Consumer Sector Outlook for 2005
It is widely recognized that the consumer is the powerhouse behind the U.S. economy, accounting for more than two-thirds of total economic activity. And, despite the 2001 recession and a relatively weak labor market since then, the consumer has proven to be far more resilient than many would have expected. In recent years, growth in after-tax income and consumer spending has been much stronger than has historically been the case around recessions, especially given the lackluster pace of hiring. In addition, consumer credit growth has been robust during a period when a retrenchment would have been expected. To stoke their spending, consumers have drawn on the increasing equity in their homes and the proceeds of substantial tax cuts during both 2001 and 2003. However, as the stimulus of the tax cuts has been absorbed and the refinancing wave appears to have abated, it is now uncertain how fast consumer spending can continue to grow.

This article reviews the various factors that will determine the pace of consumer spending and credit growth going forward. It also briefly assesses how higher energy costs may be affecting spending and concludes with an evaluation of the overall credit health of the consumer.

Job Growth Drives Income
After a few rocky years, renewed job growth is again supporting income gains. During the 2001 recession and for almost two more years, total U.S. payroll employment declined. This decline weighed on growth in employee compensation and thus on total income gains. However, since August 2003, the economy has been steadily increasing its employment base, leading to renewed vigor in total employee compensation and income growth (see Chart 1). In addition to any raises offered to existing workers, job gains should continue to support overall compensation growth in 2005 and will keep incomes and consumer spending expanding. However, two other sources of income growth in recent years—tax cuts and the liquidation of homeowners' equity—are likely to play a lesser role in 2005. Also, higher energy prices may pose a risk to consumer spending.

Chart 1
D

Tax Cuts and Income Affect Spending
The tax cuts that helped support take-home pay during the 2001 recession and the subsequent episode of weak employment growth are no longer boosting income at the margin. Household incomes usually do not decline on an annual basis. In fact, since 1950, inflation-adjusted after-tax incomes only fell once, in 1974, and then only by 0.7 percent. Rather, growth in real after-tax income typically slows when the job market weakens substantially. Although the recent experience was no exception to this trend, the slowdown in after-tax, or so-called "disposable," income growth was much milder than the historical experience might have suggested. During the early years after the 2001 recession, disposable incomes grew largely on the strength of two well-timed tax cuts that boosted after-tax income growth for those still working.

Tax rebates in the summer of 2001 and reduced tax withholdings in 2002 and 2003 increased gains in take-home pay relative to total income. Chart 2 illustrates the disparity between before- and after-tax income growth. After-tax income received a large boost beginning in late 2001 as tax cuts began to flow into the economy—a situation that carried over well into 2004. However, in July and August 2004, growth in after-tax income fell significantly below that of pre-tax income. Since the last round of payroll tax reductions was enacted a year earlier in mid-2003, the stimulative effects on growth are now diminishing. That is, while these tax cuts still provide a higher level of take-home pay for many workers, the influence of the tax cuts on income growth has ended.

D

Although tax cuts are no longer contributing to income growth, the increased pace of hiring is having a positive effect, which highlights the significance of job growth to the outlook for consumer spending and borrowing. Without a steady pace of job growth, overall and take-home pay gains would be weaker.

As total homeowner equity has risen, many households have sought to liquidate and spend a portion of the wealth tied up in their homes. One way to do so is during a mortgage refinancing transaction, where the homeowner might choose to take cash out or roll a second mortgage loan into the first mortgage. According to data provided by Freddie Mac, homeowners liquidated some $211 billion in 2003 by refinancing their mortgages. According to the Federal Reserve, another $101 billion was liquidated by increased borrowing against home equity lines of credit (HELOCs).¹ Taken together, this liquidation of home equity contributed an extra $312 billion to household cash flow during 2003, an amount of stimulus that almost equaled the $332 billion gain in after-tax income during the year.

The current combination of high home prices and historically low interest rates continues to induce homeowners to liquidate home equity, and increasingly they are doing so with HELOC borrowing. Because so many homeowners refinanced their mortgages during 2003 when interest rates were at a multi-decade low, a relatively modest rise in mortgage rates in 2004 has curbed refinancing activity substantially. The average weekly refinancing index of the Mortgage Bankers Association declined by 75 percent in the first half of 2004 from year-ago levels. Meanwhile, the total volume of home equity loans outstanding rose by 23 percent during the year ending in June, as households turned to HELOCs as a source of liquidity.

