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

Thank you, Mr. Chairman. As I worked my way through a final reading of the Greenbook this weekend, I was reminded of the old joke about the man who is told by his doctor that he has only six months to live. The doctor recommends that the man marry an economist and move to North Dakota. The man asks whether this will really help him live any longer than six months. The doctor says, “No, but it sure will feel a lot longer.” [Laughter] Part 1 of the Greenbook was its usual twenty pages, but with lengthy discussions of motor vehicle mismeasurements, PPI inventory deflator anomalies, income revisions, and footnotes on errors in Okun’s law, it sure read as though it were a lot longer than twenty pages. So, in the kinder spirit of the season, I thought that I would jump straight to the bottom line of this forecast.

The bottom line is that our outlook for economic activity has not really changed much from the one presented in the October Greenbook—or for that matter, the September Greenbook. The economy still appears to us to have entered a period of below-trend growth that will eventually relieve some of the pressures on resource utilization that we believe have developed over the past few years. As in our previous forecast, the current and expected weakness in aggregate activity is being led by a steep contraction in homebuilding. Indeed, the recent readings on housing starts and building permits were a little softer than we had been forecasting, and we have marked down our forecast of residential investment a bit further. We currently estimate that the drop in residential investment is taking about 1¼ percentage points off the annualized growth of real GDP in the second half of this year, and we are expecting a similar-sized subtraction from growth in the first quarter of next year.

We were also surprised to the downside by the October reading on construction put in place in the nonresidential sector. As you may recall, we had been expecting some slowing to become apparent by early next year as a deceleration of business sales, smaller employment increases, and less-rapid growth of equipment spending reduced businesses’ needs for space. Moreover, while fundamentals have improved in commercial real estate markets in recent years, they are best characterized as only moderately favorable; vacancy rates for office and industrial buildings are still elevated by historical standards, and rental income has been rising at only a tepid rate. All told, we have interpreted the softer readings of the past couple of months as suggesting that the slowdown in nonresidential construction has arrived a bit sooner than expected, but we don’t see an outright slump as the most likely outcome in this sector.

Business spending on equipment has unfolded pretty much as we had expected. The report on durable goods orders was widely read by others as surprisingly weak. In part, that weakness resulted from a 25 percent drop in orders and shipments of computers that we are extremely skeptical about and that the BEA will significantly downweight in estimating investment spending. Among the pieces that actually matter for gauging capital outlays, the report was close to our forecast. We have been expecting some slowing in real spending for equipment and software in the current quarter, and it looks as though that is what we are getting. But with order backlogs still ample, corporate balance sheets flush with cash, and the cost of capital low, we continue to anticipate modest gains in equipment spending in the near term.

Meanwhile, the consumer appears to be chugging along. Our forecast for 3 percent growth in real consumer spending in the current quarter is unchanged from the October Greenbook and close to the average pace of the past few years. Steady gains in employment and income, the drop in energy prices that has occurred since the summer, and higher stock prices appear, at least to date, to have offset any restraint coming from higher borrowing costs and decelerating house prices.

We have had a few upside surprises as well. In particular, government spending, at both the federal and the state and local levels, has been somewhat stronger in the second half than anticipated in our October forecast. Also, as Steve will be discussing shortly, net exports are expected to make a slightly larger contribution to current-quarter growth of real GDP.

On net, we read these data as suggesting that growth in aggregate output in the second half of this year has been slightly weaker than in our previous projection, largely on account of the softer construction figures. That’s not easy to see in our top-line forecast of real GDP because of some serious problems with the BEA’s measurement of motor vehicle output. As you know, we simply don’t believe the BEA’s estimate that motor vehicle output added ¾ percentage point to the growth of real GDP in the third quarter, in light of the fact that vehicle assemblies fell 600,000 units at an annual rate. By our estimates, the BEA’s faulty methodology caused the growth of real GDP to be overstated about 1 percentage point in the third quarter. We expect that the unwinding of some of that glitch in the fourth quarter will trim real GDP growth about ½ percentage point. So, we believe that, on net, the published growth of real GDP will be overstated about ¼ percentage point in the second half of this year. Adjusting for the measurement problems, we estimate that real GDP probably rose at an annual rate of 1½ percent in the second half, about ¼ percentage point less than in our previous forecast and noticeably below our estimate of the growth of potential.

Our view that there has been a perceptible slowing in the pace of activity has received some independent support from our measures of industrial production. Factory output increased at an annual rate of about 5 percent over the first half of the year but seems likely to increase at roughly half that pace in the second half. The cutbacks in auto production and construction, in addition to their direct effects on aggregate output, are leaving an imprint on the production in upstream industries. Even beyond these two areas, industrial activity appears to have weakened some of late.

The recent slowing in IP has occurred amid signs of some backup of inventories. Whereas a few months ago any problems seemed to be confined largely to the motor vehicle sector, there are now more widespread signs of unwanted inventory accumulation—most notably for steel, fabricated metals, mineral products, wood, paper, and plastics. The impression left by the hard data is reinforced by purchasing managers, more of whom report that their customers’ inventories are too high than was the case a few months ago. Our forecast envisions that a relatively brief period of soft manufacturing output will be sufficient to clean up these problems. But we will need to monitor this area closely in coming months because what appears relatively benign today could turn worrisome in a hurry. For now, we are reasonably comfortable that the data on both spending and production are more consistent with aggregate activity running modestly below the pace of its potential than with a more serious slowdown.

