sign in sign up
<<
Home , Economic expansion, Economic policy, Monetary policy

NOVEMBER/DECEMBER1998

John B. Taylor is a professor of at Stanford University. The article that follows is a reprint of The Homer Jones Lecture delivered at Southern Illinois University-Edwardsville on April 16, 1998. Kent Koch provided research assistance.

this lecture. This month (April 1998) the Monetary United States celebrates seven years of . By definition and The Long an economic expansion is the period between ; that is, a period of con- Boom tinued growth without a . The last recession in the United States ended in April 1991, so as of this April we have had seven John B. Taylor years of expansion and we are still going. This current expansion is a record breaker: regret that I never had the opportunity to to be exact it is the second longest peacetime work or study with Homer Jones. But I expansion in American history. Iknow people who worked and studied with But what is more unusual is that this him, and I have enjoyed talking with them and current expansion was preceded by the reading about their recollections of Homer first longest peacetime expansion in Amer- Jones. What is most striking to me, of all that ican history. That expansion began in has been said and written about Homer Jones, November 1982 and continued through is his incessant striving to learn more about August 1990. It lasted seven years and economics and his use of rigorous economic eight months. Although the 1980s expansion research to improve the practical operation of was the first longest peacetime expansion in . As a college student at American history, the current expansion Rutgers, studied under may very well continue long enough to Homer Jones. Friedman Jones as an become the first longest peacetime expan- essential influence on his own decision to sion. Either way, we are now experiencing study economics, and I want to begin this lec- back-to-back the first and second longest ture with a quote from Friedman (1976, p. peacetime expansions in American history. 436) describing certain features of Homer There is something even more extraordi- Jones character: nary. The recession that occurred between these two record–breaking expansions was— The hallmark of his contribution at least for the economy as a whole—short- is throughout those same traits that lived and relatively mild. Hence, during the exerted so great an influence on me last 15 years not only did we have the two in my teens: complete intellectual longest peacetime expansions in American honesty; insistence on rigor of history, but the sole recession we had during analysis; concern with facts; a drive these 15 years was remarkably short and for practical relevance;and, finally, a mild. This 15-year period of unprecedented perpetual questioning and reexami- stability and virtually uninterrupted growth nation of conventional wisdom. is what I refer to as the The Long Boom. The Long Boom has another character- I am going to return to these character istic which I briefly mention now and traits of Homer Jones later in this lecture discuss in more detail later, for it, too, is a for they are part of the story I want to tell. part of the story I want to tell. The inflation rate has been very low and very stable during The Long Boom, much lower and DEFINING “THE much more stable than during the years LONG BOOM” immediately prior to The Long Boom. I must begin by explaining what I mean Consider this long boom period in by the term “The Long Boom” in the title of comparison with other periods in American

