President, Federal Reserve Bank of Richmond
Danville Chamber of Commerce
It’s a pleasure to speak on the economic outlook today, because after many months I am finally able to say with a reasonable degree of confidence that the outlook is improving. We’ve endured the worst downturn that the economy has experienced in most of our lifetimes, and conditions remain distressed in many industries and localities. Yet the economy appears to have leveled out and I believe we can look forward to better times ahead.
Before discussing this outlook in more detail, I’d like to provide my perspective on the developments that led up to the recent recession. Please note that these are my own views and should not be attributed to any other person in the Federal Reserve System.
Undoubtedly, the boom/bust cycle in housing was the critical economic development of the last decade. Robust housing demand led to a boom in new construction, and the share of housing in overall economic activity increased by about 50 percent from 1995 to 2005. Why was demand so strong? Part of the answer includes healthy fundamentals, such as strong gains in real income and employment and low long-term interest rates. Also important was a public policy commitment to boost housing activity and home ownership, which was reflected in the favorable tax treatment of home ownership and the implicit subsidies to housing generated by government-sponsored enterprises. In addition, innovation in mortgage lending and securitization helped to bring home ownership to a wider segment of the population. This expansion in access to credit has benefitted many Americans, but in retrospect we can see that it was accompanied by lax underwriting standards and overly complex and opaque securitization. Moreover, a wide array of borrowers, lenders, investors and analysts acted as if housing prices were certain to appreciate over both the short and the long run.
This boom in construction was accompanied by a dramatic price increase. One popular measure of existing home prices increased 192 percent from 1995 to 2006, but by the middle of this decade, evidence began to signal that the boom had gone too far. The number of vacant homes rose to new record highs and many measures of housing activity declined precipitously after recording their peak values. The number of new housing starts, for example, fell by 79 percent from January 2006 to April 2009, and prices fell by about a third. Falling home prices have reduced home equity values and household wealth. They also have led to rising delinquencies, defaults, and foreclosures, and have therefore led to substantial reductions in the value of mortgage-backed securities on the books of many financial institutions.
In short, a substantial decline in residential investment began in early 2006. Growth in overall economic activity then slowed before the expansion officially ended in December 2007. Manufacturing began to fall in mid-2007, with pronounced weakness in building materials and autos. Income growth stagnated, and in 2007, real disposable income grew by only 0.6 percent. Consumer spending, not surprisingly, weakened as well, growing by a very modest 1.6 percent in 2007. The year 2008 began with the economy in a recession that at first seemed to be relatively moderate. Through August of 2008, for example, payroll employment fell by an average of 137,000 jobs per month. The recession then intensified, and in the last 11 months, employment has fallen by an average of more than a half million jobs per month. I could cite many other dismal statistics, but it is clear that the decline in economic activity intensified last fall and has been large and widespread since then. The result was the worst recession since the 1930s.
That’s the background. Currently, the latest data indicate that the economy is leveling out. Starting with housing, several indicators of sales and construction activity hit low points in the first quarter and have registered modest advances since then. For instance, single-family housing starts have risen by 34 percent since January and new home sales have increased by 17 percent. And there are signs that housing prices are also bottoming out. Even with these welcome gains, however, housing activity remains well below a pace that would accommodate population and income growth on a sustained basis. That’s to be expected as we work off the overhang of unsold homes in many parts of the country. But at least housing is no longer a major drag on GDP growth, and that’s a welcome improvement in comparison to the last three years.
Consumer purchases of cars and trucks also began to tail off in 2007 and then fell very sharply in 2008. Sales hit a low point this past February and then increased very gradually before the “Cash for Clunkers” program gave a jumpstart to sales in July. While that program has undoubtedly pulled forward some sales that would otherwise have occurred later in the year, I believe that even without the program we would have seen some additional firming of auto sales this year. The level of motor vehicle sales not only remains well below the pace required to keep up with population and income growth in the long run, but also appears to be below the rate necessary to maintain the existing stock of vehicles. Importantly, as with housing, autos are no longer a drag on GDP growth, a welcome contrast to the last two years.
