The economy continues to suffer from some aftereffects of the bursting of the "bubble economy" of the late 1990s. Although consumer spending is expanding moderately, business investment remains weak, and financial markets are uncertain about the durability of the current recovery. Nevertheless, the Congressional Budget Office believes that the stage is set for stronger economic activity this year--an opinion shared by many private-sector economists, as represented by the Blue Chip consensus forecast.

Much of the boom of the late 1990s was based on persistently faster growth in productivity. However, the tremendous surges in the stock market and in investment spending that occurred at that time were partly based on expectations for corporate profits that are now understood to have been unreasonable. That "bubble" part of the boom burst in early 2000, and the following year the economy entered a relatively shallow recession (as measured by the drop in output). The economy recovered in 2002, but it was buffeted by revelations that a small number of notable corporations had engaged in accounting irregularities during the bubble years. Those revelations shook the confidence of investors, consumers, and businesses. The stock market fell sharply again, and private-sector employment declined in the second half of the year.

The strength of the economy in 2003 depends in large part on whether consumer spending will continue to provide the economy's foundation. Throughout the 2001 recession and the early recovery, the household sector has been a source of strength. Expansionary fiscal and monetary policies are partly responsible for that strength: the lowest mortgage interest rates since the 1960s have triggered a wave of refinancing and contributed to a boom in housing, zero percent financing has spurred sales of cars and light trucks, and tax cuts have bolstered disposable income. Those factors have largely offset the drag on consumer spending caused by declines in the stock market. In the future, however, they will play a smaller role in supporting spending. Thus, the growth of consumer spending will depend primarily on the growth of personal income.

The prospects for personal income in the short run are uncertain, however, because demand is anemic in many other parts of the economy. Spending by the business sector remains weak, as low corporate profits and excess capacity from overinvestment during the bubble years have inhibited investment. Uncertainty about the strength of demand and about the risks arising from terrorism and war have led businesses to be particularly cautious in hiring. In addition, state and local governments have had their spending weakened by deteriorating finances.

Nevertheless, some indicators point to a brighter outlook for the economy this year. Investors and consumers appear to have gained a bit more confidence about the economy in recent months. The stock market has tentatively moved upward since its low in October. The spread between interest rates on corporate bonds and Treasury notes narrowed slightly toward the end of 2002, suggesting that credit markets are somewhat less worried about corporate finances than they were earlier in the year. Consumer sentiment and expectations also appear to have stabilized late last year. Business spending on equipment and software, particularly on information technology, appears to have strengthened in 2002, and inventories may be reaching the point at which businesses need to restock their shelves. Finally, a drop in the exchange value of the U.S. dollar is conducive to stronger growth of exports.

CBO's economic forecast expects the recovery to continue, with real (inflation-adjusted) gross domestic product growing by 2.5 percent in calendar year 2003 and 3.6 percent in 2004 (see Table 2-1). That growth is slower than in most past recoveries but is comparable to the pace after the 1990-1991 recession (see Figure 2-1). The growth of housing investment is expected to slow substantially, while real spending for personal consumption should continue to increase by about 3 percent a year. Investment in producers' durable equipment is expected to recover, but investment in structures will remain weak for some time. In CBO's forecast, the unemployment rate is stable in 2003, averaging 5.9 percent, and then edges down only to an average rate of 5.7 percent in 2004. As the recovery achieves a firmer footing, the Federal Reserve is assumed to shift monetary policy gradually from its current accommodative stance toward a more neutral one; consequently, both short-term and long-term interest rates are expected to rise in late 2003 and during 2004. In this near-term forecast, inflation--as measured by the consumer price index for all urban consumers (CPI-U)--remains below 2.5 percent a year.

