Transcripts of the monetary policymaking body of the Federal Reserve from 2002–2008.

Thank you, Mr. Chairman. I’ll wait for my colleagues to come to the table. We’ll be using the chart package that you all should have on the economic outlook. Separating the signal from the noise in the recent economic data has not been easy—what with the motor vehicle anomaly, the defense spending pull- forward, and the transitory swings in oil imports. We tried to cut through the clutter by highlighting in the Greenbook real private domestic final purchases, or PDFP— that is, the sum of consumption, residential investment, and business fixed investment. We think this aggregation, which is shown on line 3 of the table in exhibit 1, currently is giving a fair representation of the thrust of aggregate demand. The data that we have received since the December meeting have been stronger than expected. As a result, we have revised up our estimates of the growth of PDFP in both the fourth and the first quarters to annual rates of about 2 percent—roughly the same rates as those in the middle two quarters of 2006.

The remaining panels of exhibit 1 highlight some of the indicators that have informed our judgment about the current pace of activity. Starting with the labor market, the middle left panel, increases in private payroll employment averaged 119,000 in the fourth quarter, close to the average pace in the preceding two quarters. As you may remember, at the last FOMC meeting we commented on the stronger signal for activity coming from the labor market compared with the spending data. That tension seems to have been largely resolved, not because of weaker employment but because of stronger spending—especially consumption. Retail sales increased briskly in November and December; accordingly, in this Greenbook, we boosted our estimate of fourth-quarter real PCE growth, the middle right panel, to an annual rate of about 4½ percent. The fundamentals for consumption remain quite solid: steady employment gains, recent declines in energy prices that have raised real income, well- maintained consumer sentiment, and further increases in stock market wealth. That said, at least according to some of our models, the fourth-quarter pace of consumer spending was stronger than would have been consistent with those fundamentals. Our forecast for the growth of real PCE in the first quarter, at 3.6 percent, reflects a bet that some of the surprising fourth-quarter strength will carry forward for a while. Turning to housing, sales of new and existing homes—which are not shown in the exhibit—appear to have stabilized in recent months, and the ratio of new home inventories to sales has moved down a bit. As shown in the bottom left panel, the apparent stabilization of housing demand may now be starting to show through to permits and starts for single-family homes. Of course, the unusually warm weather in December makes a definitive assessment at this time particularly difficult. In the business sector, investment spending slowed appreciably in the fourth quarter. In particular, shipments of nondefense capital goods, the red line in the panel to the right, have been unexpectedly soft recently, including the December figure that we received after publishing the Greenbook. Part of the recent weakness in this category appears to be for purchases of equipment related to construction and motor vehicle manufacturing. With orders remaining above shipments, we expect real equipment spending to rise modestly in the first quarter.

Exhibit 2 takes a closer look at some recent developments, starting with an examination of the effects on the industrial sector coming from the recent sharp declines in the production of light motor vehicles and residential investment. By our reckoning, production of light motor vehicles, the top left panel, tumbled nearly 20 percent at an annual rate in the third quarter of 2006 and dropped further in the fourth quarter. Meanwhile, we estimate that residential investment, shown to the right, plunged at an annual rate of about 20 percent in both the third and the fourth quarters. In thinking about the effects of these developments on industrial production, we need to keep in mind the upstream effects. As noted in the bulleted items in the middle left panel, the drop in production of light motor vehicles affects IP not only through its direct effect on light motor vehicle manufacturing but also indirectly through its influence on production in upstream industries such as primary metals, tires, and, nowadays, semiconductors. In the case of construction, of course, all the IP effect comes through the influence on upstream industries—lumber, concrete, plumbing fixtures, and so forth. The table to the right shows the estimated effects on IP growth, including upstream effects, associated with the declines in the production of light motor vehicles and residential construction illustrated in the top panels. We have used input-output relationships to estimate the direct and upstream effects and then translated these effects into their IP contributions. Lines 2 and 3 show that, after we account for upstream influences, motor vehicles and residential construction were sizable drags on IP in the third and fourth quarters. Yet, as shown in line 4, the drag from those two sectors was not the whole story. Even so, we think the more likely track from here forward involves modest growth rather than a cumulative weakening of industrial activity, in part because we think that most producers have been moving reasonably promptly to address any emerging inventory problems.

