President, Federal Reserve Bank of Richmond
Remarks to the West Virginia Bankers Association and the Community Bankers of West Virginia
It's a pleasure to be with you. While the economy is seldom far from many people's minds, I think it's fair to say that economic conditions are garnering a bit more attention than usual right now. Housing markets have deteriorated over the last two years and the resulting losses on mortgage-related securities contributed to financial market turmoil last summer. The associated decline in employment in the construction and financial industries has contributed to a slowdown in aggregate job growth. Moreover, inflation — both overall and excluding food and energy prices — has picked up of late. As you might imagine, these developments have kept us busy at the Federal Reserve. So today, I'd like to spend some time talking about the economy. I'll begin by reviewing current economic conditions, and then go on to discuss the outlook for the coming year. Before we begin though, let me note that the usual disclaimer applies: the views I express are my own and are not necessarily shared by any of my colleagues on the Federal Open Market Committee.1Clearly, economic activity has been softening. At the end of last year, real GDP grew at a meager 0.6 percent annual rate, and most forecasters are not looking for much better growth — if any — in the current quarter. Much of this sluggishness has been due to a severe housing market downturn, along with the attendant financial market fallout. After a 10-year expansion, residential investment peaked in late 2005. Since then, construction and sales have fallen fairly sharply, first in large metropolitan areas that had seen the strongest booms, and then spreading to other markets where housing price increases were less pronounced. Despite the falloff in construction, inventories of unsold homes rose sharply. While inventory levels have actually retreated somewhat in recent months, they have not come down as rapidly as sales, and they are currently a depressing influence on home prices and new construction.
Home prices increased significantly during the long boom, particularly in local markets with restricted supply. Existing home prices increased about 90 percent between 1995 and 2005 for the nation as a whole. In the Washington, D.C., market, prices increased 148 percent from 1995 to 2005 and rose another 11 percent in 2006. And in Charleston, W.Va., prices climbed by 39 percent over the same 10-year period and increased by another 2 percent in 2006. Of course, rapid increases in real quality-adjusted prices are not indefinitely sustainable for any asset, and in the case of housing, potential buyers eventually get priced out of the market. In many markets, prices changed course quickly, but in others, prices have continued to increase. Average prices for the nation as a whole fell in the third quarter of 2007 by 0.4 percent, which is the first national price decline since 1994. And in formerly hot markets, the declines have been larger, with prices falling over 5 percent in San Diego, for example. Prices have also fallen significantly in areas with weak regional economies, like Michigan and northern Ohio. Charleston has avoided an outright decline in prices, although appreciation has remained modest, with third-quarter growth at a 2 percent annual rate.
Developments in housing finance arguably have played a substantial role in the evolving conditions in housing markets. Here the long-term story is the technology-driven wave of innovation in retail credit delivery that enhanced the ability of lenders to assess the creditworthiness of individual borrowers. This increased the pool of qualified borrowers and the range of feasible financial products, and dramatically expanded access to mortgage credit over the last decade, just as it expanded access to unsecured consumer credit earlier on. Technology also has contributed to innovation in securitization and other forms of intermediation of credit flows. This spreads risks more widely and leads to lower borrowing costs. As with any new product or service innovation, however, some experimentation and risk was involved, and in this case, some of those risks were realized. Looking back, there undoubtedly are many loans that both the borrower and the lender wish had not been made. But it is important to keep in mind that many borrowers remain better off as a result of recent lending innovations.
Future research may quantify the extent to which credit market innovations contributed to the boom in housing market activity by expanding the pool of potential homeowners. In any event, when the growth in housing demand came to an end, home prices peaked and began falling in many markets. In hindsight, it seems clear that the success of new methods of lending to riskier borrowers was to some extent dependent on sustained home price appreciation. This provided strained borrowers with the ability to refinance, thus masking the effects of more inclusive underwriting. It takes some time, however, for the ultimate loss experience of a mortgage portfolio to become evident. While observers were raising concerns early on — the late Federal Reserve Governor Ned Gramlich, for example 2 — it wasn't until last year, after home prices had peaked in some major markets, that more quantitative evidence began to emerge regarding the substantial extent to which mortgage loans made in 2006 and later would underperform previous vintages. The ensuing adjustment in underwriting standards has further contributed to the decline in housing activity.
