Growth Prospects for the U.S. Economy

William Poole*
President, Federal Reserve Bank of St. Louis

AAIM Management Association
St. Louis, MO
Dec. 20, 2002

*Kevin L. Kliesen, Associate Economist the Federal Reserve Bank of St. Louis, provided extensive assistance, but I take full responsibility for errors. The views expressed are mine and do not necessarily reflect official positions of the Federal Reserve System

Growth Prospects for the U.S. Economy

It is a pleasure to participate once again in an AAIM program. The end of the year is a traditional time to take stock of the state of the economy, and its likely future course, and that is what I'll do this morning.

The title of today's program, "The American Economy and Middle East Situation," presents considerable challenges. I'll talk about the "American Economy"—actually, the US economy-part of the topic. That is the easy part, and I'll leave the tough Middle East issues to others.

Before proceeding, I want to emphasize that the views I express here are mine and do not necessarily reflect official positions of the Federal Reserve System. I thank my colleagues at the Federal Reserve Bank of St. Louis, especially Kevin Kliesen, for their assistance and comments, but I retain full responsibility for errors.

My basic message today is that, while the economy is still working through some adverse developments that affected our economic performance over the past couple of years, it has displayed an extraordinary resiliency. The economy is fundamentally sound; our underlying economic strength has carried us through some difficult times.

Review of 2002

At this time last year, the consensus view among forecasters was that the economy, as measured by real gross domestic product (GDP), would grow by about 2-3/4 percent in 2002 and that the consumer price index (CPI) would increase a shade over 2 percent. With below-trend economic growth expected to prevail for most of the year, the unemployment rate was projected to rise from 5.6 percent in the fourth quarter of 2001 to 5.9 percent four quarters hence.

As we approach the end of 2002, the economy appears to have performed about as expected: If real GDP grows at about a 1-1/4 percent annual rate in the fourth quarter, and CPI inflation is about 2-1/4 percent at an annual rate for the quarter, as projected in the December 10 issue of the Blue Chip Economic Indicators, then economic growth in 2002 will be 2.9 percent, inflation will be 2.2 percent and the unemployment rate will probably average about 5.9 percent. In the forecasting business, last year's projections are tantamount to hitting the bull's-eye.

If the economy is performing about as forecast, why are so many so glum these days? Part of the story is that forecasters were right for the annual average but wrong on the pattern over the year. At this time last year, policymakers and business leaders alike were struggling with the economic uncertainties associated with the fallout from the horrific events of Sept. 11. The general expectation was that the economy would contract in the fourth quarter of last year and that economic growth in the first quarter of this year would be only a touch above zero. Then, the consensus view was, as the 9/11 shock wore off and as the expansionist policies put in place by the Fed and Congress took hold, the economy would be growing at about a 4 percent pace by the third and fourth quarters of this year.

Actual events have played out a bit differently. After growing at a 2.7 percent annual rate in the fourth quarter of 2001, real GDP surged to a 5 percent growth rate in the first quarter of 2002. Those two pretty strong quarters were certainly a surprise in the wake of 9/11. The economy then slowed appreciably during the second quarter, to a 1.3 percent rate, and accelerated again during the third quarter, to a 4 percent rate. Smoothing through this volatility shows that real GDP has grown at a bit less than 3.5 percent at an annual rate over the first three quarters of this year—which is pretty close to the economy's potential rate of growth. The soft patch in the fourth quarter finishes this year on a down note, and that is part of the story of why so many are so glum.

Consumer spending has accounted for about two-thirds of real GDP growth over the first three quarters of this year. Much of this strength reflects elevated sales rates of new light vehicles, the slower pace of sales in October and November notwithstanding. The two other sources of strength this year have been from a slower drawdown of business inventories and government gross investment and consumption expenditures. The latter, obviously, reflects the initial response to 9/11 events and the associated war on terrorism.

The expected slower rate of GDP growth in the current quarter reflects a projected decline in automotive output that could reduce GDP growth by more than 1 percentage point. The choppiness caused by the ups and downs of auto production and inventory investment is not a matter of fundamental concern. Indeed, some of the choppiness may well disappear in the future as revisions to seasonal factors tend to smooth short-run fluctuations.

Putting aside the extraordinary 9/11 tragedy, if that is possible, what do we make of the current recovery from the recession of 2001? The NBER Business Cycle Dating Committee has not yet announced officially that the recession is over but, assuming that the recovery began late last year or early this year, 2002 will have been one of the slowest economic recoveries in the post-World War II period. During the first four quarters of the average post-World War II recovery, real GDP increased by a little less than 7-1/2 percent.