Table 2-1.
CBO's Economic Projections for Calendar Years 2003 Through 2013
		Estimated 2002		Forecast				Projected Annual Average
				2003		2004		2005-2008		2009-2013
Nominal GDP (Billions of dollars)		10,443		10,880		11,465		14,154a		18,066b
Nominal GDP (Percentage change)		3.6		4.2		5.4		5.4		5.0
Real GDP (Percentage change)		2.4		2.5		3.6		3.2		2.7
GDP Price Index (Percentage change)		1.1		1.6		1.7		2.1		2.2
Consumer Price Indexc (Percentage change)		1.6		2.3		2.2		2.5		2.5
Unemployment Rate (Percent)		5.8		5.9		5.7		5.3		5.2
Three-Month Treasury Bill Rate (Percent)		1.6		1.4		3.5		4.9		4.9
Ten-Year Treasury Note Rate (Percent)		4.6		4.4		5.2		5.8		5.8
Tax Bases (Percentage of GDP)
	Corporate book profits	6.2		6.8		7.3		9.2		8.4
	Wages and salaries	48.1		48.1		48.1		48.0		47.8
Tax Bases (Billions of dollars)
	Corporate book profits	653		739		842		1,267a		1,474b
	Wages and salaries	5,025		5,237		5,518		6,782a		8,635b
Sources: Congressional Budget Office; Department of Commerce, Bureau of Economic Analysis; Department of Labor, Bureau of Labor Statistics; Federal Reserve Board.
Notes: Percentage changes are year over year.
Year-by-year economic projections for calendar and fiscal years 2003 through 2013 appear in Appendix E.
a. Level in 2008.
b. Level in 2013.
c. The consumer price index for all urban consumers.

 

Figure 2-1.

The Economic Forecast and Projections

Sources: Congressional Budget Office; Department of Commerce, Bureau of Economic Analysis; Department of Labor, Bureau of Labor Statistics; Federal Reserve Board.

Note: All data are annual values; percentage changes are year over year.

a. The change in the consumer price index for all urban consumers, applying the current methodology to historical price data (CPI-U-RS).

CBO's forecast assumes that there will be no significant repercussions for the U.S. economy from any war with Iraq and no shocks to the economy from major acts of terrorism. However, uncertainty about war and terrorism may continue to weigh on consumers and businesses, either directly or through its impact on stock prices. The forecast assumes that such uncertainty is not fully resolved in the near term. (For a discussion of how war might affect the U.S. economy under several alternative military scenarios, see Chapter 5.)

Beyond 2004, CBO projects that growth of real GDP will average 3.2 percent a year from 2005 through 2008 and then slow to 2.7 percent a year from 2009 through 2013. That downward trend in economic growth over the next decade primarily reflects slower growth in the labor force as the oldest members of the baby-boom generation begin to retire. The unemployment rate is expected to average 5.2 percent after 2008.

CBO's baseline projections reflect current law, which includes the expiration of the tax-cutting Economic Growth and Tax Relief Reconciliation Act of 2001 at the end of 2010. Thus, in CBO's baseline, tax rates will return to their pre-2001 levels in 2011. The expiration of that law will have complicated effects on the economy, although those effects are small relative to the overall uncertainty of the economic forecast (see Box 2-1). The most noticeable impact is that the growth of real GDP is reduced in 2011 and 2012.
 