The story behind last year's unfolding drama in credit markets was the continuing re-assessment of the fundamental value of nonprime mortgages in light of incoming data implying significantly higher ultimate losses on recent vintages of subprime mortgages. Securitization was an important part of the expansion of credit in recent years, and securities backed by pools of sub-prime or other nontraditional mortgages served as the backing for other obligations, usually issued off the balance sheets of the sponsoring institutions. As housing slowed over the summer, it became clear that some mortgage-related securities previously judged as relatively safe would suffer substantial losses. Many of these securities were the liabilities of entities with explicit or implicit bank lending guarantees. Many banks that provided such guarantees have had to either meet large funding demands or bring the impaired assets onto their balance sheets. Uncertainty about the scale of such adjustments has at times meant higher funding and capital costs, although risk premia increased far more for some institutions than for others. Since then, such institutions have taken large write-offs, and many have replenished their capital. Many affected banks have dramatically increased their advances from the Federal Home Loan Banks, where lending increased by 29 percent or about $180 billion in the third quarter.
Credit markets were hit particularly hard in August, as many participants found it difficult to refinance the asset-backed commercial paper they had issued. Banks began holding larger precautionary reserve balances then, putting upward pressure on interbank lending rates. The New York Fed injected significantly more reserves than usual via open market operations in order to relieve the pressure and keep the overnight federal funds rate near the target. In addition, the Board of Governors accepted Reserve Bank requests to lower the discount rate by a half a percentage point, reducing the spread above the federal funds rate from the traditional 100 basis points down to 50 basis points. With concern mounting that housing investment was declining more rapidly than had been expected and that the growth outlook was deteriorating as a consequence, the FOMC reduced the federal funds target rate in September and October, bringing it down 75 basis points to 4.5 percent. Financial market conditions showed some improvement in September and October, but turned problematic again in late November, a month that also saw a further deterioration in the real outlook, as measures of housing market activity continued to come in below expectations.
In December, wholesale funding markets increasingly showed the effects of heightened uncertainty surrounding financial institutions' adjustment requirements. Term funding spreads relative to expected overnight rates became quite elevated for some banks, differentiation in rates across institutions became more pronounced, and the volume of term funding contracted. Increases in interbank interest rates associated with year-end, balance-sheet considerations have occurred in the past, but market participants appeared to expect low overnight rates over the year-end this time. Rates at the Federal Home Loan Banks are closer to the expected overnight funds rate than to term LIBOR, which may explain the relatively small amount of discount window borrowing even since the August reduction in the discount rate spread over the target. All this suggests that term funding premia reflected assessments of counterparty risk rather than expectations that the funds rate would spike at year end.
Against this back-drop, the Federal Reserve introduced a new mechanism for providing term funding to financial institutions. The Term Auction Facility, or TAF, makes 28-day loans of a predetermined total amount at a rate set by auction. These loans are otherwise similar to discount window loans made by a bank's regional Reserve Bank against collateral posted with that Reserve Bank. Since these auctions began, near the end of December, spreads on interbank term loans have fallen significantly and have returned to where they were last November before the year-end funding difficulties emerged. It will be difficult to determine the extent to which the TAF contributed to this easing of rates in the term funding market, since the counterfactual will never be observed. An earlier instance of elevation in term spreads, peaking in early September, abated without such action by the Fed.