FEDERALRESERVEBANKOFST. LOUIS

3 NOVEMBER/DECEMBER1998

Figure 1 States from 1955 to the present. You can see that the economy is growing. You can A More Stable Economy also see the ups and downs: the 1981-82 Percent Billions of 1992 Dollars recession and recovery; and the 1990-91 8000 recession and recovery. The trend line in Tre nd GDP Figure 1 shows where the economy is (right scale) 6000 going in the longer term. The lower part of Figure 1 nicely illustrates the large 4000 4 Real GDP change in . It’s like a (right scale) 2000 microscope that focuses on the 2 fluctuations in real GDP around trend 0 0 GDP. It shows the GDP gap, which is defined as the percentage difference of real -2 GDP from trend GDP. On the right is Percent GDP Gap where we have been recently. On the left -4 (left scale) is where we were before. You can see -6 clearly in Figure 1 that the ups and 1955 1960 1965 1970 1975 1980 1985 1990 1995 downs— recession, expansion, recession, Since the early 1980s, real GDP has fluctuated much less than in earlier expansion—are much milder and much years—as shown by the movement of the GDP gap. less volatile in the latter period than in the earlier period. During The Long Boom, there is obviously greater stability. This history. First, go back to the 15-year greater stability is one reason why the period before The Long Boom—the late has boomed during the same 1960s and 1970s. In the same span of period, 1982 to the present. time we had four recessions, not one. The economy was much more unstable with many ups and downs. We also had the EXPLAINING THE LONG longest inflation in American history—a BOOM very unstable and high inflation—a What are the underlying reasons for remarkably different experience compared this remarkable and unprecedented period to the last 15 years. The economy was not of economic performance in America? performing well. Many explanations have been offered. For another historical comparison, go Some have to do with inevitable changes back exactly 100 years, to the . in the structure of the economy that have Compare the 1890s with the 1990s. If the had the fortunate by-product of a more 1990s were like the 1890s, we would have stable economy. Other explanations are had a recession in 1990 or 1991 that was related to economic policy—deliberate bigger, quite a bit bigger, than the one we decisions of economic policymakers to actually had. And that would not have change policy. So which is it? Good been the end of it. We would have had fortune or good economic policy? Let us another recession in 1993, a big one, right consider each in turn. when President Clinton was taking office. In 1996 we would have had yet another Good Fortune? recession. We would have just been Many have noted that the U.S. economy coming out of that recession now. So you is much more -oriented now than it can see how dramatically different the was in the past. Services—educational ser- economy has been during The Long Boom. vices, legal services, financial services—are Times have changed. generally not as cyclical as manufactured Figure 1 helps visualize this. The , such as automobiles or airplanes. upper part of Figure 1 shows real gross Recessions hit manufacturing sectors quite domestic product (GDP) in the United hard. But in the service industries, busi-

FEDERALRESERVEBANKOFST. LOUIS

4 N OVEMBER/DECEMBER 1998 ness–cycle fluctuations have been typically Long Boom. There are, of course, two small. So maybe The Long Boom, with its aspects of economic policy one should greater stability, is due to the economy focus on: fiscal policy and . becoming more service-oriented. The Let us consider fiscal policy first. problem with this explanation is that the move to a service-oriented economy has been a very gradual change occurring over many decades. It could not explain the Has fiscal policy seen a major change sudden shift toward greater economic sta- that could have led to The Long Boom? bility in the early 1980s shown in Figure 1. What about budget deficits? Hence, a more service-oriented economy is Budget deficits were huge throughout the unlikely to be an explanation. 1980s and much of the 1990s. Budget deficits Others have noted better control of were smaller in the late 1960s and 1970s. inventories. It is true that inventory sales Therefore a smaller budget deficit does not ratios are lower now because inventories seem plausible as an explanation for The are being managed better. The just in time Long Boom. approach to inventory management is now What about the ability of government much more common. During most ups fiscal policy to respond to recessions by and downs in the economy, inventories lowering or increasing spending? fluctuate widely. As the economy starts to Has that response gotten larger, quicker, or dip, firms want to cut their inventories; more efficient? No. In fact, if anything, they reduce their orders, and production the ability of the federal government to falls even more rapidly. Thus, better con- make discretionary fiscal policy changes to trol of inventories may be an explanation mitigate or offset recessions has diminished. for this greater stability that defines The President Bush proposed a small economic Long Boom. But this explanation also has package to be put in place at the problems. If you take out the fluctuations end of the 1990-91 recession, but Congress of inventories and look at what is left rejected it. In 1993 President Clinton also over—final sales—you see virtually the proposed an economic stimulus package same amount of improvement in economic and again the Congress rejected it. Hence, stability. That is, if I replaced real GDP I have to rule out fiscal policy—either with final sales of in smaller budget deficits or better counter- the economy in Figure 1, it would look cyclical policy—as a possible explanation essentially the same. of The Long Boom. A very common explanation of The Long Boom is that the U.S. economy has been lucky with respect to the shocks hit- Monetary Policy ting the economy. Recall that in the 1970s Now consider monetary policy. As is we had several large oil shocks. In the probably already obvious, I will argue that 1980s and 1990s we seem to have had monetary policy is the key factor behind fewer oil shocks. But for two reasons I The Long Boom. To do so, I must first dis- must reject this explanation, too. First, cuss briefly the monetary transmission the poor economic performance, along process in the United States. Monetary with the higher inflation, that we experienced policy is, of course, the responsibility of in the 1970s really got started before the the policymakers who serve on the Federal oil shocks began. Second, the U.S. economy Open Market Committee (FOMC). The had serious shocks in the 1980s and ultimate tool of monetary policymakers is 1990s, including the and loan the supply. An increase in money crisis, the oil when Iraq invaded supply growth will lead eventually to an Kuwait, and the East Asian crisis. increase in inflation. However, the FOMC Now let us move on to economic actually carries out its deci- policy as a possible explanation of The sions by focusing on the short-term