Box 2-1. The Economic Effects of Expiring Tax Cuts The Economic Growth and Tax Relief Reconciliation Act of 2001 (EGTRRA) is scheduled to expire in 2010. As a result, under current law, marginal income tax rates will rise in 2011, provisions for child credits and marriage-penalty relief will cease to apply, and estate and gift taxes will be reinstated. That expiration (often called a sunset) will also affect provisions in the tax code for pensions, individual retirement accounts, education, and miscellaneous items. (Those effects are described in detail in Chapter 3.) The sunset of the 2001 tax law will have a complicated impact on the economy. The expiration of some provisions (such as those affecting marginal tax rates) will reduce gross domestic product, whereas the sunset of other provisions (such as the child credits) will increase it. On net, CBO estimates, the expiration of EGTRRA will lower GDP by about half a percent by 2013. That estimate is very uncertain, however, and CBO may revise that figure as it continues to analyze the issue.1 The major economic effect of the sunset stems from the rise in marginal tax rates. Those rates influence people's incentives to work and save because they determine how much additional income taxpayers can keep when they decide to work an extra hour or save an extra dollar. The sunset will also decrease the proportion of total income that is subject to taxation--as marginal tax rates rise, more people may seek to shelter more of their income by taking it in nontaxable rather than taxable forms.2 CBO estimates that in 2011, the first year after EGTRRA expires, the effective marginal tax rate on labor will rise by about 1.8 percentage points, while the effective tax rate on capital will increase by 0.6 percentage points (see the table). Those changes in effective tax rates are smaller than the changes in statutory income tax rates that will occur, because some income is not taxed.       Effective Marginal Income Tax Rates, 2001-2013 (In percent)   Tax Rate on Labor Tax Rate on Capital 2001 20.7 15.5 2002 20.5 15.5 2003 20.7 15.5 2004 20.3 15.4 2005 20.3 15.4 2006 19.9 15.1 2007 20.1 15.1 2008 20.3 15.1 2009 20.5 15.1 2010 20.7 15.1 2011 22.5 15.7 2012 22.8 15.7 2013 22.9 15.7 Source: Congressional Budget Office. Note: Includes federal individual and corporate income taxes; excludes payroll taxes. In the three years between the end of 2010 and the end of CBO's current projection period, the largest economic effects of the higher tax rates are likely to involve labor supply, which may shrink by between 0.4 percent and 1.2 percent from what it would have otherwise been. National saving, by contrast, is likely to rise.3 But in a period as short as three years, changes in saving--and consequent increases in the capital stock--will probably not be large enough to offset the impact of a reduction in labor supply on the nation's productive capacity. Economic outcomes could also be affected by the extent to which people anticipate the 2011 tax increase ahead of time. Workers who know that taxes will rise in a few years may tend to adjust their work so as to concentrate their income in the years before taxes go up. For instance, people close to retirement may work overtime in the lower-tax years and then retire somewhat earlier when taxes increase. Second earners in married-couple households may choose to work and earn income when taxes are relatively low and then leave the labor force when taxes are high. Thus, anticipation of the tax increase might increase GDP before 2011. However, people have different opinions about when and whether the tax law will expire--and also have widely varying opportunities to shift their income from one year to another--so making projections about those anticipatory responses is difficult. CBO assumed that, on average, anticipation of the tax increase would boost the annual level of GDP by less than 0.05 percent between now and 2011. The economic effects of the sunset during CBO's projection period will also depend on people's expectations about what policymakers will do in later years (after 2013). Logically, there are several alternatives. CBO's budget baseline assumes that tax rates will be higher from 2011 to 2013, but because that baseline extends only through 2013, CBO is not required to make any specific assumption for subsequent years. One possibility is that the additional revenues and lower debt will allow taxes to be lower at some point after 2013 than they would be otherwise. If so, some people may choose to work less than they otherwise would when tax rates are high (such as between 2011 and 2013) but work more later when tax rates are low. Alternatively, people may assume that taxes will remain relatively high and that the additional revenues will lead to higher levels of spending. In that case, people will not change their labor supply as much as in the previous example. In any event, it is unclear when--or even if--people expect any of those changes to take place. Simulations from economic models suggest that assumptions about future policy can significantly influence the long-term impact of a tax increase. If people expect that paying more taxes now means that tax rates can be lower in the future, GDP is generally higher in the long run. But if people think higher tax rates now mean that government consumption can be higher in the future (rather than taxes lower), then GDP is likely to be lower in the long run. However, those uncertainties affect the period after 2013 much more than the years from 2011 to 2013. CBO's simulations suggest that regardless of the policy choices made after the projection period, the sunset of EGTRRA will decrease GDP in the last three years of that period, although the amount of the decrease varies according to what is assumed about future policy. CBO was unable to determine what assumption about future policy was most appropriate. Thus, in constructing its baseline, CBO simply chose to use an average from a number of different assumptions and different models of the economy. The estimated budgetary implications of those scenarios are strikingly small compared with the overall uncertainty of 10-year budget projections. (That uncertainty is detailed in Chapter 5.) The economic weakening caused by even so large a tax increase as the one that will occur when EGTRRA expires could reduce revenues by about $40 billion: $6 billion in 2011, $15 billion in 2012, and $18 billion in 2013. (The tax increase itself is expected to raise annual revenues by a total of about $600 billion over those three years). To the extent that people anticipate the tax increase and boost their taxable income in the lower-tax years before the sunset, revenues could be increased in those years. As a result, the economic repercussions of the sunset are likely to reduce revenues by less than that $40 billion over the entire 10-year period. By contrast, the difference between reasonably optimistic and pessimistic budget projections could amount to more than $6 trillion over those 10 years (see Chapter 5)--more than 100 times the difference caused by the tax increase. Clearly, even large percentage errors in calculating the economic impact of the sunset would play little role in the overall uncertainty of long-term budget projections. A sudden tax increase such as that caused by the expiration of EGTRRA after 2010 might also risk creating a short-term economic slowdown. CBO does not attempt to forecast the cyclical movement of the economy more than two years ahead, so its baseline does not contain a recession in 2010. In the case of EGTRRA, moreover, it may not be reasonable to expect that the sunset would cause much of a slowdown. To the extent that disruptions would predictably affect the unemployment rate and inflation, the Federal Reserve could anticipate and offset those disruptions. Its task might be more difficult, however, if tax policy remained unclear in the years before the sunset. 1.  The effect of taxes on the economy remains an unsettled area of economics. Some models suggest that GDP could decline by more than half a percent from the sunset of EGTRRA; other models suggest that GDP might increase. 2.  Estimates of the increase in the extent of tax sheltering are normally the responsibility of the Joint Committee on Taxation. Preliminary CBO estimates are reflected in the Box 1-2 table in Chapter 1 and in Table 3-11 in Chapter 3. 3.  National saving includes both government saving and private saving. Although private saving will probably decline because of the increase in marginal tax rates, government saving will rise (under current law) from the additional tax revenues. Simulations with several models suggest that, on net, national saving is likely to increase.