Aside from housing and cars, we also have seen a considerable improvement in the foreign trade picture. The trade deficit was above $60 billion per month for much of last year, but has fallen below $30 billion for the last five months. While the recession has hit almost all countries simultaneously, the current pickup in growth among some of our major trading partners is likely to boost the demand for American exports in coming months, and would provide a welcome lift to domestic producers.
I’ve identified three areas on the demand side of the economy where we have some grounds for cautious optimism, namely housing, autos and exports. We also have seen some promising news on the supply side – namely, in the cyclically sensitive manufacturing sector. There, output fell by 17 percent from the cyclical peak through June, but rebounded in July by a full percentage point. Much of that increase resulted from the resumption of production by GM and Chrysler, but there is evidence that the improvement in manufacturing is more widespread. The composite index of manufacturing activity published by the Institute for Supply Management has increased substantially this year to a reading just a smidgeon below the level that would indicate an expanding manufacturing sector. Closer to home, our own Richmond Fed Fifth District Index of manufacturing activity in the Fifth District has risen sharply this year to a level consistent with solid growth in manufacturing. And just as with the national index, our index of new orders has shown a striking improvement in recent months.
We also have seen a significant improvement in financial conditions since the turmoil last fall. Corporate borrowing costs have declined considerably, as interest rates on commercial paper and corporate bonds are now much lower than they were last year. Many major banks have sold stock successfully and now have the capital to support new lending, even if conditions turn out worse than expected. Although many borrowers face tougher credit terms in a soft economy, the banking system as a whole appears capable of supporting business investment and expansion.
Academic and industry economists have taken all this into account, and most now see the second quarter as the low point for GDP in this cycle. The typical forecast calls for positive GDP growth in the current quarter and a gradual improvement beyond that. I agree with this outlook. Indeed, since the beginning of this year I have been expecting positive growth before year-end – but I must emphasize that the recovery is likely to be slow and uneven for some time. We obviously have major difficulties to overcome before we can feel really good about the economy.
I won’t be encyclopedic, but the labor market is a prime concern. We lost 247,000 jobs last month, and until employment is back on a solid growth path, income prospects will be suspect. In addition, average wage growth has slowed significantly; last year, average hourly earnings grew 3.9 percent, but so far this year they have only increased at a 1.5 percent annual rate. Thus household incomes are facing a double whammy from falling employment and sluggish wage growth, and weak incomes will likely put a damper on consumer spending growth. As overall activity regains momentum, the labor market should gradually improve and lead to a better outlook for incomes and spending. But right now that’s just a forecast, and we will clearly need to monitor the situation carefully.
The outlook would not be complete if I did not mention monetary policy and the outlook for inflation. Last year the FOMC lowered its target for the federal funds rate to essentially zero. Moreover, the Fed has aggressively expanded its balance sheet; its size has more than doubled in less than a year. The magnitude of these actions is unprecedented in the Fed’s history and should help underpin the nascent recovery.
Regarding the outlook for inflation, economic forecasters are divided. One school of thought expects inflation to steadily fall for a considerable period in view of the fact that economic slack (as indicated by the high unemployment rate, for example) is high and likely to decline only slowly. For the record, the core price index for personal consumption expenditures increased 1.5 percent in the last 12 months. That’s right on target, as far as I’m concerned, and so a large further decline would be unwelcome.
Other economists, though, point out that the notion that economic slack invariably depresses inflation, which is based on a correlation called the “Phillips Curve,” has often led forecasters astray. For example, many forecasters who relied on a Phillips Curve expected a greater decline in core inflation than we saw last year. That approach greatly underestimated inflation for much of the 1970s, as well as during the rise of inflation in 2003-2005. The situation in late 2003 is especially instructive, in my view. Then, we were emerging from a recession and the unemployment rate was stubbornly high. Yet core inflation did not fall or even remain unchanged; instead, it rose from 1.5 percent in 2003 to 2.2 percent in 2004 and remained elevated, above 2 percent for several more years. That episode illustrates well the pitfalls of overestimating the influence of economic slack on inflation.