As one would expect, revised assessments of mortgage lending risk have resulted in a tightening of credit standards for businesses and consumers. Many lenders are requiring larger down payments, and mortgage rate spreads have increased significantly for riskier borrowers and riskier products. Mortgage rates have come down since December, however — the rate on conventional 30-year fixed-rate mortgages has fallen about 45 basis points. And even though the spread between jumbo and conforming mortgages has widened a bit, jumbo rates have also eased in recent weeks, coming down about 20 basis points. Spreads on investment-grade corporate bonds have widened over the last month, but still, the level of yields on such debt has fallen. On the other hand, interest rates on high-yield debt and commercial mortgage-backed securities moved up in the last half of 2007, and have increased further since the beginning of the year. The strong differentiation in the response of lending spreads across borrower classes suggests that increasing spreads have been driven mainly by changing risk assessments rather than bank funding pressures per se. Higher risk spreads and generally tighter lending terms will tend to restrict spending in the near term. But the fall in short- and long-term Treasury rates over the last few months has offset the upward movement in higher spreads for a wide range of borrowers. The net effect has been lower rates for all but the highest-risk borrowers. In fact, lower reference rates have meant that more adjustable-rate mortgage borrowers will see their interest rate go down rather than up.
The economic outlook for 2008 has worsened in response to the developments of the last six months, and the recent flow of data has heightened the downside risks. The housing sector has been and will continue to be affected by the tightening we've seen in lending standards. New home sales have fallen 64 percent from their peak in October, 2005. Home construction is unlikely to bottom out this year, and I expect housing investment to continue to be a drag on growth through at least year-end.
Business investment has contributed positively to growth over the last year, but I expect it to grow less robustly than in 2007, since some firms are experiencing a higher cost of capital and most firms face an uncertain demand for their products. A particularly dramatic change is likely to occur in commercial construction, which is a key segment of business investment. Construction spending for new stores and offices grew by a healthy 10 percent after inflation last year, but we have heard reports from our District contacts of a significant softening of conditions lately, with major projects being deferred or cancelled outright. In addition, vacancy rates for retail space have increased over the last year, which should lead to less construction going forward. The most recent investment data we have are for December, and those reports indicate continued growth in construction activity and new orders for business equipment.
Exports are likely to remain a source of strength next year, however, as a weaker dollar and continuing growth abroad support demand for U.S. goods and services. Accordingly, I expect the trade deficit to continue to narrow, providing modest support to real GDP growth. On the other hand, we are hearing reports of unexpectedly low tax revenues in the state government sector, which will likely mean some pruning of expenditures in coming months.
The main story in the forecast, though, remains household spending, which accounts for 70 percent of GDP. Consumer spending held up well until the end of last year, having grown at a solid 2.2 percent rate in real terms during the three months ending in November. In December, however, real spending was flat. Higher energy prices and falling home prices are cited often as factors that could dampen consumer spending, and these are legitimate concerns.
In addition, we will see more moderate growth in household income in the year ahead due to a weaker labor market. Job growth slowed over the course of 2007, and in January employment was reported to have fallen by 17,000 jobs. The unemployment rate has risen by a half percentage point since March and now stands at 4.9 percent. More industries now show declines, rather than increases, in employment. Fewer small businesses plan to increase hiring. And in our own surveys of economic activity in the Fifth District, we are hearing that an increasing number of firms have cut back on hiring plans recently. Other indicators are flashing less discouraging signals, however. Layoff announcements have continued to fall through December, and the U.S. Department of Labor's measure of job openings has remained at a relatively high level for over a year. My own expectation is that job growth will be lethargic, at best, for much of this year.
Putting it all together, we have obviously experienced a significant decline in the growth of overall economic activity since August, with much of the decline occurring in the last two months. My sense is that we will see sluggish growth for at least half a year before a gradual firming begins. A legitimate question is whether conditions will weaken further — in other words, whether the economy will enter a recession. There are two keys to answering that question. The first is business investment; as I mentioned, there are some early signs are that investment is slowing, but the most recent monthly indicators still suggest some forward momentum. The other key is the jobs market. There is a fair amount of month-to-month volatility in the employment numbers, so it is quite possible that the underlying trend is stronger than the January reading by itself would suggest. If job growth is positive in the months ahead, and if wages can stay ahead of inflation, then income growth should be sufficient to support consumer spending gains and allow us to skirt the boundary of recession.
As I said, my sense is that the most likely path is sluggish growth in the near term. But I can also see the possibility of a mild recession, similar to the last two we have experienced — in other words, shallow and with a slow recovery. What I don't expect is a more severe recession, like those we saw in 1982 or 1974. Keep in mind that monetary policy has moved aggressively in recent months, and that inflation-adjusted interest rates are now very low by historical standards. That by itself won't solve all our problems, but it will help support activity enough to at least avoid the worst outcomes, and possibly avoid a recession altogether.