 

Recent Economic Developments

The slow recovery from the 2001 recession continues. Consumer spending is still rising--helped by moderate growth in wages and salaries, the contribution of lower income tax rates to disposable income, and proceeds from the refinancing of home mortgages, but hindered by a decline in stock market wealth. The housing market, fueled by low interest rates, has been a consistent source of strength. Investment in business equipment has begun to revive, as some of the excess capacity built up in the late 1990s has been worked off. But that investment remains weak because of subdued demand.

Financial Market Conditions
The Federal Reserve has eased monetary policy aggressively since the beginning of 2001, including cutting the federal funds rate by 0.5 percentage points in November 2002 (see Figure 2-2). Nevertheless, overall conditions in financial markets have not been conducive to economic growth. The plunge in stock values last year has substantially reduced household wealth and at the same time has raised businesses' cost of capital. Meanwhile, overall interest rates on corporate bonds have not fallen in tandem with rates on long-term Treasury securities because investors continue to perceive businesses as having a high risk of default. That perception has also caused banks to keep loan standards tight for many corporate borrowers. Those standards, along with weak demand for loans, have contributed to a relatively large drop in bank loans to businesses, even though the banking system is in good shape.
 

Figure 2-2.

The Federal Funds Interest Rate

Sources: Congressional Budget Office; Federal Reserve Board.

Note: The federal funds rate is the interest rate that banks charge for overnight loans.

One way to assess the impact on the economy of overall conditions in financial markets is to use an index--such as the one calculated by Macroeconomic Advisers (MA), a private forecasting firm--that combines the stance of monetary policy with a quantitative assessment of the channels through which that policy operates. MA's index draws on statistical relationships between GDP and financial variables such as interest rates, exchange rates, and measures of the stock market. It suggests that despite the Federal Reserve's policies, financial market conditions deteriorated sharply in 2002 (see Figure 2-3). The stimulative effect of the decline in short-term interest rates has been more than counteracted by the drop in the stock market and the still-elevated interest rates on corporate bonds, especially for riskier companies.
 

Figure 2-3.

An Index of Monetary and Financial Conditions

Sources: Congressional Budget Office; Macroeconomic Advisers, LLC.

Note: The index measures how financial variables such as interest rates, exchange rates, and the stock market affect the growth rate of real (inflation-adjusted) GDP.

Although the Federal Reserve acted quickly and aggressively to bolster the economy in 2001--before the recession was generally acknowledged--by early in 2002 its rate-cutting cycle appeared to have ended. The March 2002 statement of the Federal Open Market Committee (FOMC) noted that with a recovery under way, risks to its twin goals of price stability and sustainable economic growth had become balanced. By the committee's August meeting, however, the recovery seemed to be in danger of stalling, and the FOMC shifted back toward the view that risks were more heavily weighted toward economic weakness than toward inflation. That shift was followed by a cut in the target federal funds rate (to 1.25 percent) in early November, when the FOMC cited "greater uncertainty, in part attributable to heightened geopolitical risks, . . . currently inhibiting spending, production, and employment." The FOMC suggested that after the November cut, risks were once again in balance; as of mid-January, financial markets believe that further rate reductions are unlikely.

The stimulative effect of that monetary policy has been partly offset by a moribund stock market. The market typically rises at the beginning of a recovery, but the broad-based Standard & Poor's 500 index fell by 23 percent last year--the third consecutive year of decline. Analysts believe that decline was caused not only by uncertainty about the viability of the recovery but also by new concerns about corporate governance and the integrity of corporate earnings reports.