FEDERAL RESERVE OF ST. LOUIS

5 NOVEMBER/DECEMBER1998

Figure 2 look at a number of factors, including the inflation rate and real GDP. Is it possible A Change in to see a change in this decision-making Monetary Policy process that could explain The Long Boom? Has anything important changed Rate about monetary policy? To explore these questions, consider a numerical example 5 that illustrates how the FOMC might respond to a change in the economy, such as an increase in the inflation rate. First, 4 imagine that the FOMCgets reports that the inflation rate is 1 percentage point higher. 3 Let us suppose that in this circumstance the FOMC decides to raise the rate by .75 percentage points. The inflation rate 2 is up by 1 percentage point, and the is up by three-quarters of a percentage 1 point. The FOMC has raised the interest rate in response to inflation. But what hap- penedto the difference between the federal 0 funds rate and the inflation rate, a measure 0 1 2 3 4 5 of the ? According to this measure, the real interest rate has gone down Rate by a quarter of a percentage point. In other The response of the to changes in inflation is words, the federal funds rate did not go up more aggressive (solid line) in The Long Boom than in the prior by enough to raise the real interest rate. It 15 years (dashed line). The dotted line has a slope of 1 and is the real interest rate that affects spending. So by allowing the real interest rate to fall, represents a constant real interest rate. the FOMC would be doing exactly the opposite of what it should do when the inflation rate rises. The FOMC members interest rate—the federal funds rate—that would be to stimulate the economy commercial charge when they loan just when they should be trying to cool off funds to each other. The federal funds rate an inflationary surge. So this policy, with a is the instrument of policy that the FOMC response of .75 percentage points, is not a members vote on. The federal funds rate, good policy. It adds fuel to the inflation fire. of course, has a big impact on all other Now consider an alternative monetary interest rates, especially other short-term policy response. Again, start with the interest rates, but also on longer–term same scenario: The inflation rate rises by 1 interest rates. Interest rates have a big percentage point. But now suppose the effect on the economy. Higher interest FOMC, instead of raising the federal funds rates tend to slow down the economy; rate by .75 percentage points, raises it by lower interest rates tend to stimulate the 1.5 percentage points. In other words, as economy. To be sure, these changes in the inflation rate goes up by 1 percentage interest rates are closely related to the point, the interest rate goes up by 1.5 per- money supply, and if the FOMC tried to centage points. The real interest rate now keep interest rates too low for too long, the goes up by half a percentage point, and that money supply would have to increase and is the right thing for the FOMC to do. The this would cause inflation to rise. FOMC has removed fuel from the inflation When the members of the FOMC fire because the higher real interest rate is make decisions about the interest rate— going to reduce demand for automobiles, whether back in the 1970s or today—they houses, and other goods. Even though in

