Rising Household Debt: Context and Implications

Rising Household Debt: Context and Implications
June 17, 2008
Brian W. Cashell
Specialist in Macroeconomic Policy
Government and Finance Division

Rising Household Debt:
Context and Implications
Summary
The financial condition of households has important implications for a number
of economic issues relevant to public policy. In the short term, unsustainable growth
in debt carried by the household sector may increase the likelihood of a slowdown
in consumer spending and a sluggish economy. In the long term, households that save
and accumulate wealth may promote economic growth. Households that accumulate
debt may affect the national rate of investment in productive assets. Moreover,
households that accumulate debt to finance consumption spending without building
wealth will likely find themselves ill prepared for retirement.
How much debt a household accumulates depends on its willingness and ability
to borrow money. Current income and interest rates play a role in how much debt a
household is willing to take on. However, one critical variable is difficult to
measure: individual expectations about future income. Because debt represents an
obligation to make future payments, future income is at least as important as present
income. Expectations about future prospects are also subject to change over time.
An increase in optimism about the future is likely to increase the amount of debt a
household is willing to carry; an increase in pessimism is likely to reduce it.
Household debt is at historically high levels, both in absolute terms and relative
to personal income. The best measure of the burden of debt on households may be
the required payments relative to the income available to make those payments. Data
indicate that the debt service burden is higher now than it has been for most of the
past 25 years.
Three developments in recent years have complicated the interpretation of
available measures of household debt. The first was the phase-out of the
deductibility of interest payments on consumer loans on federal income taxes. The
second was an increase in leasing as an alternative to financing the purchase of
automobiles. The third was an increase in the “convenience use” of credit cards as
a substitute for other means of payment for goods and services.
Most studies of the relationship between household debt and economic growth
suggest that rising debt is not a threat to economic growth. Rather than a harbinger
of economic hard times, increases in the dollar value of household debt have been
associated with a growing economy. Changes in consumer debt tend to be a leading
indicator of consumer spending and thus of overall economic growth. One reason
may be that increases in consumer borrowing are an indication of confidence in the
economy, both on the part of borrowers and lenders. Moreover, what is a burden for
borrowers is income for lenders. As long as borrowers are able to meet their debt
obligations, payments are simply a transfer of income. Although such payments may
constrain the discretionary spending of borrowers, they increase lender resources.
This report will be updated as conditions warrant.

Contents
In troduction ......................................................1
Why Do Households Borrow?........................................1
Recent Trends in Household Debt.....................................3
Alternative Debt Measures ......................................5
The Demographics of Household Debt............................12
Household Debt as an Economic Indicator.............................14
List of Figures
Figure 1. Outstanding Household Debt.................................3
Figure 2. Household Debt as a Percentage of Aftertax Income...............6
Figure 3. Household Debt as a Percentage of Household Assets.............7
Figure 4. Debt Service and Financial Obligation Ratios....................9
Figure 5. Homeowner and Renter Financial Obligation Ratios..............10
Figure 6. Consumer Debt as a Percentage of Aftertax Income..............11
List of Tables
Table 1. Share of Households Holding Debt of Any Kind in 1998, 2001,
and 2004, by Income Percentile..................................12
Table 2. Ratio of Debt Payments to Income............................13
Table 3. Families With Payment 60 Days or More Late...................13

Rising Household Debt:
Context and Implications
Introduction
The financial condition of households has important implications for a number
of economic issues of relevance to public policy. In the short run, there may be
concern that rising levels of household debt may mean an increase in the likelihood
of a slowdown in consumer spending and a sluggish economy. In the longer run,
households promote economic growth by saving and building wealth. If instead,
households are accumulating debt, that may have an effect on the national rate of
investment in productive assets. Moreover, those households that accumulate debt
to finance consumption spending and not investment may find themselves ill
prepared for retirement.
Household debt is currently at historically high levels, both in absolute terms
and relative to personal income. High levels of debt, it is feared, constitute a risk that
sooner rather than later consumers must cut back their spending. One review of the
popular press found that most references to consumer debt referred to its potential for¹
negative effects on economic growth. Some economists say that may be mistaken.
Debt, in their view, is not necessarily a bad thing. While interest payments may be
a burden to those who are in debt, they are a plus for those who hold the securities.
This report examines the economics of household debt.²
Why Do Households Borrow?
Households take on debt for a variety of reasons, and the effect debt has on the
economy depends on the reason for the debt. In some cases, individuals borrow to
acquire assets and in some cases they borrow to consume more than their current
income would allow.
A standard economic assumption regarding household behavior is that people
prefer to smooth out their consumption over the course of their lifetimes. In other
words, they seek to insulate their standard of living from both short-term fluctuations

¹ Thomas A. Durkin and Zacharia Jonasson, “An Empirical Evaluation of the Substance and
Cyclicality of Financial Reporting: The Case of Consumer Credit,” Board of Governors of
the Federal Reserve System, April 1998.
² This report does not examine the relationship between debt and bankruptcy. See CRS
Report RS20777, Consumer Bankruptcy and Household Debt, by Mark Jickling and Heather
D. Negley.