The corporate bond market has also counterbalanced some of the stimulative impact of monetary policy, as rates on corporate bonds have fallen less than interest rates on Treasury bonds of comparable maturity. In fact, the spread between interest rates on Treasury bonds and rates on corporate bonds--including those of investment grade--has increased to levels not seen since the early to mid-1980s (see Figure 2-4). The bond market is still plagued by the lingering effects of the late 1990s boom and its aftermath, when a number of once-high-flying firms (such as Enron and WorldCom) wound up defaulting. Through the end of 2002, credit-rating firms continued to issue more downgrades than upgrades. That situation, along with the perception that default risks are still high, is keeping the spread between interest rates wide, in contrast to the marked narrowing that typically occurs during the early stages of a recovery. Although conditions in the bond market appear to be stabilizing, any improvement in that market remains tentative, hampered by uncertainty about the durability of the recovery.
 

Figure 2-4.

Interest Rate Spreads on Corporate Bonds

Sources: Congressional Budget Office; Federal Reserve Board.

Note: These spreads measure the difference between interest rates on corporate bonds with an Aaa or Baa rating and interest rates on 10-year Treasury notes. The higher the spread, the riskier that investors believe corporate bonds to be.

Even so, less risky industrial and financial borrowers can still raise funds in credit markets, albeit subject to those wide spreads. The level of net new issues in the domestic bond market (although down by 26 percent from its high in 2001) amounted to nearly $500 billion during the first three quarters of 2002. New debt backed by collateral amounted to another $360 billion, up by 12 percent from a year earlier. Insurance companies and mutual funds have been significant buyers of corporate bonds, and foreigners remain substantial purchasers.

The banking system as a whole is healthy, although lending standards are still tight. Unlike in the early 1990s, few banks face difficulties from inadequate capitalization. In fact, bank capitalization has improved since the start of the recession. Nevertheless, banks have tightened their standards and terms of lending in the face of heightened uncertainty about the economy. Consequently, overall bank lending has grown at a tepid pace--one that is characteristic of recessions and early recoveries rather than expansions.

The Household Sector
Spending by households held up well last year despite the continued drop in the stock market. Real personal consumption expenditures rose at an average annual rate of 3 percent during the first three quarters of 2002, only about half a percentage point less than the average growth rate during the post-World War II period. (Those expenditures rose at a slightly higher rate, 3.1 percent, excluding spending on motor vehicles and parts.) In the fourth quarter of 2002, nominal retail and food-service sales grew by only 1.2 percent overall--but by a stronger 4.4 percent excluding motor vehicles and parts.⁽¹⁾ Both new and existing home sales reached record highs in 2002.

Household spending last year was bolstered by strong gains in disposable income, rising home values, near-record-low mortgage rates, and sales incentives for motor vehicles. Moderate growth in wages and salaries supported the growth of disposable income, which received a sharp boost from lower income tax payments. The continued rise in home values in many areas, combined with low mortgage interest rates, encouraged homeowners to refinance their mortgages to reduce their interest costs. Many homeowners also took out some equity from their homes when they refinanced so they could spend more on consumer goods and home improvements or repay other debts. Particularly attractive sales incentives boosted automobile purchases at the end of 2002. Strong growth in household borrowing, despite the opportunity to reduce debt-service burdens through refinancing, led to a slight deterioration in the financial health of households last year.

Employment and Income. A slight decline in employment was the reason that wages and salaries grew only moderately last year. Private nonfarm payroll employment decreased by 0.4 percent (or 438,000) between December 2001 and December 2002, despite the growth in real output (see Figure 2-5). Although employment appeared to stabilize during the middle of 2002, it began declining again, with a net 189,000 jobs lost in November and December. The manufacturing sector, which accounted for much of the total employment loss, continued to shed jobs at the end of last year, albeit at a slower pace than during the recession. Manufacturing employment looked poised for recovery in the spring of 2002, as the average workweek rose from its low of late 2001 and the pace of job loss slowed. After that, however, the gains in average weekly manufacturing hours disappeared, and the rate of job loss quickened. The temporary-help industry exhibited modest increases throughout the spring and summer of 2002, but they mostly evaporated late in the year. Employment in services (excluding temporary help) has resumed growing, but at a pace that is slower than typically occurs during a robust recovery.
 

Figure 2-5.