A key issue for the consumer spending outlook is whether home appreciation can continue to provide a ready source of consumer cash. During 2002 and 2003, despite the liquidation of roughly $540 billion in homeowner equity and an overall increase in household mortgage debt of $1.4 trillion, total homeowner equity still rose by $1.2 trillion. Strong home price gains depend on continued robust housing demand, but rising interest rates may cut into demand and slow the pace of price increases. Some analysts have expressed concern that the recent rapid pace of home price growth in many major markets across the country could lead ultimately to a downturn in home prices and housing market activity.² Should home prices decline in these markets, homeowner equity—and spending—could also be adversely affected.

Higher Energy Costs Affect Spending
The recent run-up in energy costs may pose a downside risk to consumer spending growth in 2005. Inflation-adjusted consumer spending appeared to take a breather in second quarter 2004 after an extended period of strong growth. After advancing by roughly 3 percent in both 2002 and 2003 and then at a 4 percent annual rate in first quarter 2004, spending rose at only a 1.6 percent annualized rate in the second quarter. Sluggish real retail sales growth persisted through August, with motor vehicle sales being the weakest component. Automakers curtailed many of their incentives in early 2004 and auto loan rates rose modestly, causing weakness in auto demand. However, with the reintroduction of incentives and a brief summertime easing in energy prices, auto and overall retail sales bounced back in September. As a result, inflation-adjusted U.S. consumer spending advanced at a sturdy 4.6 percent annualized pace during the third quarter.

Higher energy prices likely played a key role in dampening consumer goods spending in mid-2004. Federal Reserve Chairman Alan Greenspan recently noted that the 2004 surge in energy costs—and the price of gasoline in particular—weighed on consumer spending growth and contributed to a mid-2004 "soft patch" in economic growth.³ Although spending for gasoline and oil accounts for an average of only 2.5 percent of after-tax personal income (compared to 4.4 percent in 1981), lower-income households are typically more affected by rising energy costs (see Chart 3).

D

Consumer sales data offer another indication that lower-income households may be feeling a greater effect from energy price increases. Michael Niemeira, chief economist for the International Council of Shopping Centers, estimates that for every 10 percent increase in the price of gasoline, sales at discount retailers drop 0.66 percent, versus only 0.33 percent at department stores.⁴ This may help explain why recent same-store sales data have shown stronger growth among high-end retailers than among discounters (see Chart 4). Other analysts have pointed out that sales growth for high-end items, such as private aircraft and pleasure boats, was stronger this year than in the past few years. It is likely that other factors, such as the relatively faster income growth seen by higher-income households recently, have contributed to the divergence in sales trends. Nonetheless, the retail data provide some anecdotal evidence that higher oil and energy prices affect high-income households less than other income-level households.⁵

D

Although the recent increase in energy costs may have weighed on consumer spending growth during the past year, consumers likely will continue to adjust their consumption patterns over time to adapt to the higher level of energy prices. Given that adjustment process, the effect of rising energy costs on consumer spending may be less severe during 2005, even if prices remain elevated.

Increasing Consumer Indebtedness and Worsening Credit Quality a Concern
The continued expansion in consumer indebtedness and the rising rate of personal bankruptcy filings in recent years has raised concern. If job or income growth were to slow or interest rates were to rise significantly, consumer credit quality could deteriorate.

Current levels of consumer debt are probably still manageable for most households. At just over 21 percent in mid-2004, consumer debt as a share of net worth was near its high point for the past 55 years (see Chart 5). This level of indebtedness has been rising steadily over the past several decades. Much of this rise is due to two factors: increasing access to credit by certain households and rising homeownership. In particular, rising homeownership seems to have played a significant role in the increase in indebtedness over the past several years. During the five years ended mid-2004, household mortgage debt increased by $2.8 trillion, which accounts for 80 percent of the total increase in consumer debt over this period. At the same time, however, the value of residential real estate owned by homeowners rose by $5.9 trillion. So the dramatic increase in mortgage debt during the past five years was accompanied by an even larger increase in net worth, the net effect of which is over $3 trillion.

D

Although personal bankruptcy filings have declined in recent months, they remain near their all-time highs. During 2003, personal bankruptcy filings reached a record high of 5.5 per thousand persons; however, filings declined an encouraging 4.2 percent on a year-ago basis in second quarter 2004. This was the largest year-over-year decline since mid-2000. Any sustained trend toward fewer personal bankruptcies should indicate improving consumer credit performance.

So far, the credit position of the consumer appears to be stable. According to the Federal Reserve, the aggregate consumer debt-service ratio—which reflects minimum required payments on credit cards, home mortgages, and other consumer loans (but not leases)—has been stable at around 13 percent of disposable income since the end of the 2001 recession (see Chart 6).

