U.S. Economic Outlook*
Remarks by Michael H. Moskow
President and CEO, Federal Reserve Bank of Chicago
Federal Reserve Bank of Chicago
Levy Economics Institute
Embargoed for release:
Friday, April 23, 2004
11:45 a.m. E.T.
10:45 a.m. C.T.
or upon delivery.
When I think about monetary policy, I often find it useful to think in terms of the output gap—the gap, or difference, between the levels of actual and potential output. This gap, of course, is a convenient shorthand concept. It summarizes the effect of numerous frictions in the economy that can slow the adjustment of aggregate demand and productive resources towards a sustainable equilibrium. Sometimes these frictions cause substantial underutilization of resources, and the resulting positive output gap puts downward pressure on inflation—this is the environment we've been in most recently. At other times, demand exceeds the long-run sustainable productive capacity of the economy, and the resulting negative output gap generates inflationary pressures.
Unfortunately, the output gap is not something that can be measured with precision. Even the best quantitative analysis is only an estimate and needs to be supplemented with judgement. To give an example that's particularly important right now, consider the remarkable acceleration in productivity over the past couple of years. Some fraction of this acceleration represents a permanent increase in trend productivity growth, but some of it is purely transient. The more permanent the gains, the larger the output gap. But statistical analysis can't give us more than a rough guess about how much is permanent, and how much is transient. So, in measuring the output gap, we must include a dose of good judgement as part of the mix. More generally, we always have to formulate monetary policy in an environment of irreducible uncertainty.
One way to help deal with this uncertainty is to examine a variety of statistical models and the behavior of a number of different indicators—such as inflation and labor market conditions—to gauge the size of the output gap. On balance, these suggest that the current level of actual output is still below potential. But, with the impressive growth of economic output in the past nine months, the output gap has narrowed. Part of the Fed's current policy challenge is to help accommodate economic activity in closing the remaining gap, without—sometime down the road—overshooting the level of potential output and generating worrisome inflationary pressures.
Of course, as the economy moves toward a sustainable growth path, imbalances may develop in individual sectors or industries, and policymakers must consider whether these pose a risk to the economy at large. But monetary policy is a blunt tool. It is wholly incapable of addressing the concerns of individual sectors of the economy—it can only be used to influence the overall economy, taking into account the aggregate impact of the individual sectors.
Today, I'd like to review recent economic developments in the context of the output gap, discuss labor market conditions, talk about the outlook for economic growth and inflation, and discuss what they mean for monetary policy.
Recent economic conditions
Our current output gap can be traced back to the recent recession, which the NBER tells us began in March 2001 and ended in November 2001. During this period, real GDP declined as businesses significantly cutback on their capital investments and inventory spending. By historical standards, the recession itself was mild, both in terms of its duration and its effect on total output. The recession ended just eight months after it began, and real GDP contracted only 1/2 percent between its peak and trough—about one-quarter the average decline during the other recessions since 1960.
The subsequent recovery began at a moderate pace, in part, because of the mild nature of the recession and, in part, because of a series of shocks that hit the economy. The start of the war on terrorism, revelations of corporate malfeasance, and the buildup to the war in Iraq all led to heightened uncertainty and diminished confidence about the economy.
As a result, while we experienced a few short bursts of strong demand growth early in the recovery, the economy failed to sustain a vigorous pace of expansion. During 2002 and the first half of 2003, real GDP growth averaged about 2-3/4 percent—which is below the 3 to 3-1/2 percent that we estimate as the potential output growth rate. So, even though the economy stopped shrinking at the end of 2001, the output gap did not close.
As we moved through 2003, we saw signs of improvement in several key sectors of the economy. Household spending, which had held up quite well throughout the recovery, posted solid gains. But, more importantly, business confidence improved and spending on capital equipment rose strongly in the second half of 2003. Inventory investment also increased.
The economy's momentum has continued, and, according to the median forecast from the latest Bloomberg survey, real GDP increased 5 percent in the first quarter of this year. If this estimate is correct, real GDP increased at more than a 5-1/2 percent annual rate over the past three quarters. This is a good deal stronger than even the most optimistic estimates of the rate of growth in potential output, and implies that we have made significant progress in narrowing the output gap.
Job growth has been the missing link for much of the recovery. True, the recent increase in demand and confidence has led to some improvement in labor markets. Unemployment insurance claims and other indicators suggest that the pace of layoffs has slowed substantially. And we have seen some job gains recently—308,000 new jobs in March and an average of 171,000 new jobs per month in the first quarter.
But even with the recent gains, payrolls are still below where they were when the recession ended. In the two years after a typical recession, employment increases by 5 percent. Even two years after the 1990-91 recession, which was the original jobless recovery, employment had not only surpassed its level at the end of that recession, but also its previous peak.