Employment in the Private Nonfarm Sector

Sources: Congressional Budget Office; Department of Labor, Bureau of Labor Statistics.

Despite a choppy monthly pattern, the broad movement in the unemployment rate reflects the weak employment picture. That rate reached a cyclical high of 6.0 percent in April 2002, up from an average of just 4.0 percent in 2000 (see Figure 2-6). The unemployment rate subsequently declined to 5.6 percent before climbing back to 6.0 percent at the end of 2002.
 

Figure 2-6.

Civilian Unemployment Rate

Sources: Congressional Budget Office; Department of Labor, Bureau of Labor Statistics.

In spite of the decline in employment, real wage and salary income has begun increasing, offering modest support for household spending (see Figure 2-7). Wages and salaries in the private sector rose at an annual rate of 3.1 percent in the second quarter of 2002 and 3.7 percent in the third quarter; they appear to have risen at a 3 percent to 4 percent rate in the fourth quarter. Because productivity is growing rapidly, employers have been able to increase workers' real hourly wages without hampering profits. That wage growth has outstripped price increases (consumer price inflation is running in the 2 percent to 2.5 percent range), which has allowed for a modest recovery in households' purchasing power.
 

Figure 2-7.

Growth in Disposable Income

Sources: Congressional Budget Office; Department of Commerce, Bureau of Economic Analysis.

In addition to higher wages and salaries, lower tax payments substantially augmented the growth of disposable income and supported consumer spending in late 2001 and 2002. Most households received tax rebates in the third quarter of 2001 (up to $600 for joint tax returns). At the same time, a decline of 1 percentage point in tax rates for people in the 28 percent and higher brackets went into effect. Beginning in January 2002, rates of withholding from paychecks were adjusted to take into account the new 10 percent bracket. Those various tax cuts reduced tax payments by about $67 billion in calendar year 2002. The amount of taxes owed by households fell significantly more than that, however, because of the weak economy, reduced realizations of stock options and capital gains, and fewer people in the highest tax brackets.

In all, real disposable personal income rose at an annual rate of 7.0 percent between the fourth quarter of 2001 and the third quarter of 2002--a stronger pace than in most past recoveries. More than half of that growth resulted from lower tax payments rather than higher pretax income. Unless lawmakers reach agreement on current proposals for additional fiscal stimulus, tax cuts will not provide further stimulus this year. In that case, additional increases in disposable income will have to come mainly from improved labor market conditions and wage gains.

Household Net Wealth. The continued drop in the stock market further eroded the net wealth of households last year (see Figure 2-8). Between the end of 2001 and the third quarter of 2002 (the latest data available), net household wealth dropped by $2.8 trillion because of the decline in stock prices. That decline probably reduced nominal consumer spending by around $100 billion, or slightly less than 1½ percent. Given the small rise in the stock market at the end of 2002, it seems likely that net wealth did not deteriorate further in the fourth quarter.
 

Figure 2-8.

Household Net Wealth

Sources: Congressional Budget Office; Federal Reserve Board.

Thus far, the personal saving rate has not responded noticeably to last year's drop in net wealth, and the possibility exists of a sharp rise in the saving rate (and a concomitant decrease in consumer spending), which would reduce economic growth. That risk is not included in CBO's forecast (see Box 2-2).
 