The bottom panels widen the scope from the industrial sector to the economy as a whole and address the question of whether developments in less-cyclical industries have been helping support economic activity. My colleague Stephanie Aaronson divided the establishment survey employment data into three categories—highly cyclical industries, moderately cyclical industries, and acyclical industries—based on the correlation of individual industry employment changes with the GDP gap. The bottom left panel presents some history, with the highly cyclical grouping plotted by the black line and the moderately cyclical plotted in red. To keep the chart easier to read, the acyclical group is not plotted. The chart shows what you might have expected ex ante: Fluctuations in both series are highly correlated, but the amplitude of swings in the moderately cyclical is more damped. The panel to the right puts a microscope on the past few years—note the change in scale. As you can see, despite the step-down of employment gains in the highly cyclical industries, employment in the moderately cyclical industries has continued to grow apace. This suggests that the softness we’ve seen lately in residential construction and some parts of manufacturing has not spilled over to other parts of the economy.

That conclusion is an important factor that has shaped our view about the longer- run outlook for the economy—the subject of exhibit 3. As shown in line 1 of the table in the top panel, later this year the growth rate of real GDP is expected to move back up toward our estimate of the growth rate of potential, and it stays there in 2008. This basic pattern is unchanged from the last Greenbook. The bullets in the middle panel highlight some of the major forces shaping this projection. The most important is our forecast that the restraint from housing will diminish this year and that its contribution to GDP growth will turn slightly positive next year. Second, the recent declines in oil prices, plotted in the bottom left panel, have raised real income; we believe that the drag from the earlier increases in oil prices should dissipate in the near term, and over time the stimulus from the recent price declines should begin to predominate. Third, federal fiscal policy, the bottom right panel, also is a bit stimulative, although the impetus is projected to ebb over the next two years. Finally, given our conditioning factors, the assumed path of the nominal federal funds rate is consistent with a real funds rate that closes the output gap over time.

Exhibit 4 focuses on the components of PDFP. As I noted earlier, the leveling-off of home sales, the uptrend in mortgage applications, and the improvement in homebuying attitudes suggest that housing demand may be leveling off. The top panel shows the historical relationship between housing demand, as measured here by sales of new homes (the red line), and housing construction, shown here by single- family housing starts (the black line). The shaded areas highlight previous housing downturns as well as the current situation. As you can see, cyclical recoveries in sales and starts have generally been fairly coincident historically. You’ll have to take my word for it, but this has been the case even when the inventory of unsold homes has been high. Accordingly, we think that the recent stabilization of sales should be accompanied soon by a stabilization of starts. Then, as sales move up, so should starts. The middle panels focus on the consumption forecast. We expect real PCE, the red bars in the left panel, to increase 2¾ percent this year and next. The forecast reflects two main crosscurrents. On the one hand, real income growth, the blue bars, is projected to be robust, reflecting, in part, continued increases in real wages as well as further employment gains. On the other hand, the wealth-income ratio, plotted by the black line in the panel to the right, falls in our forecast as house prices appreciate only about 1 percent per year. With slower gains in wealth, and spending gradually coming back into line with fundamentals after the current period of unexplained strength, the saving rate should rise.

The bottom panels present some details on the outlook for business fixed investment. As illustrated in the bottom left panel, total real outlays for equipment and software, excluding the volatile transportation equipment component, are projected to increase about 6 percent both this year and next. You can see from the red portion of the bars that the bulk of the support comes from spending for high-tech equipment as telecommunications service providers further expand their fiber optic networks and as businesses continue to invest in information technology equipment and software. We expect the contribution from the other equipment category (the blue portion) to narrow this year and then to edge up in 2008, largely reflecting the pattern of changes in the growth of business output. The bottom right panel shows our forecast for nonresidential structures excluding drilling and mining. The incoming information on construction outlays for nonresidential buildings and the forward-looking indicators that we monitor suggest that spending growth has downshifted. Accordingly, after rising 12¾ percent in 2006, real outlays for this component of nonresidential structures are expected to decelerate to a pace of 5½ percent this year. Our projection for 2008 brings growth in this component of nonresidential structures down to its long-run average. Bill will now continue our presentation.

Keyboard shortcuts

j previous speech k next speech