If economic slack does not drive inflation what does? Some economists stress the importance of expectations in determining price setting behavior as an alternative source of inflation. According to this view, if expectations are firmly anchored, then the behavior of individual buyers and sellers will be consistent with the expected outcome and inflation will remain fairly steady. I have a lot of sympathy for this view, because it also matches up well with modern macroeconomics. Unfortunately, we have less than ideal measures of the expectations that form the basis of individual behavior. What we do have suggests that inflation is not likely to decline significantly from here. First, there is evidence from monthly surveys that puts the expected long-term growth in the Consumer Price Index at about 3 percent, near the center of where it has tracked over the last decade or so, a period in which inflation has averaged over 2 percent. Second, we can get an implicit measure of longer-run inflation expectations from the prices of certain financial security, and those readings imply that inflation is expected to be higher a few years down the road. While neither type of measure is without its flaws, both seem to suggest that inflation is more likely to rise than to fall.
This evidence does not illuminate why people might expect inflation to be higher. Market commentary, however, suggests that uncertainty over the likely course of monetary policy might be important. I mentioned that the Fed’s easing was unprecedented in magnitude. Market participants have at times expressed some doubts that the Fed will be willing and able to reverse course promptly enough to keep inflation in check. From a technical point of view, I do not see a problem – we do have the tools to contract our balance sheet and remove monetary stimulus when we need to do so, as Chairman Bernanke explained in detail in last month’s Monetary Policy Report to Congress. The harder problem is choosing when and how rapidly to remove stimulus as the recovery begins. I am certainly aware of the danger of aborting a weak, uneven recovery if we tighten too soon. But there can be a strong temptation to hesitate when emerging from a recession, awaiting conclusive signs of robust growth. Keeping inflation well-contained may require action before a vigorous recovery has had time to establish itself.
Judging when to withdraw monetary stimulus by raising our policy interest rate is hard enough. But assessing the degree of stimulus provided by our expanded balance sheet poses special challenges. The rapid growth in the Fed’s assets and liabilities occurred during the exceptionally turbulent financial market conditions of last fall. In response to heightened creditor anxiety, many banks built up large buffers of highly liquid assets to protect themselves against a possible loss of investor confidence, some of which they held in reserve account balances at the Federal Reserve Banks. The increase in banks’ demand for reserve balances last fall was met in part by the reserves supplied via the Fed credit programs supporting the commercial paper market and money market mutual funds. The rest of the increase was met through banks’ borrowings from the Fed, either through the standard discount window facility, the Term Auction Facility or by borrowing dollars from foreign central banks. Usage of the commercial paper and money market programs has dropped off since last fall, but in the meantime the Federal Open Market Committee instituted a large-scale asset purchase program, which authorized purchases of up to $300 billion in U.S. Treasury securities, up to $200 billion in agency debt, and up to $1.25 trillion in agency mortgage-backed securities.
In a statement after the last FOMC meeting, the Committee announced that it would gradually slow the pace of Treasury purchases and complete them by the end of October. Purchases of agency debt and agency mortgage-backed securities continue, however, and those purchases supply reserves which reduce the amount that banks need to borrow from the Fed to satisfy their elevated demand for reserve account balances. Should those purchases continue at their current pace, there will come a point at which the banking system will no longer need to borrow to obtain the desired level of reserve balances. At that point further asset purchases would then push the supply of reserve balances beyond demand, and would necessitate a downward adjustment in other yields to induce banks to voluntarily hold large balances. This would provide discretely more monetary stimulus than past asset purchases have provided thus far, since arguably such purchases have until now simply displaced bank borrowing from the Fed. With the economy leveling out and beginning to grow again later this year, and with bank reserve demand ebbing as financial conditions improve, I will be evaluating carefully whether we need or want the additional stimulus that purchasing the full amount authorized under our agency mortgage-backed securities purchase program would provide.