FEDERALRESERVEBANKOFST. LOUIS

6 N OVEMBER/DECEMBER 1998 both cases the FOMC raised the interest sound like an overly simple description of rate when inflation rose, in one case the how the Fed makes its decisions, it is actu- policy was right, and in the other case the ally very accurate. Empirical estimates of policy was wrong. the Fed’s reaction function (regressions of In fact, these two cases are not the interest rate on inflation and other vari- hypothetical examples. They turn out to ables) show that the dashed line be actual descriptions of monetary policy corresponds to the late 1960s and 1970s before and after the start of The Long period (the great inflation and all those Boom, respectively. The first case charac- business cycles) and the solid line refers to terizes monetary policy during the late the more peaceful economic times associ- 1960s and 1970s period, while the second ated with The Long Boom. case characterizes monetary policy during What are the implications of this assess- The Long Boom of the 1980s and 1990s. ment of policy for the future? Put as simply By responding in this more aggressive as possible, the Fed should continue to way during The Long Boom, the Federal respond to inflation according to a 1.5 Reserve (the Fed) has been able to keep the response coefficient. If it continues doing inflation rate lower and much more stable that, it will be able to keep the economy than in the earlier periods. In my view, that stable, making future long booms more change in policy has been the key to common, not avoiding recessions completely, keeping the real economy stable. Every of course, but making recessions smaller and recession in the post-World War II less frequent. Focusing on keeping the of the United States has inflation rate low and stable and responding been preceded by a run-up of inflation. So aggressively with interest rates is the most by keeping the inflation rate low and stable important thing the Fed can do to keep the through this policy (taking the fuel off the economy stable. fire when inflation heats up) the Fed has succeeded in stabilizing the economy, and making recessions less frequent, smaller, and WHY THE CHANGE IN shorter. That has made all the difference, MONETARY POLICY? which Figure 2 illustrates. The graph has Now, what has caused this shift in the interest rate—in particular, the federal monetary policy? Did economic research funds rate, the variable that the FOMC is play a role? To answer these questions we making decisions about—on the vertical must look at some of the history of the axis. On the horizontal axis is the inflation . Go back to the 1950s. rate. The dashed line is the bad policy, where The 1951 Accord between the Federal the coefficient is .75. For this policy, when Reserve and the released the Fed the inflation rate rises by 1 percentage point, from the job of assisting the Treasury bor- the FOMCraises the interest rate by only .75 rowing by keeping interest rates low, as it percentagepoints.For the solid line, when had done during World War II. But after the inflation rate rises by 1 percentage the Accord, the Fed actually had to decide point, the FOMC raises the interest rate by what to do with the interest rate. One 1.5 percentage points. For reference,I also widely discussed suggestion was to lean show a dotted line in Figure 2 for which against the wind, raising the interest rate the response is exactly one. The real when the economy grew more rapidly or interest rate is constant along that line. By inflation started to pick up. Leaning being more responsive than that dotted against the wind seemed to have the direc- line, the Fed lets the real interest rate rise tions of interest rate adjustments right, but when inflation rises. By being less respon- it had nothing to say about the size of the sive (following the dashed line), the Fed adjustments. What is the wind? How do lets the real interest rate fall when inflation you measure it? What do you lean with rises. These two lines characterize the and by how much? There were many decisions of the Fed. Although this may important, but unanswered, questions.

FEDERAL RESERVE BANK OF ST. LOUIS

7 N OVEMBER/DECEMBER 1998

Meigs (1976, p. 440) described the situ- Homer Jones’ character traits (quoted from ation in the 1950s as follows,“...The Manager Milton Friedman at the start of this lecture) [of the open market account] generally tried were for stimulating and motivating others to to keep free or borrowed reserves...at a carry out this important research agenda. level, he thought would satisfy FOMC mem- In my view, the research directed by bers’ desires for a little more, or a little less, Homer Jones was an essential part of a gradual or about the same ‘degree of restraint.’” But process through which the Fed learned more the degree of restraint was not quantitatively about the conduct of monetarypolicy. Of defined and the impact of changes in the course, others participated in this process— degree of restraint was uncertain. As Meigs the staff at the Federal Reserve Board and (1976) put it, “We were as uncertain about other District banks as well as academic how monetary policy worked as were our . In my view, the result of this colleagues at the other Reserve Banks and gradual learning process was a recognition the Board.” A similar accounting comes by the 1980s that changes in interest rates from Board member Sherman Maisel (1973, had to be larger and quicker if they were to p. 77), who admitted “that after being on keep inflation and the overall economystable. the Board for eight months and attending Focusing on the monetary aggregates— twelve open market meetings, I began to especially during the disinflation period of realize how far I was from understanding the late 1970s and early 1980s when interest the theory the Fed used to make monetary rates had to rise by a very large amount— policy... Nowhere did I find an account of and emphasizing the distinction between the how monetary policy was made or how it real and nominalinterest rate were part of operated.” the means towards this end. Of course, there This was the situation when Homer were other factors that led to the change in Jones arrived at the research department of monetary policy. The Fed learned from the the of St. Louis in great inflation experienceof the 1970s and 1958. He and others were uncomfortable discovered through increasing evidence that with this vagueness and about there was no long-run tradeoff. the operations of monetary policy, and he tried to make the FOMC decisions more specific. The vagueness was why, according to Meigs (1976), Homer Jones “undertook A PROBLEM WITH THE the extraordinary program of monetary research LONG BOOM to which all of us are indebted today.” So far in this lecture I have emphasized The research program undertaken by the many good features of The Long Boom Homer Jones helped change this situation and the role of economic policy in helping in several ways. First, the research improved to bring them about. Now let me move on the money supply statistics. This allowed to some not-so-good features and the role policymakers to see how money supply growth that economic policy might have in allevi- targeting would work and to measure the move- ating them. These are what we mentin interest rates that would accompany should focus on in the future. money growth targets. The St. Louis Fed The main problem with The Long Boom Model of the U.S. economy provided an is that growth is much lower analytical structure through which different than during past periods of U.S. history. By monetary policy procedures could be explored. definition, productivity is the amount that The research was also essential in helping poli- workers produce on average during a given cymakersdistinguish between real interest time at work. Labor productivity growth rates and nominal interest rates. The neglect means that for the same number of hours of this distinction is at the heart of the poor of work, a worker can produce more. Pro- monetary policy performance in the late 1960s ductivity growth is the means through and 1970s when the interest rate reaction was which people improve their living standards, too small. One can easily imagine how useful because more production per worker means