Box 2-2. The Wealth Effect and Personal Saving The unusually low rate of personal saving in recent years prompts concern about the strength of consumer spending in 2003. Between 1994 and 1999, the personal saving rate (personal saving as a percentage of disposable income) averaged only 4.7 percent, considerably below the average of 8.7 percent before 1994. Economists believe that a key reason for that low rate was a tremendous increase in stock prices and thus in consumers' net wealth. Between 1993 and 1999, consumers' net wealth rose by an astounding $18.3 trillion, and the ratio of net wealth to disposable personal income grew from 4.9 to 6.4--the highest level since at least 1952. That sharp rise in wealth allowed consumers to increase their spending faster than their income rose, causing the personal saving rate to plummet--from 7.1 percent in 1993 to 2.6 percent in 1999. Since 1999, by contrast, consumer net wealth has fallen markedly, and the ratio of net wealth to income has declined nearly to its value in 1993. But the personal saving rate has not risen to anywhere near its 1993 level. If consumers curtail their spending in an attempt to raise their saving rate to levels typically seen before the 1990s, they could undermine the economic recovery. Current data, however, suggest that the personal saving rate may not return to the levels that prevailed before the 1990s. The reason is that the relationship between the personal saving rate and the ratio of consumers' net wealth to disposable income seems to have undergone a fundamental shift. That change is visible in the figure below. The higher group of data points shows the relationship between the saving rate and the wealth-to-income ratio from 1952 to 1993; the lower set of points shows that relationship from 1994 to 2002. Trend lines drawn through the two groups of data points illustrate the shift. Although the wealth-to-income ratio in the third quarter of 2002 (4.9, the latest figure available) is within the 1952-1993 range of values, the personal saving rate in that quarter (3.8 percent) is below even the post-1993 trend.   Personal Saving Rate Versus Net Wealth Sources: Department of Commerce, Bureau of Economic Analysis; Federal Reserve Board. Why the relationship shifted in 1994 is unclear. One possibility is that the change is a statistical artifact that will disappear in future data revisions. In recent years, the Department of Commerce's Bureau of Economic Analysis has frequently revised the saving rate upward on the basis of more complete data and other changes when it annually revises the national income and product accounts. Another possibility is that changes in the markets for consumer credit and mortgage loans have made it easier and cheaper for consumers to borrow. As a consequence, consumers do not need to save as much in advance for purchases and for down payments on homes. The shift does not appear to depend on the definition of the personal saving rate. The saving rate used in the figure is the measure from the national income and product accounts. It considers saving to be all income from current production that is not spent on consumer goods and services, interest paid by persons, and personal transfer payments to the rest of the world. A different measure comes from the flow-of-funds accounts maintained by the Federal Reserve Board.1 That measure defines personal saving as the household sector's net acquisition of financial assets plus the net investment in tangible assets minus the net increase in liabilities. A shift is apparent using that measure. Other measures of personal saving do not appear to explain the shift either.2 1.  Board of Governors of the Federal Reserve System, Flow of Funds Accounts of the United States (December 5, 2002). 2.  Examples of other measures are described in Maria G. Perozek and Marshall B. Reinsdorf, "Alternative Measures of Personal Saving," Survey of Current Business (April 2002), pp. 13-24.

The effect of falling stock prices on household wealth has been counteracted, to a limited degree, by rising housing prices. In the third quarter of 2002, prices of single-family homes were 6.2 percent higher than in the same quarter a year earlier, according to the Office of Federal Housing Enterprise Oversight. Those high housing prices have combined with low interest rates to trigger a boom in mortgage refinancing. Refinancing activity last year surpassed the record pace of 2001 by 37 percent. When homeowners refinance mortgages, many of them convert some of their accumulated housing equity into cash. Survey data indicate that roughly half of those proceeds are typically used for either consumer spending or home improvements. Thus, the refinancing boom probably contributed a few tenths of a percentage point to last year's growth in personal consumption spending.

The Financial Health of the Household Sector. Consumers' financial health has eroded slightly, and households are more indebted than they were before the 2001 recession. As a result, the household sector is vulnerable to financial problems should the growth of income falter.

Real household debt has risen much faster than is normally seen during a recession and early recovery. The growth of real mortgage debt continued to accelerate in 2002, to its fastest pace since 1990, and consumer credit grew a bit more slowly than disposable personal income. Because interest rates have stayed low, the rapid rise in debt has not increased households' debt-service burden markedly. But that burden has not fallen, as it typically does during and immediately after a recession.

The rate of delinquencies on conventional mortgages has increased in the past few years (although it is lower than in the 1981-1982 recession and about the same as during the 1990-1991 recession). The delinquency rate is especially large on higher-risk FHA loans (see Figure 2-9).
 

Figure 2-9.

Mortgage Delinquency Rates

Sources: Congressional Budget Office; Mortgage Bankers Association.

Notes: FHA = Federal Housing Administration; VA = Department of Veterans Affairs.

However, mortgage delinquencies and foreclosures appear to be lagging indicators, so they may peak soon if the economy continues to recover. Indeed, mortgage delinquency rates edged down in the third quarter of 2002.

The delinquency rate on a broad range of consumer loans at commercial banks, by contrast, is lower than it was at the start of the 2001 recession. That relatively better rate may reflect the fact that households used some of the proceeds from refinancing mortgages to pay down consumer loans. In addition, banks have kept a tight rein on standards and terms of such loans, helping to minimize delinquencies. Nevertheless, the delinquency rate on credit cards surged in 2001 and remained at a very high level in 2002, suggesting credit problems among some borrowers (see Figure 2-10).
 