In a typical recovery, the key drivers of growth are new private home construction (residential fixed investment), household expenditures on durable goods, like motor vehicles, furniture and home furnishings, and business spending on plant, equipment and software. This recovery has been unusual because these sources of strength have been much weaker than usual, resulting in below average growth.

The 2002 recovery has also been unlike the typical recovery in that equity prices have continued to be weak, which may be part of the reason that demand growth has been modest. All else equal, rising stock prices increase consumer wealth, leading consumers to increase their purchases by more than their income growth. For business, rising stock prices spur equity issue, which is used to finance investment in plant and equipment. When equity prices fall, the opposite effects occur. Ultimately, stock prices rise or fall because profits rise or fall. This year's poor stock market performance is readily explained by this year's dim profit performance: As reported by the Bureau of Economic Analysis, after-tax economic profits are down about 13 percent over the first three quarters of 2002.

Another reason for the weak business investment in 2002 may be related to the extraordinary rates of capital investment seen during the latter part of the 1991-2001 business expansion; the capital stock in a number of industries got ahead of demand. But most fundamentally, the modest recovery reflects the modest nature of the recession of 2001. The fact that the economy never sank very far means that there wasn't much of a bounce back in the cards anyway. Most noteworthy in this regard is the steady strength of housing demand. Housing construction usually declines substantially during a recession, and then bounces back during the recovery phase of the cycle. This time, housing was strong through the recession, and continued on a high plateau during the recovery. Housing growth was modest because the industry operated at a high level all along.

Nevertheless, the year does seem to be ending on a disappointing down note, and that is part of the reason many are glum. Perhaps a more important source of gloom is that employment growth has been essentially zero. But the flip side of the employment story is the remarkable surge in productivity growth. This topic is so interesting that it deserves some elaboration.

Productivity Surprise

One of the most noteworthy features of this recovery is the weak demand for labor. In fact, the labor market performance this year bears an eerie similarity to the so-called "jobless recovery" after the 1990-91 recession. In the first year of both recoveries, growth of nonfarm payrolls from the cyclical trough was negative and the unemployment rate was modestly higher.

At one level, the reluctance of firms to vigorously compete for labor in today's economy reflects the modest growth of output and aggregate demand. Yet at another level, firms may be reluctant to boost employment during this recovery because they have found ways to meet demand through improved production processes. Productivity growth—the growth of output per hour of labor input—has this year been considerably higher than what is normal.

To be more specific, the growth of labor productivity over the first year of the last four recoveries (excluding the short 1980-81 recovery) averaged about 4-1/4 percent. Assuming that the 2001 recession ended sometime during the last quarter of 2001, growth of labor productivity over the first three quarters of 2002 has been nearly one percentage point faster. A crude back-of-the envelope calculation shows that this extra one-percentage point of labor productivity growth over the first three quarters of 2002 has effectively been substituted for a net job creation of a little more than 2 million workers, which would have occurred had the average job growth of the past four recoveries held. In fact, 2 million may be too low, since the average job growth includes the 1990-91 jobless recovery.

In no way do I mean to imply that this extra one percentage point of productivity growth during the 2002 recovery has been bad for the economy. Strong productivity growth entails far more benefits than costs. Indeed the long-term benefits of, say, 3 percent productivity growth versus 2 percent productivity growth are huge. The problem is not that productivity growth is too high but that GDP growth is too low to create satisfactory employment growth.

In the short-run, firms have been able to substitute capital for labor because of the tremendous rates of capital investment they made during the 1990s and the application of improved business practices that are coming closer to realizing the full potential of the enlarged capital stock. What really matters is not that a PC replaces a typewriter but that the PC makes possible changes in business practice that squeeze more output from each hour of labor input. This very speech is a simple example: Based on an outline I sent by email, Kevin Kliesen drafted it and I put it into my own style working on a computer at the St. Louis Fed branch in Louisville, with the assistance of comments Bob Rasche sent me by fax.

Recent research by experts in this field suggests that a reasonable estimate of long-term US labor productivity growth going forward is between 2 and 3 percent(1). If so, then productivity growth will be significantly higher than it was from 1973 to 1995, when US productivity growth averaged about 1.4 percent per year.