Moreover, it would be a mistake to focus only on the downside risks because there are some prominent upside risks to our forecast as well. To my mind, the performance of the labor market continues to provide the clearest challenge to our view that the growth of activity has slipped below its potential. To be sure, last Friday’s labor market report came in very close to the projection in the December Greenbook. But the last two labor market reports taken together were stronger than we were expecting back in October. Payroll employment gains have slowed from the more rapid rate seen over the past few years, but only to a pace consistent with something close to trend growth in output—not the below-trend pace that we estimate has prevailed over the second half of this year. Moreover, the unemployment rate has declined about ¼ percentage point in recent months, an outcome more consistent with above-trend growth than with below-trend growth.

Our forecast assumes that signs of greater weakness in labor demand will become more apparent in the months immediately ahead, with increases in private payrolls slowing to about 75,000 per month in the first quarter and the unemployment rate returning to 4¾ percent. The recent modest backup in initial claims gives some support to this expectation. But slowing in labor demand is, for now, just a forecast.

We considered another possible interpretation of recent labor market developments, which is that, despite the downward adjustments that we have made to our estimates of the growth of structural productivity and potential output, we remain too optimistic in our outlook. The unemployment rate has been moving lower despite growth in real GDP that we estimate to have been below 2 percent. We have been surprised again by the weakness in labor productivity. We have not bought into this interpretation largely because the tensions between the labor market signals and GDP are relatively recent and are not especially large. So we are inclined to gather a bit more evidence before making any further adjustments to the supply side of our forecast. But the recent readings do point to a bit more downside risk than upside risk to our estimates of structural labor productivity and potential output.

Moving beyond near-term developments, we continue to expect that the period of below-trend growth will extend through the middle of next year. As in our previous forecast, the weakness in activity is led by large ongoing declines in residential investment. Moreover, we are expecting the deceleration in home prices to weigh on the growth of consumption next year through the typical wealth channel. With final sales and output slowing, the usual accelerator effects put some brakes on outlays for consumer durables and business investment spending. Those influences are reinforced in this forecast by a modest backing up of long-term interest rates, as financial market participants come to realize that monetary policy will not be eased on the timetable that they currently envision.

We see forces at work that, by the middle of next year, should result in a gradual reacceleration of activity back to a pace in line with the growth of the economy’s potential. Importantly, we are expecting some lessening of the contraction in residential investment. Housing starts have now fallen by enough that, if home sales stabilize at something around their recent pace—and I recognize that this is a big if— homebuilders will be able to make substantial headway in clearing the backlog of unsold homes. As they do, we expect construction activity to level off in the second half of 2007 and then to stage a mild upturn in 2008. Another factor working in the direction of some acceleration in activity is a diminishing drag on spending and activity from the earlier run-up in oil prices. By our estimates, the rise in oil prices has held down growth in real GDP by about ¾ percentage point this year but should be a roughly neutral factor for growth in 2007 and 2008.

On balance, our forecast is identical to that in the October Greenbook, with the growth of real GDP projected to be 2¼ percent in 2007 and 2½ percent in 2008. There were, however, a few modest offsetting influences. A stronger stock market and a lower foreign exchange value of the dollar would, all else being equal, have resulted in a somewhat stronger projection for real activity. But those effects were counterbalanced by the substantial downward revisions that the BEA made to its estimates of labor compensation in the second and third quarters. As you know, we had been expecting about half of the first-quarter surge in labor compensation to be reversed in subsequent quarters. But in the event, it was completely reversed, leaving the level of real income about $60 billion below our previous forecast. In response, we lowered our consumption projection, just as we had raised it earlier when income had been revised up. On balance, the effects of the lower income offset the influences of a stronger stock market and lower dollar, and our GDP projection was left unchanged.

Like our forecast for real activity, our forecast for inflation also has changed little over the past seven weeks. As we had anticipated, this autumn’s drop in consumer energy prices has resulted in outright declines in headline consumer prices. Core consumer prices came in close to expectations as well—though the core CPI was bit below our forecast and the core PCE a bit above. I wouldn’t make much of either surprise. Our miss on the CPI was concentrated in apparel, used cars, and lodging away from home—all components characterized by low signal-to-noise ratios. Core PCE came in only a couple of basis points above our forecast, with the surprise here mostly in the nonmarket component of medical care costs, specifically the BEA’s estimate of Medicare hospital reimbursement rates—all in all, pretty small potatoes.

Our longer-term outlook for inflation also remains unchanged. The slightly tighter labor market incorporated in this projection, along with the lower dollar and higher attendant import prices, would have led us to mark up a bit our inflation projection. But the downward revision to labor compensation implies smaller gains in labor costs and a noticeably higher level of the price markup, suggesting a little less prospective upward pressure on prices. As in our past forecasts, we expect a gradual slowing in core consumer prices over the next two years as the pass-through of higher prices for energy and other commodities runs its course and as the current tightness in labor and product markets diminishes and a small gap in resource utilization eventually opens up. Both the CPI and PCE measures of core prices are currently running a bit below the pace of this past spring—by enough to encourage us in our view that core inflation is more likely to fall than to rise over the next two years but not by nearly enough to cinch the case for our position.

I have so much more to say, but I’d better stop here, or you will think that we’ve begun our honeymoon together in North Dakota. Steve Kamin will continue our presentation.

Keyboard shortcuts

j previous speech k next speech