This essay is available here as an Adobe Acrobat PDF.

September 2005

Among the more remarkable features of the U.S. economy over the past five years--through a tech-stock collapse (from which we have still not recovered), the 9/11 disaster, and numerous chastening corporate scandals--has been the extraordinary resilience of American consumers. To paraphrase H. L. Mencken, no one has ever gone broke (at least not recently) by overestimating the willingness of Americans to spend money.

Lower Saving, Higher Spending

The unflappability of the American consumer over the past year is especially impressive. As higher energy prices have reduced discretionary income by well over $100 billion (the equivalent of adding a 1 percent national sales tax) and the Federal Reserve has boosted short-term interest rates by enough to nearly double the prime-lending rate, consumption has continued to grow. Second quarter real (that is, inflation-adjusted) consumption this year was 3.9 percent above its level a year earlier, well above the five-year average growth rate of 2.4 percent and comfortably above the ten-year average of 3.3 percent.
 
Achievement of above-average consumption growth during a time when employment and income growth were tepid by the standards of most economic recoveries required that Americans drive their savings rate to zero by June of this year--down from the already low levels of 2.1 percent in 2003 and 1.7 percent in 2004. (The ten-year average savings rate is a paltry 2.7 percent, although it averaged about 7 percent during the decade ending in the early 1990s. Savings rates in Germany and Japan are much higher--about 10 percent in Germany and 8 percent in Japan, with the Japanese rate down from 12 to 14 percent during most of the 1980s and 1990s.)
 
A drop in the U.S. savings rate by 1.7 percentage points, while seemingly small, actually represents a substantial contribution to consumption growth of about $175 billion or 2 percentage points. More broadly, about one-third of the 3.9 percent growth rate in gross domestic product (GDP) over the past year was accounted for by the elimination of what was already low U.S. saving. (Two percentage points of consumption growth contribute about 1.33 percentage points to aggregate growth.) It was as if Americans, faced with increasing uncertainty about higher interest rates and higher energy prices, decided to eliminate the tiny share of income that they had still been devoting to providing for the proverbial “rainy day.”
 
I hasten to add two comments. First, I am well aware of the imperfections of the measure of how savings rates are derived and, indeed, will suggest below that such imperfections, including the omission of wealth gains on the housing stock and the savings component of expenditures on long-lasting durable goods account for some of the reductions in the measured rate of U.S. saving. For now, suffice it to say that America’s imperfectly measured savings rate has been dropping steadily for years and has reached zero.
 
Second, I will neither decry nor praise America’s low and falling savings rate beyond noting that, in a world of substantial excess capacity, a large nation that boosts global demand growth by saving less and spending more provides welcome relief for producers, along with the gratification that steady consumption growth provides to its consumers. Asian exporters, China the most prominent among them, have acquired hundreds of billions of dollars’ worth of U.S. government securities while intervening to avoid appreciation of their currencies, a development that would, in turn, curb access to American consumers at low prices. Certainly these exporters have recognized the serendipity between America’s big spenders and their ability to produce more goods at attractive prices.
 
More U.S. Saving Would Be Problematic

The American consumer has been the primary source of demand growth, a spending hero, in a global economy plagued with excess capacity. In the process, foreign lenders have acquired claims on the United States worth about $2 trillion. That is a large number, but America is a wealthy nation with total household assets of about $60 trillion. Owing just over 3 percent of those substantial assets to foreigners is unusual for an advanced industrial country, but not unmanageable.
 
Had American net liabilities to the rest of the world not increased--alternatively, had not foreigners lent heavily to Americans to finance a current account deficit amounting to 6 percent of GDP--world growth would have been substantially lower. Were the United States to move toward elimination of its current account deficit by boosting its savings rate modestly to the 2004 rate of 1.7 percent, U.S. growth would slow by 1.3 percentage points, and global exports to the U.S. would drop sharply.

If the rise in the U.S. savings rate were accompanied by a weaker dollar, foreign exporters would face a shrinking share of a shrinking market for exports to the United States. No wonder export-oriented Asian nations, not to mention oil-exporting nations, have been willing and eager to support U.S. consumption with massive lending to the United States. The relationship between Asia’s exporters and oil producers as lenders, and American consumers as borrowers, is a mutually beneficial one that neither side seems eager to terminate. What is remarkable is the low interest rates at which this global lending and borrowing market clears. Either America is seen as a low-risk borrower or lenders have few attractive alternative uses of their funds. Both are true in today’s world of excess capacity to produce traded goods such as automobiles, other durables, and clothing.
 
Why Is U.S. Saving So Low?

While the drop in U.S. saving that has helped to sustain consumption growth has been fortunate for global growth and has been much discussed, less has been said about why U.S. consumption growth has actually accelerated over the past year in the face both of steady interest-rate increases by the Federal Reserve and of a rising energy “tax.” Part of the explanation lies in the substantial, $1.5 trillion boost in net housing wealth which has occurred over the past year, but other forces are also at work. It is important to understand them, if for no other reason than to help anticipate what may be expected in the savings rate going forward. That rate will play a large role in determining the pace of U.S. and global economic growth.
 