Figure 2-10.

Delinquency Rates on Consumer Loans at Banks

Sources: Congressional Budget Office; American Bankers Association.

The Housing Market. The market for housing has been a source of strength in this recovery. Real residential investment surged to all-time highs in each of the first three quarters of 2002, and housing starts for the year as a whole were at their highest level since 1986. Moreover, sales of both new and existing single-family homes reached record levels in 2002 (see Figure 2-11). Those sales have been fueled by the lowest mortgage rates since the 1960s (see Figure 2-12). According to Freddie Mac, late in 2002, interest rates were just above 6 percent for 30-year fixed-rate mortgages, around 5.5 percent for 15-year fixed-rate mortgages, and between 4 percent and 4.25 percent for one-year adjustable-rate mortgages. All of those rates were about a percentage point lower than they were early in 2002.
 

Figure 2-11.

Sales of New Homes

Sources: Congressional Budget Office; Department of Commerce, Bureau of Census.

Note: Data are three-month moving averages.

 

Figure 2-12.

Mortgage Interest Rates for Existing Homes

Sources: Congressional Budget Office; Federal Housing Finance Board.

Several indicators suggest, however, that the housing market may decelerate soon. Nationally, the increase in housing prices has slowed, suggesting lower growth in demand, and prices in some areas have begun to decline. Some analysts suggest that housing prices may have risen by more than the underlying conditions of supply and demand warrant, at least in some metropolitan areas, which means that prices in those areas could fall. In addition, the rise in delinquencies among high-risk borrowers could cause mortgage lenders to tighten credit terms and standards for such borrowers.

Motor Vehicles. Purchases of cars and light trucks have been another important element bolstering consumer spending over the past year. After the terrorist attacks of September 11, 2001, automakers feared that consumers would stop buying major items such as cars. To prevent that from happening, General Motors offered its customers zero-interest financing beginning in October 2001; Ford and the Chrysler unit of Daimler-Chrysler quickly matched that offer. As a result, sales of cars and light trucks reached a near-record level that month--an annual rate of 21.1 million vehicles--and remained at high levels throughout most of 2002 (see Figure 2-13). Some industry observers fear that those incentives may soon lose much of their impact, but vehicle sales remained strong at the end of 2002.
 

Figure 2-13.

Sales of Cars and Light Trucks

Sources: Congressional Budget Office; Department of Commerce, Bureau of Economic Analysis.

Note: Data are three-month moving averages of annual rates.

The Corporate Sector
Whereas spending by the household sector has helped the economy recover, weakness in the corporate sector restrained growth last year. Excess capacity, weak corporate profits, the high cost of raising funds for investment in either the stock or bond market, sluggish growth of final sales, and pervasive uncertainty have all inhibited companies from making new investments in plant and equipment, rebuilding inventories, and restoring the growth of employment.

Corporate investment has been on a roller-coaster ride in recent years. It grew explosively during the late 1990s, fueled by rising stock prices, strong growth in demand, and excessive investment in information technology (computers, software, and telecommunications equipment). Real investment in producers' durable equipment and software surged at a rate of 11.6 percent a year, on average, between 1994 and 2000. Although much of that growth came from purchases of computers and software (prompted in part by rapid declines in quality-adjusted computer prices), other investment in producers' durable equipment rose at a healthy pace.

In late 2000, however, investment growth slowed sharply as stock prices fell and businesses began to pull back from investing in information technology. In 2001, investment in overall producers' durable equipment and software declined by 6.4 percent. Investment in nonresidential structures (which had stayed strong through the summer of 2000 before declining in early 2001) plummeted at an annual rate of 30 percent in the fourth quarter of 2001 and continued to fall at double-digit rates throughout 2002. Today, equipment investment appears to be recovering modestly, mainly because businesses have eliminated much of the overhang of excess investment in information technology built up during the boom years. Nonetheless, business fixed investment is unlikely to return to the high share of GDP that it constituted in the late 1990s, because the factors that caused that share are not expected to recur on a sustained basis.

An important factor inhibiting a revival of investment so far is excess capacity. The rate of capacity utilization in manufacturing plunged from 82.2 percent in the first half of 2000 to 73.4 percent in the fourth quarter of 2001, driven by a decline in demand for goods (see Figure 2-14). That drop left the capacity utilization rate considerably lower than during the 1990-1991 recession

---