Of course, there is always the possibility that some part of the recent surge in productivity growth will be revised away. For policymakers, one of the difficulties in discerning the trend rate of productivity growth stems from revisions to the data, which can sometimes be quite substantial. For example, the 1998 and 1999 annual revisions indicated that productivity growth was noticeably faster than originally thought. Conversely, the 2001 and 2002 revisions significantly lowered estimates of productivity growth. During these periods, revisions to output growth were the driving factor, as hours were hardly revised.

Some preliminary research at the St. Louis Fed shows that, since 1985, there has been little correlation between the initial productivity figures released by the Bureau of Labor Statistics and final figures that incorporate annual revisions. For example, if the initial published productivity growth rate is 2 percent, an approximate 95-percent confidence interval for the final revised figure is 0 to 4 percent, a huge range. For this reason, I'm hesitant to put too much emphasis on this year's strong productivity numbers. Knowing this history, I don't want to go out on a policy limb, or any other limb, based on the early estimates of productivity growth.

That said, this year's numbers are consistent with an evolving upward trend in labor productivity growth, and they are consistent with anecdotal information from talking with business leaders. I don't want to get bogged down in the details, but the experts suggest that a large percentage of this acceleration is IT-related—both quantitatively and qualitatively. From a quantitative standpoint, there was a surge in business capital spending over the past several years: Real investment in equipment and software as a share of real GDP effectively doubled to 6 percent from 1995:Q1 to 2002:Q3.

While the tremendous increases in physical volumes of capital goods have been important, so has the qualitative aspect of these capital goods. Qualitatively, the new technologies embodied in these computers, servers, telecommunications equipment, and the like has also enabled firms to produce more with less. At some point, though, as aggregate demand growth starts to pick up and firms, responding in kind, begin to raise the pace of their hiring, the current exceptionally high rates of productivity growth will slow to rates approximating their long-term trend. The capital-labor trade-off may be more permanent in some industries than others, but in the aggregate the faster rate of growth of living standards enabled by enhanced rates of productivity growth is unambiguously a net plus for the economy.

Outlook for 2003-04

I'll now turn to the outlook for the US economy over the next year or two. Inherent in any forecast of the future are many leaps of faith. In my mind two stand out. First, any forecast that is generated from a macroeconometric model presumes that economic outcomes are tied together in a certain fashion. A model's structure reflects both the theoretical biases of the forecaster and the historical regularities in the data. Clearly, the theory can be wrong or incomplete, and the interpretation of the data flawed.

The second key assumption, or leap of faith, that forecasters have to make concerns the likely path of economic variables in the model. Key assumptions in this regard are the path of oil prices, the Fed's interest rate target, tax rates, depreciation rates on capital goods, foreign exchange rates and—sigh—the course of the stock market. Clearly, a forecast depends on a number of factors that may change unexpectedly—9/11 is a perfect example of an important but inherently unforecastable event. With that in mind, let me walk you through a scenario that seems credible knowing what we know now.

When thinking about the outlook, I am always well aware that the only sensible stance for me is to be an informed consumer of forecasts. I am not myself an expert forecaster, and do not—cannot—spend enough time forecasting to expect to outperform those who make a living from that occupation. I am well aware of the range of forecasts—expert forecasters disagree. Research suggests that the best forecast for a consumer like me is some sort of average of the range of forecasts.

For these reasons, I place a great deal of weight on the so-called Consensus forecast published in the Blue Chip Economic Indicators. Each month, about 50 of the nation's top private-sector forecasters are polled on their outlook for several key economic indicators (e.g., growth of real GDP, the CPI, industrial production and the level of the unemployment rate) over the next several quarters. The Consensus is just the average of these individual forecasts. However, and this is an important point, I am always impressed by how quickly the consensus forecast can change. I am not willing to bet my house—or my policy position—on the forecast. Experts differ, and over a few months time they sometimes change their forecasts significantly.

With that introduction, I'll offer an overview of the Consensus forecast for the US economy. First, it seems likely that the growth of consumption—that is, spending by households on durable and nondurable goods, and services—will remain around 3 percent, which is modestly slower than the growth seen during the latter part of the 1990s (and even over the past four quarters). Although strong growth of labor productivity will continue to undergird household incomes, current saving rates seem abnormally low in light of the pending demographic challenges associated with the retirement of the baby boom generation. Ultimately, some rebalancing of spending relative to incomes seems necessary—though this process could play out over several years. In addition, some of the factors that have boosted consumption in recent years—rapid growth of household wealth, mortgage refinancing and mortgage cash-outs-are not likely to be the source of extra spending that they were a few years back. In short, consumption growth is likely to be a solid 3 percent, more or less, but to lag income growth slightly as households return the saving ratio to a more normal level.