I would emphasize that this outlook does not incorporate an overly sanguine view of either the housing market or financial markets. The swollen inventories of unsold homes that we see in most major markets is a clear reason to project further weakness in new home construction. Until home inventories fall to sustainable levels, I would expect further declines in home prices and soft demand, and so my overall growth outlook incorporates a continued drag from housing this year. Declining home prices will further increase the number of borrowers with negative equity in their home. Since this is a key driver of mortgage defaults and foreclosures, especially for less creditworthy borrowers, I expect continued increases in subprime loss rates.
Sound valuations of mortgage-backed securities already account for higher ultimate loss rates as various mortgage vintages age. Those are just projections at this point, and actual experience could come in either better or worse than expected. Unexpected reductions in the values of mortgage-related securities could spark new episodes of financial market turmoil. But I believe that financial market participants will find ways to work through problems as the year progresses. Financial intermediaries will continue to re-adjust balance sheets and replenish capital as needed, and will strengthen risk management practices as they take on board the lessons of the last year. Investors will continue to reallocate portfolios and their heightened desire for transparency will help shape the next generation of financial innovations.
Risks are not limited to the outlook for real economic growth. Inflation has stepped up recently. As measured by the 12-month change in the PCE price index, inflation was 3.5 percent in June 2006. That measure of inflation fell to 1.8 percent in August 2007. Similarly, core inflation, which omits volatile food and energy prices, was 2.5 percent in August 2006, and then declined to 1.8 percent in August 2007. Those declines were heartening, and when the financial market turmoil intensified in August, the improving inflation picture allowed even an inflation hawk to endorse an easier monetary policy stance. Since then, however, the inflation picture has deteriorated. From August through December, the overall PCE price index rose at a 4.3 percent annual rate, and the core index rose at a 2.8 percent rate. These numbers do display transitory swings, so I wouldn't extrapolate them forward indefinitely. Still, I have to say that I am uncomfortable with the inflation picture, and disappointed that the improvement we saw earlier this year was not more lasting.
I am also troubled by the lengthy divergence we've seen between overall and core inflation. Some of you may recall that core inflation was devised in the 1970s to filter out some of the more volatile consumer prices and get a better read on inflation trends. For several decades, core inflation seemed to work well due to the fact that food and energy prices had no clear trend relative to the overall price level. In the last few years, though, overall inflation has been persistently above core inflation, and few observers expect oil prices to go back below $20 per barrel. Because the job of a central banker is to protect the purchasing power of currency, it is overall inflation that we need to keep down, not just core inflation. Going forward, markets expect oil prices to back off slightly from their current level, and I hope they have the direction right this time.
In the last few weeks, the Fed has responded to signs of weakening economic growth with further cuts in the federal funds rate, bringing the cumulative reduction to 2 ¼ percentage points. A slowing economy requires a lower inflation-adjusted interest rate because it means softer relative demand for resources now compared to the future. In my view, the prominence of downside risks means that further easing ultimately may be warranted. My expectation that growth is likely to be sluggish this year figured prominently in my thinking about policy last month, however, so if incoming data is not weaker than expected over the next several months, it's not clear further rate cuts would be warranted.
And let me end with one final thought: inflation also presents risks. Throughout the period since 2005, when inflation rose, eased off, then rose again, longer-term inflation expectations have remained fairly stable. This has been comforting, and makes it easier for me to support interest rate cuts when a weakening outlook calls for it. The longer we go experiencing only upside inflation misses, however, the more we risk losing the credibility we have fought so hard to maintain.
This is a revised and expanded version of a speech I gave to the Richmond Chapter of the Risk Management Association on January 18, 2008. I am grateful to Roy Webb and John Weinberg for help in preparing this speech.
Edward M. Gramlich, "Subprime Mortgages: America’s Latest Boom and Bust" (Washington, DC: Urban Institute Press, 2007).