The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
For release on delivery 1:00 p.m. E.S.T. February 28, 1989 Issues Facing Monetary Policy in 1989 Address by Manuel H. Johnson Vice Chairman Board of Governors of the Federal Reserve System before Seminar Sponsored by The National Association of Business Economists Washington, D.C. February 28, 1989 Issues Facing Monetary Policy in 1989 It is a pleasure to be here this afternoon to address the National Association of Business Economists. I always enjoy talking with this group about issues relating to monetary policy. One way of addressing this topic would be to discuss the specifics of the Fed's current concerns and goals for policy in 1989. addressed these points However, Chairman Greenspan has at the Humphrey-Hawkins hearings before Congress just last week and I see no need to repeat all the points in his statement. Instead, I would prefer to concentrate on the following question: to what extent is the real performance of the economy relevant for monetary policy? Conceptual Foundations Recently, many analysts have argued that monetary policy should be set to prevent real economic growth from increasing too rapidly in order to avoid higher inflation. According to this view, appropriate growth rates near full employment depend on the growth of capacity or resource constraints. Economic growth in excess of capacity resource growth inevitably leads to inflation. numerical estimates for non-inflationary rates or Sometimes of real economic growth are made explicit; these days 2.5 percent appears to be a consensus figure for such a growth rate. But estimates have been both higher and lower. - 2 - Of course, this approach suggests that differing economic circumstances would lead to alternative policy prescriptions. Should economic growth be sluggish and less than growth, capacity for example, then the logically consistent policy prescription would be a policy of monetary stimulation. This view is certainly appealing. If timely and accurate measurements and projections of capacity growth or aggregate supply could be made and if macro policies could precisely control aggregate demand, this model would be quite valid. In this case, it would be possible for growth in and capacity matched by resource monetary utilization expansion. to be Aggregate accurately demand could always be set equal to aggregate supply; monetary policy would produce inflation. full utilization of resources without Macroeconomic equilibrium would always prevail. It is easy to understand why such a view is so attractive to many economists. Theoretical academic economists can assume both accurate projections of capacity and precise control of aggregate demand by the monetary authority. A variation of this approach has been appealing to certain policymakers. for years by Macroeconomic policy was dominated theories that emphasized demand-side manipulation of the economy whereby changes in aggregate demand readily influenced economic activity or growth. or determined alterations in - 3 - In part, the influence of these demand oriented theories is a legacy of the 1930s and the explanations developed to assess those years. During that period aggregate demand had collapsed leaving the economy with a great deal of excess supply or unused resources. of this idle capacity readily available, unconcerned about explaining the aggregate supply. Rather, bulk of the attention. With all economists were growth of capacity or aggregate demand received the In these circumstances, movements in aggregate demand were associated with similar movements in employment, aggregate production, demand, and economic growth. not aggregate supply Changes in or capacity, therefore, were seen as being closely related to economic growth. Accordingly, some policymakers readily embraced theories prescribing the manipulation of aggregate demand to influence real economic variables. When circumstances changed so that the economy approached full aggregate demand employment, unanticipated were interpreted still increases as temporarily influencing real variables and economic activity. circumstances, economic however, activity led such demand-induced to inflation. in In these increases in Accordingly, a trade-off between higher levels of real economic activity and inflation was seen to exist. Therefore, lower inflation could only be attained from a real economic slowdown. In this sense, it is easy to understand the attachment of some - n - policymakers to views that effectively treat changes in aggregate demand and real economic activity as synonymous. Unfortunately, an exaggerated emphasis on the practical application of these views has led to arguments for the use of real economic variables as policy indicators or even policy targets. Theoretical Concerns This approach directly conflicts with some long standing and fundamental lessons of monetary thought. Since the dawn of economic reasoning a basic message has been that monetary growth cannot permanently or predictably influence real economic variables like wealth or output growth.1 far back century, as the both late David eighteenth Hume and and Henry early As nineteenth Thornton clearly established that monetary expansion could only temporarily influence economic activity. 