We faced some rather considerable challenges in assembling this forecast—the largest one, of course, being to find a way to be helpful to you in setting monetary policy in a period when the probability of imminent military conflict with Iraq has become very high. Early last fall, the Chairman commenced one of our pre-FOMC briefing sessions by asking whether the Greenbook forecast was ignoring the elephant in the room—the elephant being the potential for war. We responded that we were well aware of that possibility. But we believed that a forecast constructed to take explicit account of military conflict would focus attention on areas about which we knew very little and would distract us from providing you with our interpretation of incoming data and the underlying forces operating on the economy. Still, we recognized then that, by incorporating developments in oil futures and financial markets, our forecast was, in effect, reflecting the evolving likelihood of military action in Iraq.
Well, in this forecast, the room has become very small and the elephant very large. As you know, we opted for an approach that we think preserves maximum continuity with recent forecasts. We have continued to take on board the path of oil prices implied by futures markets, current readings in financial markets, and recent measures of consumer sentiment. These indicators, of course, embody a probability- weighted average of a wide range of possible outcomes, reflecting the collective judgment of market participants and survey respondents. At this point, it seems safe to presume that they place a very substantial weight on an imminent war. As events unfold in the coming days and weeks, readings on many variables conditioning our baseline forecast will shift, as outcomes currently embodied in that probability- weighted distribution become more or less likely. Oil prices, equity values, and sentiment are almost certainly going to deviate—perhaps substantially—from the paths assumed in the forecast. We attempted to give you a sense in the Greenbook of those factors on which we will be most focused and some rough quantitative assessment of the sensitivity of the forecast to movements in those variables.
At this point, you probably need little convincing that we don’t know what will happen in the weeks ahead with respect to conflict with Iraq or its economic consequences. But, you may reasonably ask, what are the economic data and financial market developments of the past seven weeks telling us about the underlying condition of the U.S. economy as we head into this critical period? If anyone’s hopes have risen momentarily that I will answer that question clearly, let me admit up front that I cannot. A quick look at the top line of our forecast might suggest that we have learned a great deal about the economy and that what we’ve learned has been decidedly negative. After all, we’ve revised down projected growth in the first half of this year to about 2¼ percent at an annual rate, about ½ percentage point less than in our January forecast. But in point of fact, owing to an upward revision to growth in the fourth quarter of last year, the level of real GDP is actually higher, on average, in the first half of this year than in our last forecast. One of the key surprises in the incoming data was that inventories were not as lean in the fourth quarter of last year as we had thought. Higher inventory investment accounts for all the 1¼ percentage point upward adjustment to the growth of real GDP last quarter; final sales were about unrevised. With that stockbuilding already behind us, we see less upward impetus to activity from this source in the first half of this year. Indeed, a smaller contribution from non-auto inventories more than accounts for the downward revision that we have made to our projection of activity in the first half of this year.
The recent data on spending, labor markets, and production have presented the usual mixed bag. For the most part, the strength in spending that was apparent late last year carried into January. To be sure, sales of new motor vehicles in January dropped off from the very high level observed in December, but other consumer outlays advanced at a rapid clip. Housing activity—both sales and construction— jumped to very high levels, aided by good weather and very favorable financing conditions. In the business sector, orders and shipments of capital goods in January exceeded our expectations, with spending for high-tech equipment posting an especially sharp rise. But the picture in February and into March was not nearly so bright. Motor vehicle sales posted another steep decline last month. Snowstorms in the East evidently clobbered sales in late February, but reports for early March suggest that there has been only a modest bounceback so far. Faced with sagging sales and mounting inventories, vehicle makers have announced significant cutbacks in production in the second quarter—a sign that they, too, are concerned about some underlying weakness in demand. Retail sales apart from motor vehicles declined last month as well, and this morning’s housing starts release—which showed single- family starts off nearly 14 percent, to 1.3 million units in February—can be added to the list of downbeat indicators for that month.
Readings on the labor market and industrial activity also worsened in February and have shown little sign of improvement this month. We were especially struck by the widespread weakness implied by last month’s labor market report. Private payrolls contracted by 321,000 in February. No doubt, bad weather contributed to the declines in construction employment, and the call-up of reservists likely crimped payrolls more broadly. But try as we might, we could find no silver lining in February’s employment situation. Our January forecast was built on the assumption that employment would be about flat early this year; instead, job losses have averaged about 100,000 over the past three months. Industrial production is also coming in below our earlier expectations. Declines late last year now look to be larger than previously estimated, and factory output is expected to eke out a much smaller gain this quarter than we had projected in January. Although the declines in production witnessed last fall seem to have abated, there appears to be little upward momentum to manufacturing activity. That view is supported by the available purchasing surveys, which have recently retraced earlier gains.
It is, of course, exceedingly difficult, if not impossible, to determine whether this most recent softening in the data reflects underlying weakness in the economy or the intensifying effects of war-related fears and uncertainties. Virtually all our major economic indicators at this point could be directly or indirectly influenced by actions or expectations related to the possibility of war. Is consumer confidence low because of higher war-related energy prices or because the labor market is so soft? Is the labor market soft because firms are hesitant to hire in advance of an imminent war or because product demand is weak? I don’t know of any way to confidently break into that argument. Moreover, I don’t think this is a signal extraction problem faced solely by the staff and other economic forecasters. I seriously doubt that households and businesses can reliably disentangle underlying economic trends from effects related to possible military conflict. Suffice it to say that, whatever the fundamental causes, the staff sees the economy on a somewhat shallower trajectory than we had earlier expected.