FEDERAL RESERVE BANK OF ST. LOUIS

8 N OVEMBER/DECEMBER 1998 that workers can earn more. Productivity A GOAL growth is why the standard of living is I propose that we adopt a goal. The now so much higher than it was in the goal is much easier to write down than to days before the . carry out. I write in bold-faced characters: Productivity growth in the 1950s and +1%. The goal is to raise productivity 1960s was about 2 percent per year. During growth by 1 percentage point per year, so the period of The Long Boom, productivity that productivity growth averages 2 percent has been about 1 percent per year. Steady 2 per year for the 21st century. In other percent per year productivity growth means words, we would see 2 percent per year that the average worker in America can pro- productivity growth rather than 1 percent duce 2 percent more next year compared to per year productivity growth for the next this year for the same amount of time on the 100 years. job, and 2 percent more the next year and Now, I will be the first to admit that so on. The 2 percent accumulates and this is a goal that is very difficult to achieve. compounds. Unfortunately, the data indicate But I think goals are useful for bringing atten- that productivity growth is only half as much tion to a problem, for focusing policymakers as it was in the 1950s and 1960s. efforts, and simply for getting things done. Now, to be sure, there are some signs Even if we got halfway to that +1% goal, it that productivity growth has increased would make a tremendous difference. The recently. In fact, a recent buzzword in the U.S. economy reached that 2 percent goal financial press is the “New Economy.” The in the past, so it may indeed be achievable. New Economy is characterized by higher What can we do to help achieve the productivity growth. Others argue that +1% goal? By way of offering some there is a problem in measuring productivity, examples, I would like to mention four especially in a computer age, and that pro- economic policy measures that—taken ductivity is actually higher than we think. together—would achieve the goal. But Still others have argued that an even bigger the important point is that setting such a productivity spurt is about to happen: We goal would improve the debate about have all this great technology—computers, policy alternatives that have the best biotech, and telecommunications—ready chance of achieving the goal. to be used in the workplace to make people Let us consider budget policy first. more productive. But, in my view, it is too The debate about the federal budget has soon to conclude that we are now in a changed remarkably in the last two years, period of persistently higher productivity from debate about how to end the deficit growth. If productivity has not, or does to a debate about how to use the surplus. not, pick up, then improvements in living Simply running a budget surplus would standards are going to be much less than help achieve the productivity growth goal. in most of U.S. history. Why? Because by running a budget surplus, If we could get productivity growth up by the federal government can add to 1 percent per year, from 1 percent to 2 percent the economy rather than subtract saving per year, it would make a huge difference for the from the economy. More saving means future. As I noted, that would take us to where more , raising , and productivity growth was in the 1950s and 1960s. increasing productivity. However, it is Such a growth rate would remove many prob- nearly impossible for politicians to run a lems facing us in the future: Social budget surplus for any length of time. would no longer be the problem it is today, the Every time we have seen projections of income would improve, and poverty surpluses in the past, we also have seen rates would decline. That is what happened proposals for spending the surpluses. And, during the 1950s and early 1960s when produc- of course, spending the surplus means we tivitygrowth was very strong. The income do not have a surplus. distribution narrowed, because the productivity Thus, a better approach is to establish gains spread across the whole population. a fiscal policy that will increase economic