The acceleration of U.S. demand growth in the past six months has been largely due to stronger consumption growth. The consumption growth rate has held between 3.5 and 4.3 percent (above its longer-term trend) since the first quarter of 2004. As already noted, consumption growth accelerated to a 3.9 percent year-over-year rate during the second quarter of 2005.

It may be that the Federal Reserve has, in a counterintuitive way, kept the U.S. demand boom running longer by means of its predictable, yet very gradual, tightening strategy. That strategy has caused households to hasten discretionary spending on durables, even if it requires a savings drawdown. The promise that interest rates will be higher next year has led households to defer saving, especially given price cuts on consumer durables such as automobiles.

The purchase of durables carries a saving component, since they provide a stream of consumption services over time. Instead of putting funds into certificates of deposit or other “safe” assets that have only been paying 2 or 3 percent over the past year, households have elected to “save” by purchasing houses and cars, which yield a stream of services over time. Why spend less and save more now when interest rates are at historically low levels and when the Fed is promising that interest rates next year will be higher, probably over 4 percent and possibly 5 percent? Beyond that, housing carries with it the additional attraction of tax-free capital gains.
 
Faster Spending Growth on Cheaper Durable Goods

Since 2004, the growth of consumption has been concentrated more on goods than on services. On average, over the five years ending in 2004, year-over-year spending on goods and services was equal at 2.5 percent. During 2004, the year-over-year growth of real spending on goods soared to 7.3 percent while growth of spending on services remained close to the average of 2.7 percent. Households have substituted goods accumulation for saving because interest rates have been so low and goods prices have been stable or falling.
 
The growth of durable goods spending has exceeded the growth of other categories of goods consumption during the accelerating phase of this recovery. During 2003 and 2004, real durables spending rose at an average annual rate of 7.2 percent while services spending growth averaged just 2.8 percent. Durable spending slowed to an annual growth rate of 4.7 percent in the first quarter of 2005 only to rebound sharply to a 6.8 percent growth rate in the second quarter. Meanwhile, services spending growth remained at about 3 percent.
 
Lower prices along with low interest rates played a key role in boosting sales of durables. Services prices have risen steadily at about 3 percent over the past five years. By contrast, after falling at a 2.4 percent rate in 2002, durables prices rose by only a cumulative 1.5 percent during 2003 and 2004--ending up nearly 1 percent lower than they had been in 2001. Moreover, in the last three quarters, durables deflation has resumed mildly, with prices falling at an average annual rate of 0.4 percent. Today durables prices are, on average, a little more than 1 percent below their 2001 level. Meanwhile, nondurable goods inflation, boosted substantially by rising energy costs, has averaged close to the level of services inflation--about 3 percent.
 
The Energy Cost Trap

More spending on cheaper durables raises the possibility that the Fed’s slow tightening strategy may, in one sense, have a destabilizing effect. With heavy spending on transportation equipment concentrated on fuel-inefficient, incentive-laden light trucks, American households have accumulated an unusually large, fuel-inefficient stock of transportation capital. They have also purchased additional living space that must be heated and cooled with more expensive energy. After inducing aggressive purchases of energy-consuming durables (housing and cars) by virtue of low interest rates, the Fed has made slow, gradual increases in interest rates, which were complemented by falling prices of durable goods. With assistance from dollar recycling by Asian countries and OPEC, the Fed may consequently have unwittingly contributed to the rise in energy prices. Rising house prices have excited more purchases of real estate for investment purposes, but those who have purchased prospective rental properties now face lower-than-expected rents because of excess supply and higher-than-expected carrying costs resulting from rising energy prices.

Sometime during the next year the stock of durable goods and housing will probably be seen as too large (above desired levels) as energy costs and interest rates rise. Purchases will drop sharply, and the wealth creation that has accompanied rising real estate prices will be reversed unless the Fed ceases tightening and reverses the rise in interest rates. Such a Fed reversal is an unlikely event, at least in the early stages of a housing slowdown. In fact, Fed tightening sometimes overshoots the mark, and that raises the risk that a housing slowdown could turn into a housing collapse. Furthermore, sharply rising energy costs could exacerbate the slowdown.
 
There are limits to the growth of American consumption in excess of income growth, another way of describing a falling rate of saving. Given that the conventionally measured saving rate has reached zero, prolonged dissaving--while possible--seems unlikely in view of the rising uncertainties associated with the continued rises in energy prices and short-term interest rates.

Ending the Party

Given the potentially destabilizing pattern of increased spending on energy-intensive automobiles and housing, the Fed may decide to end the party a little earlier. The method would be a “surprise” in the form of a 50-basis-point increase in the fed funds rate sometime this fall in lieu of the anticipated 25-basis-point increase. While such a move would break the pattern of measured and gradual rate increases by the Fed in place since June 2004, it might actually be a gift from Alan Greenspan to his successor that would keep the consumption party from running on too long. Absent a larger Fed rate boost this fall, it may take an even bigger surprise in 2006 to slow down the remarkably resilient American consumer. That would constitute a difficult beginning to the new Fed chairman’s term.

John H. Makin is a visiting scholar at AEI.