Moreover, a number of other factors led us to mark down a bit our projected growth of real GDP over the next year and a half. For one, stock prices at the close of the Greenbook were about 10 percent below those anticipated in our previous forecast, and we lowered our projected path by a similar amount. The effect of lower equity prices was only partly offset by the accompanying downward revision that we have made to long-term interest rates. I should note, however, that the rise in equity markets since we closed the forecast, if sustained, would, all else being equal, lead us to add back in much of the final demand we had taken out of the forecast. Higher current and projected oil prices, especially over the next few quarters, are expected to take a larger bite out of household incomes than in the January projection. Perhaps reflecting to some extent higher energy prices, consumer sentiment has weakened noticeably since the end of last year. Consumers’ views of their own financial conditions, broader business conditions, buying conditions, and the labor market situation have all deteriorated noticeably in recent months. Sentiment has dropped by more than can be explained by macroeconomic conditions, and in the forecast we expect this unusual weakness in household attitudes and their effect on consumption to fade only gradually over the next year and a half.
Not all of the key forces operating on the economy have been negative. Fiscal policy, in particular, is providing more upward stimulus to activity than in our previous projection. We have defense spending ramping up more sharply in this forecast compared with the last. We have added about $20 billion to projected defense spending in both 2003 and 2004. Separately, we have retained the package of tax cuts assumed in the January forecast, though we have delayed its projected implementation to the third quarter of this year. Meanwhile, tax receipts have been coming in much lower than we had expected, largely reflecting another year of unexpectedly large refunds. Smaller tax payments are providing a noticeable buffer against weaker labor income in this forecast and help to sustain the projected growth in consumer spending.
Despite the small net downward revision that we have made to the projection, we continue to expect a gradual acceleration of activity in the second half of this year and into 2004 for reasons that will sound quite familiar to you. Stimulative monetary and fiscal policy contribute to the pickup, as does a gradual diminishment of the drag from earlier declines in the stock market. Productivity growth shows no signs of flagging and should feed gains in both household incomes and corporate profits. Finally, we expect the unusual restraint on household and business spending, which we are attributing in part to pessimism and uncertainty, to fade over the projection interval. As has been the case for some time, this feature of the forecast must be considered more art than science. Even as art, I’ll admit it’s closer to Jackson Pollock than to Rembrandt. [Laughter] With slack in resource utilization persisting throughout the forecast period, core inflation is expected to recede modestly in 2003 and 2004. Indeed, core PCE price inflation drops to 1 percent in the second half of next year.
In working on this forecast, we—perhaps like many others in the economy—have felt as if we were biding our time, waiting for events to unfold and the situation to clarify itself. But how much clarification of economic circumstances can be expected? Certainly, a quick and successful resolution of the conflict with Iraq could be followed by a sharp decline in oil prices and a jump in equity values. Household and business sentiment could improve swiftly and persistently. With just a short lag, some high-frequency readings on the economy might provide evidence of follow- through to the real economy; initial claims, chain store sales, and reports from automakers would provide the early evidence on that score. With luck, those readings would be confirmed by broader monthly measures of spending, production, and labor market conditions. While it is less pleasant to contemplate, circumstances could clearly indicate a much less favorable outcome. The use of weapons of mass destruction, substantial damage to the oil fields, or slow military progress could result in sizable adverse reactions in oil and financial markets and a further souring of sentiment. The consequences of these developments could be reflected clearly in the flow of data on the real economy. In either event, a significant narrowing of uncertainties would allow us to get a much better fix on macroeconomic conditions and provide you with clearer direction on the appropriate setting of monetary policy.
But a third possibility seems eminently plausible—namely, that there will not be in the months immediately ahead a dramatic shrinkage in the uncertainty surrounding underlying economic conditions. There are a number of reasons for suspecting that this might be the case. Under the best of circumstances, financial markets will be volatile, and the data on the economy will be both noisy and often available with only a frustrating lag. It could take months to detect the economic consequences of the events that will unfold in the period ahead. Moreover, the best of circumstances may not present themselves. Even if the military situation concludes as quickly and successfully as it did in 1991, broader geopolitical risks may linger in a way that they did not after the earlier Gulf War. The threat of terrorist responses may not decline and could conceivably increase even in the wake of a successful military campaign. As a consequence, the current hesitance to spend and hire may not diminish rapidly. Moreover, there have been additional sources of geopolitical tension in recent months—North Korea and Iran, to name just two. We will need to be watchful of euphoric reactions in financial markets and among households and businesses. Those initial readings may signal a more vigorous rebound from recent doldrums, but that enthusiasm could prove unjustified over time by the underlying fundamentals. If so, the favorable initial reaction signaled by asset prices and sentiment would fade.
Finally, there could be a real economic bounce following the successful conclusion of a war. Some firms might act on spending and hiring plans that were delayed by war fears. Households, too, may boost spending that had been deferred by worries about war and the accompanying high levels of prices for gasoline and heating fuel. Witnessing such a rebound, however, will provide no assurance that the economy is moving onto a more persistently vigorous growth path. If the underlying health of the economy remains lackluster, a temporary period of above-trend growth could give way, yet again, to subpar performance. In effect, circumstances might be similar to a year ago. As the worst case scenarios following the attacks of September 11 failed to materialize, we saw a rapid acceleration of activity that could not be sustained by the underlying condition of the economy. I don’t want to exaggerate in either direction here. Clearly, we’re likely to know a great deal more than we do now should a swift and successful military action be completed in the period ahead. But I suspect that, even in those circumstances, you will see only a partial lifting of the fog of uncertainty that currently shrouds the economic outlook. Karen Johnson will now continue our presentation.