2 Correspondingly, it was determined that only nominal prices would be permanently influenced by monetary expansions or contractions. another way of stating the widely This is accepted homogeneity postulate or the long-run neutrality of money which stands as a fundamental microeconomic pillar principles of monetary of economics thought. tell us Indeed, that the maximizing of utility and profit by individuals and firms implies that demand curves and supply curves are homogeneous of degree zero in nominal prices; that a movement in all nominal prices has no effect on real supplies or demands. In other words, microeconomic principles of economics tell us 5 - - that, theoretically, inflation has no necessary relation to real economic activity. These observations have a very important policy implication. near-term If monetary expansion or restraint produces a stimulation or contraction of real economic activity but no permanent, long-run influence, then clearly there must exist some intermediate period between the near-term change and the long-term neutrality in which real economic variables, if used to guide monetary policy, would produce highly misleading signals. Measurement Problems But concerns, aside a host from these of practical troubling problems theoretical associated with attempts to implement a real variable strategy are also relevant. For example, timely and accurate measurements of aggregate resource usage or capacity constraints must be readily obtainable. While current measurements of such variables are likely the best available and are useful for many purposes, they do not appear to be ideal for such a strategy. It has become increasingly evident that there are serious problems associated with our current measures of real economic activity as well as economic potential. Recent work by Professor Eisner describes many of these problems related to the national income accounts that I will not repeat difficulties, here. 3 But aside from actual measurement various real economic data are plagued by - 6 - sizeable revisions and seasonal adjustments. And these data often have significant lags associated with them, complicat ing their usefulness for monetary policy. Measures of potential output and utilization have their own set of problems, capacity despite the diligent efforts of talented analysts who compile them. According to one study, for example, when compared to other groups of economic data classified as cyclical indicators, capacity utilization series rank in the lowest group with respect to "statistical adequacy." 4 None of the available statistics measure economic capacity directly; surveys and estimates of potential capacity necessarily are used to construct this series. Furthermore, most capacity utilization numbers are not comprehensive measures for the macroeconomy. they apply only to a portion of including manufacturing and mining. the economy, Instead, usually These sectors, however, may have declined in overall economic importance in recent decades. Service sectors, agriculture, government, and other non-manufacturing sectors are not normally included in these measures. become Yet some of these excluded sectors increasingly important, suggesting that have capacity utilization estimates may pertain to a smaller portion of the economy than was earlier the case. Moreover, capacity numbers often attempt to measure the degree of existing capital utilization but not that of other factors of production. Utilization rates for _ 7 - labor (and human capital), land, or natural resources, for example, are often not included. Yet these other factors of production may are important and sometimes serve as substitutes for capital. Finally, given our more integrated world economy, domestic capacity usage rates have less and less relevance as measures of overall resource constraints. Foreign or world capacity is obviously more pertinent today than in the past. Accordingly, a monetary policy strategy attempting to equate real economic growth with the growth of potential or full capacity is risky at best. Corroboration with the Empirical Evidence In addition to these theoretical concerns and measurement problems, there is another major reason why it is dangerous to generalize that changes in real economic growth lead to changes in prices. Sustained shifts in real economic growth can only occur because of non-monetary, supply-related adjustments, not monetary induced changes in aggregate demand. technology, labor For example, or changes capital in investment, productivity, or entrepreneurial and innovative activity can all promote more or less real economic potential. Similarly, broad changes in trade barriers or other tax distortions can also work to foster permanent shifts in real growth rates since these actions alter factor productivity by allowing more or less specialization. 8 - On the one hand, - long-run increases in aggregate supply are not inflationary. However, because changes in real output can be associated with either supply factors or demand pressures, there is no consistent between economic growth and inflation. and rapid economic growth have relationship Periods of sustained been associated with deflation, inflation, and stable prices. In the U.S., for example, in the period from 1865 to 1879 there were rapid increases in output and falling prices. Similarly, the decade of rapid growth and stable prices.