Analysts have proposed many explanations as to why we haven't seen the kind of job growth we typically do early in an expansion. One we hear frequently is described as unusually high sectoral reallocation. Proponents of this explanation claim that extensive changes in the economy have resulted in a greater-than-normal need for workers to move among industries and occupations. As evidence, they point to the sharp declines in manufacturing employment and to the increased ability to outsource both manufacturing and service jobs to other countries. Certainly these developments are noteworthy. But do they imply that sectoral reallocation is unusually high?
After all, our economy is very dynamic—it is constantly changing due to a variety of factors. Consumers shift their demand for products and services. International markets open up new opportunities for trade. And new technologies emerge that create entire new industries. As a result, jobs disappear from industries that are shrinking and many new ones are created in industries that are expanding.
The monthly job numbers don't fully reflect the huge amount of churning that actually takes place in labor markets. The 308,000 new jobs in March was only the net change in employment. In fact, more than 2-1/2 million jobs are created on average each month, and approximately the same number are destroyed.
As one would expect, the gross rate of job destruction spiked during the recession but has fallen back to the low level that prevailed before then. In contrast, the gross rate of job creation fell during the recent recession and, unfortunately, has been slow to recover. But the fact that both rates are now relatively low suggests that the current pace of job reallocation is not especially high.
Furthermore, despite all of the discussion about the decline of manufacturing and the increase in international outsourcing, neither factor seems to be unusual enough to make this a period of heightened reallocation.
With regard to manufacturing, factory jobs are usually hit harder by business cycle downturns and periods of weak growth. Manufacturing's share of total employment has been trending down for decades. This long-run trend is largely due to higher productivity growth in the manufacturing sector compared to the overall economy. This evidence is consistent with research by one of our economists at the Chicago Fed, Ellen Rissman(2003). She found that most of the recent changes in employment shares across manufacturing and other broad industries appear consistent with long-run trends and the usual changes over the business cycle.
With regard to outsourcing to other countries, accurately measuring its impact is quite difficult, and we do not have good information on the number of jobs that have moved overseas. But, according to one estimate from Goldman Sachs, only 300,000 to 500,000 jobs have been lost to outsourcing since 2000. For an economy the size of ours, these are not large numbers—especially in relation to the 2-1/2 million jobs created and lost each month in the U.S.
Also, outsourcing and other forms of international competition are not new. Over the past 50 years, we have been challenged by growing competition from Japan, the Asian Tigers, Mexico, and, now, China and India. Yes, certain jobs are lost forever, but our economy is remarkably flexible. Technological advances often give rise to entire new industries, new products, new services, and, ultimately, new jobs that replace the jobs lost to foreign competition. That's why over the past 50 years, even while there has been a decline in the manufacturing sector's share of employment and increases in international competition, the U.S. economy has generated over 80 million net new jobs.
Because these new jobs are often the result of unforeseeable innovations in technology, it is impossible to say now what kind of jobs will be created in the future. When I was in Milwaukee recently, I learned that GE Healthcare had announced plans to expand their Information Technology and Ultrasound operations in Wauwatosa, Wisconsin. As many as 2500 employees at the new facility will engineer and manufacture innovative medical equipment, including computed tomography, or CT systems. These products allow doctors to readily diagnose health problems like cancer or aneurysms without exploratory surgery, and the machines are now widely available. It would have been virtually impossible to anticipate these jobs five years ago, because no one could predict how the use of CT technology would evolve.
Now, I do not want to minimize the impact of the job losses associated with changes in the economy, no matter what their source. The human costs of this process are significant and compelling. Factories or offices may close and workers lose their jobs. We must always strive to ease the transition for these workers and their families, whether it is through financial assistance, retraining programs, or other efforts. But the dynamic changes in the economy are important if we are to continue to increase overall incomes and our standard of living.
If there isn't an unusual amount of reallocation taking place, what else might have caused weak employment growth? A number of other factors could be in play.
For example, new employment practices, which some call just-in-time hiring, have likely played some role in restraining employment growth. Firms have greater flexibility in adjusting their work forces, thanks to developments like the rise of temporary help services and consulting industries, the widespread use of the Internet for job posting and application, and the growth of businesses that match workers with firms. Take the temporary help industry for example. Twenty years ago, the temp industry was very small, representing only about 1/2 percent of nonfarm payroll employment. Since then, it has increased significantly and now comprises close to 2 percent of employment. At some firms, the majority of employees are temporary workers, allowing these firms to easily vary the scale of their operations. Furthermore, temp agencies now provide more light industrial and technical workers than they did twenty years ago.