D

Although the aggregate consumer debt-service ratio remains near its all-time high, some of the reasons for this may be structural in nature. The increased availability of credit in recent years, rising homeownership, and more sophistication on the part of borrowers and lenders (which can be reflected in such developments as increased "convenience" use of credit card debt) may all be contributing to this higher ratio of minimum debt payments to income.⁶ The fact that delinquency rates on consumer loans and credit card debt at FDIC-insured institutions have held relatively steady during the past seven years offers perhaps the strongest evidence that consumers continue to be able to service their debts. Between 1997 and mid-2004, noncurrent credit card loans ranged between 1.84 and 2.00 percent of all credit card loans, while noncurrent loans in other consumer loan categories ranged between 0.91 and 1.06 percent.⁷

D

However, not all mortgage debt has a fixed rate. Freddie Mac estimates that between 15 percent and 20 percent of total outstanding mortgage debt had adjustable rates in 2003. Adjustable-rate mortgages (ARMs) have become more popular in the mortgage market, accounting for at least 30 percent of mortgage origination activity over the spring and summer of 2004.⁸ While many consumers will be able to manage rising mortgage payments on ARMs, the segment of newer ARM holders that relied on the lower cost of this type of loan as their only means to afford a monthly mortgage payment may face some difficulty servicing their debt once interest rates rise.⁹ Although there are many types of ARM products, a significant share of ARMs still reprice in less than a year.¹⁰

The potential for ongoing increases in interest rates over the next year poses another possible risk to consumer debt service capacity. Rising interest rates may hurt housing affordability, thereby curbing demand for new homes and home price appreciation. So, although many consumers have been able to substitute lower priced fixed-rate mortgage debt for higher-priced revolving debt in recent years, this may be less feasible in 2005. Should home price appreciation subside significantly, many households may meet their incremental credit needs by shifting back toward higher-priced revolving credit lines instead of refinancing their homes at lower fixed rates and extracting equity.

Overall household debt levels remain near all-time highs, but the recent performance of consumer loan credit quality suggests that most consumers are fully able to service their current debt loads. Although interest rates likely will keep rising, some factors will insulate many consumers from the risks of rising rates. For example, much of the growth in consumer debt in recent years came in the form of fixed-rate mortgages, where monthly payments are not subject to rising interest rates. Furthermore, many consumers have swapped more costly forms of debt, such as credit cards, for lower-cost fixed-rate mortgage debt.

Such actions have reduced overall consumer exposure to rising interest rates, but not all consumers have been able to avail themselves of these developments. Increasing interest rates could begin to strain some consumers' finances over the coming year, and it will be important to monitor certain consumer segments in 2005 for emerging signs of weakness. In particular, weaker borrowers and consumers with significant leverage and exposure to variable-rate debt will remain vulnerable to increases in interest rates.

Nathan Powell, Financial Economist

¹ See "Home Equity Lending: Growth and Innovation Alter the Risk Profile" by Cynthia Angell in this issue.

² See "A Turning Point Ahead? National and Regional Trends in Residential Real Estate Markets" in this issue.

³ Chairman Alan Greenspan of the Federal Reserve Board, "Economic Outlook," testimony before the Committee on the Budget, U.S. House of Representatives, September 8, 2004, http://www.federalreserve.gov/ boarddocs/testimony/2004/200409082/default.htm.

⁴ Barbara Hagenbaugh, "$50 Oil Probably Won't Cause a Huge Economic Mess," USA Today, September 27, 2004.

⁵ Andy Kish, "What We're Doing Without," Economy.com, September 28, 2004.

⁶ Alan Greenspan, "The Mortgage Market and Consumer Debt," speech at America's Community Bankers Annual Convention, Washington, D.C., October 19, 2004, http://www.federalreserve.gov/ boarddocs/ speeches/2004/20041019/default.htm.

⁷ Noncurrent loans are defined as those that are 30 to 90 days past due as well as those in nonaccrual status.

⁸ Mortgage Bankers Association, Weekly Mortgage Applications Survey.

⁹ For a more detailed analysis of the U.S. housing market, see: Cynthia Angell, "Housing Bubble Concerns and the Outlook for Mortgage Quality," FDIC Outlook, Spring 2004.

¹⁰ Based on 1993, 2000, and 2003 Federal Housing Financing Board survey data that reflect responses from its members regarding all fully amortized purchase-money conventional first-mortgage loans used to finance the purchase of single-family nonfarm homes, including individual townhouse, condominium, and cooperative units.