real wages. It is finessed by ignoring that were being inquired about. Fay and Medoff (1985, wage data. Prescott's own work with Rajnish Mehra p. 653) summarize their results by stating that "the (1985) indicates that the asset pricing implications of evidence indicates that a sizeable portion of the swings models like the one he considers here are decisively in productivity over the business cycle is, in fact, the rejected by nearly 100 years of historical experience. I result of firms' decisions to hold labor in excess of simply do not understand how an economic model can regular production requirements and to hoard labor." be said to have been tested without price data. According to their data, the typical plant in the U.S. I believe that the preceding arguments demonstrate manufacturing sector paid for 8 percent more blue- that real business cycle models of the type surveyed by collar hours than were needed for regular production Prescott do not provide a convincing account of cycli- work during the trough quarter of its most recent down- cal fluctuations. Even if this strong proposition is not turn. After taking account of the amount of other accepted, they suggest that there is room for factors worthwhile work that was completed by blue-collar other than productivity shocks as causal elements in employees during the trough quarter, 4 percent of the cyclical fluctuations. 25 . . . And Exchange Failures? process? Convincing evidence of the types of mecha- A fourth fundamental objection to Prescott's work is nisms that can lead to breakdowns of the exchange that it ignores the fact that partial breakdowns in the mechanism comes from analyses of breakdowns in exchange mechanism are almost surely dominant fac- credit markets. These seem to have played a crucial role tors in cyclical fluctuations. Consider two examples. in each of the postwar recessions. Indeed, while it is Between 1929 and 1933, the gross national product in hard to account for postwar business cycle history by the United States declined 50 percent, as employment pointing to technological shocks, the account offered fell sharply. In Europe today, employment has not risen by, for example, Otto Eckstein and Allen Sinai (1986) since 1970 and unemployment has risen more than of how each of the major recessions was caused by a fivefold in many countries. I submit that it defies credu- credit crunch in an effort to control inflation seems lity to account for movements on this scale by pointing compelling to me. to intertemporal substitution and productivity shocks. All the more given that total factor productivity has Conclusion increased more than twice as rapidly in Europe as in the Even at this late date, economists are much better at United States. analyzing the optimal response of a single economic agent to changing conditions than they are at analyzing If some other force is responsible for the largest the equilibria that will result when diverse agents inter- fluctuations that we observe, it seems quixotic method- ologically to assume that it plays no role at all in other act. This unfortunate truth helps to explain why macro- smaller fluctuations. Whatever mechanisms may have economics has found the task of controlling, predicting, or even explaining economic fluctuations so difficult. had something to do with the depression of the 1930s in Improvement in the track record of macroeconomics the United States or the depression today in Europe presumably have at least some role in recent American will require the development of theories that can ex- cyclical fluctuations. plain why exchange sometimes works well and other times breaks down. Nothing could be more counterpro- What are those mechanisms? We do not yet know. ductive in this regard than a lengthy professional detour But it seems clear that a central aspect of depressions, into the analysis of stochastic Robinson Crusoes. and probably economic fluctuations more generally, is a breakdown of the exchange mechanism. Read any account of life during the Great Depression in the United States. Firms had output they wanted to sell. Workers wanted to exchange their labor for it. But the exchanges did not take place. To say the situation was constrained Pareto optimal given the technological decline that took place between 1929 and 1933 is simply absurd, even though total factor productivity did fall. What happened was a failure of the exchange mechanism. This is something that no model, no matter how elaborate, of a long-lived Robinson Crusoe dealing with his changing world is going to confront. A model that embodies exchange is a minimum prerequisite for a serious theory of economic downturns. The traditional Keynesian approach is to postulate that the exchange mechanism fails because prices are in some sense rigid, so they do not attain market-clearing levels and thereby frustrate exchange. This is far from being a satisfactory story. Most plausible reasons why prices might not change also imply that agents should not continue to act along static demand and supply curves. But it hardly follows that ignoring exchange failures because we do not yet fully understand them is a plausible strategy. Where should one look for failures of the exchange 26 Lawrence H. Summers Skeptical Observations References Altonji, Joseph G. 1982. The intertemporal substitution model of labour market fluctuations:An empirical analysis. Review ofEconomic Studies 49 (Special Issue): 783-824. Barro, Robert J., and King, Robert G. 1982. Time-separable preferences and intertemporal-substitution models of business cycles. Working Paper 888. National Bureau of Economic Research. Berndt, Ernst R. 1980. Energy price increases and the productivity slowdown in United States manufacturing. In The decline in productivity growth, pp. 60-89. Conference Series 22. Boston: Federal Reserve Bank of Boston. Eckstein, Otto, and Sinai, Allen. 1986. The mechanisms of the business cycle in the postwar era. In TheAmerican business cycle' Continuity and change, ed. Robert J. Gordon, pp. 39-105. National Bureau of Economic Research Studies in Business Cycles, vol. 25. Chicago: University of Chicago Press. Eichenbaum, Martin S.;Hansen, Lars P.; and Singleton, Kenneth J. 1986. A time series analysis of representative agent models of consumption and leisure choice under uncertainty. Working Paper 1981. National Bureau of Eco- nomic Research. Fay, Jon A., and Medoff, James L. 1985. Labor and output over the business cy- cle: Some direct evidence. American Economic Review 75 (September): 638-55. Ghez, Gilbert R., and Becker, Gary S. 1975. Theallocation of time and goods over the life cycle. New York: National Bureau of Economic Research. Mankiw, N. Gregory; Rotemberg, Julio J.; and Summers, Lawrence H. 1985. Intertemporal substitution in macroeconomics. Quarterly Journal of Eco- nomics 100 (February): 225-51. Mehra, Rajnish, and Prescott, Edward C. 1985. The equity premium: A puzzle. Journal of Monetary Economics 15 (March): 145-61.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'