To understand how just-in-time hiring could restrain overall employment, consider the staffing decisions of a firm. Initially, hiring new permanent workers takes time and money, and once they are hired, they may be expensive to lay off. When you combine these restrictions with an environment of uncertainty, the firm faces a tough choice. On the one hand, if it increases its employment and demand is weak, then high payroll costs will crimp profits. On the other hand, if the firm maintains its current level of employment and demand picks up, then it may initially be unable to increase production in line with orders.
The past few years had been marked with especially heightened uncertainty-even more so than the usual uncertainties prevalent when an economy recovers from recession. At the same time, the burgeoning temp help services industry and the spread of consulting services gave firms greater flexibility in their hiring decisions. Before hiring workers on a permanent basis, they had the option of waiting until they were more confident that demand would strengthen, because they could rapidly increase employment when needed (Aaronson, Rissman, and Sullivan, 2004).
These explanations suggest there's been a structural change in the labor market. However, we shouldn't forget an old and pretty fundamental explanation—employment growth will be slow if aggregate demand is weak relative to gains in the productive capacity of the economy. Aggregate demand languished in 2002 and early in 2003. But, firms continued to be successful in developing innovative ways to reorganize their production and distribution networks, especially by exploiting advances in technology. In addition, they were able to take advantage of the significant capital investments of the late 1990s. As a result, productivity increased rapidly and the potential productive capacity of the economy likely advanced much more sharply than demand. Since the middle of last year, however, aggregate demand outpaced even the highest estimates of potential output growth, and led to an increase in employment.
Going forward, I believe that economic growth will remain solid, so I am optimistic that employment will accelerate. Why? Productivity gains, fiscal stimulus, the Federal Reserve's highly accommodative monetary policy stance, replacement demand for capital equipment, and improving conditions abroad all should contribute to continued strong economic growth. In the Fed's Monetary Policy Report to Congress in February, the central tendency forecast was for real GDP to rise 4-1/2 to 5 percent this year. Given the data we've seen since then, I still believe this forecast is a good one.
On the price front, the most recent readings on CPI inflation are up from the extremely low inflation rates we saw late last year. This has not just been a pickup in the volatile food and energy components—core inflation is up as well. And we are all reading more stories in the press and hearing more concerns from various contacts about rising inflation.
To be sure, there are factors in play that could boost inflation: Unused resources could fail to move efficiently to the sectors where demand is growing; or the fall in the dollar could show through more in the price of imports; or higher commodity prices could be passed through to the consumer level. Finally, as I mentioned earlier, no one can measure the output gap precisely—it may be the case that trend productivity growth is lower and, therefore, the output gap is narrower than many believe.
These are risks to watch. But even with the recent increases, the level of consumer price inflation is still very low. And, at the macro level, we have yet to see the kinds of pressure on labor and capital resources that would foreshadow a worrisome increase in inflation. Thanks to strong sustained productivity growth, unit labor costs—that is, compensation per unit of output produced in the economy — have been falling for the past two years. Even though outright declines are unlikely to continue as labor markets tighten, I think strong productivity trends should help keep overall cost pressures in check.
Nonetheless, we must remain vigilant. As I discussed earlier, the output gap is narrowing, and part of the Fed's current policy challenge is to avoid overshooting the level of potential output and generating inflationary pressures. Currently, the federal funds rate is very low. Indeed, if recent developments have raised inflationary expectations, then the real funds rate has come down—even though we did not change the nominal funds target. Clearly, we cannot maintain this degree of policy accommodation indefinitely. As the output gap narrows further, the real funds rate will have to rise to a level more compatible with long-run sustainable economic growth.
So, in conclusion, I'd say that the outlook for growth in 2004 is encouraging. A year ago, we were worried about unwelcome disinflation; that concern has now subsided. But I do not see broad-based pressures developing that would lead to a significant increase in inflation. Moreover, our economy has begun to meet the challenge of the jobless recovery.
But this is not surprising. Time and time again our economy has proven itself resilient in the face of challenges. In spite of the jobless recovery of the early nineties, the expansion lasted a decade and created 24 million new jobs, at the same time that inflation generally continued to drift down. With its entrepreneurial culture, market-based principles, and continuing technological advances, I am confident that our economy has the ability to handle its current challenges, and the foundation to enjoy solid growth and price stability in the years ahead.
*The views presented here are my own, and not necessarily those of the Federal Open Market Committee or the Federal Reserve System.
Aaronson, Daniel; Ellen Rissman; and Daniel Sullivan (2004). "Assessing the Jobless Recovery," Federal Reserve Bank of Chicago, Economic Perspectives (second quarter, forthcoming).
Rissman, Ellen (2003)."Can sectoral labor reallocation explain the jobless recovery?" Federal Reserve Bank of Chicago, Chicago Fed Letter (December).