Second, whether near-term real GDP growth is faster or slower than the Consensus expects will depend crucially on business outlays for capital goods. At present, a majority of forecasters expect the pace of fixed investment to be stronger during the second half of 2003. This outlook is also consistent with a couple of recent surveys of firms published by the Federal Reserve Bank of Philadelphia and the Institute for Supply Management. However, the timing of the recovery in fixed investment is very uncertain, owing to some key factors. These include current low rates of capacity utilization, due to the high rate of investment in the late-1990s; relatively high vacancy rates in commercial and industrial buildings; a guarded outlook for corporate profits; uncertainties associated with a possible war with Iraq; and the threat of future terrorist attacks occurring on US soil or affecting US interests abroad.

I've offered a formidable list of factors holding back investment in 2003, but there are also reasons to be optimistic about the pace of capital spending. First and foremost, high-tech capital depreciates rapidly. Hence, fairly rapid growth of high-technology investment—an increasing share of investment spending—is needed to keep net investment rates stable or growing. Replacement investment is important when standard practice is, for example, to depreciate a PC over three years. Second, business planning could become appreciably less challenging once geopolitical uncertainties are resolved. And if the uncertainties are not resolved, from my experience what happens is that businesses learn to live with the problems and get on with their pursuit of new products and new markets. Indeed, if we can use the stock market as a barometer, it appears that some uncertainty has been whittled away of late, as equity markets have rallied modestly since setting five-year lows in early October. Finally, economic activity should continue to be boosted by an accommodative stance of monetary policy, with some assistance from fiscal policy.

Fundamentals of the Long-Term Outlook

Given the uncertainties I've noted above, the timing of the transition to higher growth remains unclear, but probably not the eventual outcome. To believe otherwise implies an alternative view that something is structurally unsound in our economy—either much weaker economic growth is likely, or inflation is poised to accelerate. At this point, I'm willing to place my marker on the side of those who point to flexible markets, rapid innovation, high productivity growth, and low inflation as the linchpins for a return to solid economic growth over the medium term.

Too often, though, observers of the economic scene get caught up in the high frequency data and neglect longer-term issues. The central bank's primary input to the mix of conditions that promotes high growth in employment and output is to pursue policies that maximize prospects for low and stable inflation. The evidence is that the Federal Reserve has been successful in this regard. Current rates of inflation are low. Just as important, inflation rates expected by financial markets, consumers and forecasters give no credibility to the assertion that the inflation outlook is poised to deteriorate anytime soon—either on the upside or on the downside.

That said, the current situation is not one in which monetary policymakers can afford to relax. Monetary policy is very accommodative right now. Short-term interest rates are exceptionally low and money growth is strong. Given the downward pressures that have slowed the economy's recovery, this stance has been appropriate. As I have emphasized on several occasions, this policy stance has been possible because the Fed's credibility is high. The markets believe that the Fed will reverse course when necessary to prevent inflation from rising, and I'll certainly do everything I know to do to assure that outcome.

Having made this point, I do not want to be read as hinting that I'm currently on the edge of pushing for a tighter policy. The objective is sustained low and stable inflation; given all the short-term economic negatives I've discussed, the Fed needs to be alert to deflationary as well as inflationary dangers in the months ahead. If necessary, there is room to cut the federal funds rate target some more, and to pursue other policies if we run out of room on the funds rate policy instrument. I think it is unlikely that we'll run into that problem, but thinking through every possible contingency is what creates a robust and competent monetary policy. In short, monetary policy changes in the future will be driven, as they have in the past, by arrival of new information on how inflation prospects and growth prospects are evolving.

Thank you, and I'd be happy to entertain a few questions.

Footnote:

(1) Jorgenson, Dale W., Mun S. Ho, and Kevin J. Stiroh (2002), "Projecting Productivity Growth: Lessons from the US Growth Resurgence," and Stephen D. Oliner and Daniel E. Sichel (2002), "Information Technology and Productivity: Where Are We Now and Where Are We Going?" both in Federal Reserve Bank of Atltanta Economic Review, third quarter, 2002.

Back to top