^ inflation actually 1920s experienced To be specific, during the 9-year period from 1921 to 1930, while the real GNP growth was 4.1% fell 1-1%! Furthermore, correlations between real GNP growth and inflation in the U.S. show no consistent relationship between these variables.^ Many countries with records of sustained strong real GNP inflation. growth have low rather than The record of Asia's newly high rates of industrialized countries certainly supports this contention. On the other hand, sustained periods of slow or sluggish economic growth are often not associated with low rates of inflation. Adverse supply-side shocks to the macroeconomy can retard real growth while contributing to price pressures. The experience of the U.S. economy in the 1970s serves as an all-too-familiar example. 9 - - And domestic policy mistakes, such as increases in protectionism, work to higher, tax rates, stifle economic not lower, or regulatory burdens, growth but will prices. Similarly, can all likely promote other distortions imposed on the price system such as minimum wage laws, usury ceilings, rent controls, or other forms of price controls work to adversely affect resource allocation thereby retarding economic growth while at the same time setting the stage for higher prices. Moreover, there can be little doubt that countries enduring long periods of inflation normally do not also experience rapid economic growth. The experience of many Latin American economies is an obvious example. Implications for Monetary Policy In sum, there are problems with the view that monetary policy should target real GNP growth. If real GNP growth advances because of increases in aggregate supply, Federal Reserve attempts to "prevent overheating" would lead to suboptimal employment of resources and policy error. Similarly, if adverse supply shocks or macro- economic policy errors work to stifle macroeconomic growth, then monetary stimulation is not an appropriate policy response by the Federal Reserve. In fact, monitoring or targeting real economic variables can mislead monetary policymakers and sometimes work to promote a destabilizing variety of "fine-tuning." As a consequence, such an approach can prevent monetary - policy from achieving 10 the - one goal it is capable of achieving: namely, the provision of price stability. For this reason, it is critical that monetary policy focus on nominal and not real economic and financial variables as policy indicators or targets. For this reason and the fact that monetary aggregates have recently become less helpful, I have advocated the use of nominal financial auction market prices as appropriate indicators for policy. Use of such indicators (which I have spelled out elsewhere) makes theoretical sense and is certainly consistent with monetary thought. Employing auction market prices as policy indicators also makes practical sense in that many of the above-cited measurement and data problems do not apply to these variables. Moreover, their use is consistent with what we know actually does work. that this approach will receive economists such as yourselves. Thank you. As a consequence, I hope fair consideration by Footnotes See, for example, David Hume, "of Money," and "of Interest" in David Hume, Writings on Economics edited by Eugene Rotwein, Books for Libraries Press, 1955; pp. 33, 37, 48; Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, edited by Edwin Cannon, University of Chicago Press, 1976, Volume I, Book IV, pp. 450-473. David Hume, op. cit, pp. 38, 40; Henry Thornton, An Enquiry into the Nature and Effects of the Paper Credit of Great Britain (1802) edited with an Introduction by F.A. Hayek, Augustus M. Kelley, Fairfield, 1978, p. 238. Robert Eisner, "Extended Accounts for National Income and Product," Journal of Economic Literature, Vol. XXVI (December 1988) pp. 1611-1684. Victor Zarnowitz, "On Functions, Quality, and Timeliness of Economic Information," Reprint No. 250, National Bureau of Economic Research, February 1982. According to Zarnowitz, statistical adequacy consists of "a number of attributes such as the quality of the reporting system, coverage of process and time unit, availability of estimates of sampling and reporting errors, frequency of revisions, length of series, and comparability over time." See, for example, Milton Friedman, "The Supply of Money and Changes in Prices and Output," in The Optimum Quantity of Money and other Essays, by Milton Friedman, Aldine, Chicago, 1969, pp. 183-184. See for example, Thomas F. Cooley and Charles I. Plosser, "Does A Strong Economy Mean Higher Inflation?", Unpublished manuscript, p. 1; Eugene F. Fama, "Inflation, Output, and Money", Journal of Business, vol. 55, no. 2, April 1982; Milton Friedman and Anna J. Schwartz, Monetary Trends in the United States and the United Kingdom, University of Chicago Press, Chicago, 1982, chapter 9, esp. pp. 403-404; Milton Friedman, Inflation and Unemployment, The 1976 Alfred Nobel Memorial Lecture, published by the Institute of Economic Affairs, London, 1977, esp. pp. 18-23.
@FedFRASER