FEDERAL RESERVE BANK OF ST. LOUIS

9 N OVEMBER/DECEMBER 1998 growth that does not require running a sur- state and local , interfere with plus.One idea is a recent proposal to production—in our case the production of partly privatize social security. It would research and teaching. To reduce this use the projected surplus to allow people interference and stimulate productivity to put funds equal to a fraction of their growth, we should have a regulatory policy payroll into a private savings account. that applies an effective cost/benefit crite- Those funds would then be part of national rion. If there is going to be a saving and would increase investment, and imposed on a university, on a private busi- thereby increase productivity growth. Other ness, on any other organization, it should people might have other ideas; with the pass the criterion that any other good +1%goal in mind, those alternatives can policy would pass—that the benefits be discussed and debated. outweigh the costs. Second, consider tax reform. Several proposals for tax reform would increase productivity growth. A flat tax, by exempting CONCLUSION investment from taxation, would stimulate In summary, during this lecture I have investment. Replacing the income tax with a pointed out both the good and the bad of retail sales tax would increase the tax on this period that I have called The Long relative to saving, encouraging Boom. As I see it, monetary policy deserves people to do more saving. Permitting much of the for what is good—the people to save in more tax exempt accounts— unprecedented degree of economic stability. extensions of the educational savings Of course, there will be recessions in the accounts and IRAs—would increasesaving. future, but if the Fed can maintain the Again there are alternatives,but with +1% kind of monetary policy rule we have had goal, we would have to adopt one of them. during The Long Boom and if the same Next, consider education reform. I type of monetary policy can be used by think this is the most important thing we other central banks in the world, then can do for productivity growth, though the more long booms will occur in the future. payoff will not start right away. Education Even with such stability, however, if reform would address the problem of the United States is not successful in workers not having enough skills or developing and implementing economic enough training to make use of new tech- policies to raise productivity growth, then nology in the workplace. Education reform the 21st century will not be one of much is controversial. I am in favor of greater progress and the current problems of choice of schools through vouchers, but social security and again the point here is to discuss, debate, will get worse. If the United States is in a and adopt some reform that will help period of a New Economy, as some argue, achieve the +1%goal. then it only just arrived and we should Finally, consider regulatory policy. adopt policies to maintain it. If the The United States has had deregulation of United States is not yet in a period of a entry and in a number of industries New Economy, which is more likely in including trucking, airlines, and telecom- my view, then we should adopt policies to munications, and that has been good for establish it. That is the purpose behind the economy. But there is another kind of the +1% goal for productivity growth that regulation called social regulation—which I suggested. includes environmental regulation, work I am sure rigorous economic research of and safety regulation—that has not been the kind Homer Jones advocated, directed, reduced. These types of affect promoted, and carried out will be essential everyone, not only business firms. At my to developing and adopting policies to raise university, for example, one can see many productivity growth and achieve such a goal. examples where regulations, not just from We need a new Homer Jones to help us find the federal government but also from the policies for just as we were

FEDERAL RESERVE BANK OF ST. LOUIS

10 N OVEMBER/DECEMBER 1998 lucky to have had the original Homer Jones to help us find policies for economic stability.

REFERENCES Friedman, Milton. “Homer Jones, A Personal Reminiscence,” Journal of (November 1976), pp. 433-36. Maisel, Sherman J. Managing the Dollar, W.W. Norton & Company (1973). Meigs, A. James. “Campaigning for : The Federal Reserve Bank of St. Louis in 1959 and 1960,” Journal of Monetary Economics (November 1976), pp. 439-53.

FEDERAL RESERVE BANK OF ST. LOUIS

11 N OVEMBER/DECEMBER 1998

FEDERAL RESERVE BANK OF ST. LOUIS

12