in income as well as the typical rise and fall in income that occurs over the course of
their lifetimes. One way they do that is to vary their saving rate, saving relatively less
early and late in life and saving relatively more during their peak earning years.
Another way to do that is to borrow. Borrowing makes it possible to separate
the cost of consumption from the consumption itself. Households can consume more
than they might otherwise and shift the cost to a time when their income is higher and
it is less of a burden to pay.
Not all household debt is used to finance consumption spending. Borrowing to
finance the purchase of a home is different, in that the good is not used up. Repaying
a mortgage is a way for many to accumulate wealth.³ In the case of durable goods,
such as household furnishings or automobiles, borrowing allows consumers to match
a stream of payments more closely to the service life of the good.
Even households with considerable wealth may find it advantageous to borrow
instead of selling some of their assets. Some assets are less liquid than others. That
is, they may take a considerable amount of time to sell, and there are costs associated
with selling them that might make it preferable to borrow.
How much debt a household accumulates depends on both its willingness and
ability to borrow money. Current income and interest rates obviously play a role in
how much debt a household is willing to take on. But one critical variable is difficult
to measure, and that is individual expectations about future income. Because debt
represents an obligation to make future payments, future income matters at least as
much as present income. But there is always some uncertainty regarding an
individual’s economic prospects, and for that reason many may take on more or less
debt than they would if they had perfect foresight. Expectations about future
prospects are also subject to change over time. An increase in optimism about the
future would be likely to increase the amount of debt a household would be willing
to carry, and an increase in pessimism would reduce it.
If households overestimate future income they may end up holding more debt
than they want, or than they can afford. In recent years, some households apparently
took on mortgage debt that they otherwise might not have, or that might not
otherwise have been made available to them, in anticipation of continued house price
appreciation.
The ability of households to get credit depends on their income and also on the
intended use of the debt. Those loans that are used to finance the purchase of a car
or a house are secured and so carry less risk to the lender. Those loans that are used
to finance current consumption expenditures are unsecured, pose a greater risk to
lenders, and so may have higher interest rates and be more difficult to get.

³ Taking on mortgage debt does, however, make households vulnerable to fluctuations in
house prices which may affect the rate of wealth accumulation.

Recent Trends in Household Debt
Household debt grew rapidly during the economic expansion of the 1990s.
Between the first quarter of 1991, at the beginning of an economic expansion, and
the first quarter of 2001, when that expansion came to an end, total household debt
doubled, increasing at an annual rate of 7%. Total consumer credit outstanding also
doubled during that period, as did the total value of outstanding mortgage debt. When
the economic contraction began in March 2001, household debt continued to grow,
as has also been the case since the latest expansion began in November 2001. This
suggests that growth in household debt is not necessarily related to the ups and
downs of the business cycle. Over the long run, household debt reaches a new high⁴
about every second quarter. Reaching a record level of household debt would appear
to be an unremarkable event. Figure 1 shows recent trends in the level of household
debt.
Figure 1. Outstanding Household Debt

^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
⁴ François Velde, “The Household Balance Sheet — Too Much Debt?,” Chicago Fed Letter,
September 2002, no. 181a.

At the beginning of 2000, mortgage debt accounted for 74% of total outstanding
household liabilities. At the end of 2007, it accounted for more than 80% of total
household debt.⁵ The share of household debt accounted for by mortgage debt has
been increasing, and between 1999 and 2007, the increase in mortgage debt
accounted for 86% of the increase in overall household debt.⁶
Three developments in recent years may have changed the significance of
available measures of household debt. The first was the phase-out of the deductibility
from federal income taxes of interest payments on consumer loans. The second was
an increase in leasing as an alternative to financing the purchase of automobiles. The
third was an increase in the “convenience use” of credit cards as a substitute for cash
or checks as means of payment for goods and services.
The Tax Reform Act of 1986 phased out the deductibility of interest payments
on consumer loans. That led to a boom in home equity lending. Home equity loans
accounted for 13.4% of total home mortgage debt outstanding in the second quarter
of 2002. Because of the change, measures of consumer debt may underestimate the
extent to which households are borrowing to finance current spending.
In the 1990s, the popularity of automobile leasing increased dramatically.
Between 1992 and 2001, the share of households leasing a vehicle rose from 2.9%
to 5.8%, although it fell to 4.0% in 2004.⁷ Leasing has a number of advantages. For
one, the consumer need only finance a portion of the total value of the car. Second,
the leasing company can write off the depreciation on the car and some of that benefit
may be passed on to the consumer. But much like a consumer loan, a lease
agreement obligates the consumer to a stream of payments. That it is not part of total
consumer debt may cause the financial vulnerability of the household sector to be
underestimated.⁸
A considerable share of credit card debt is paid off before it accrues any interest
charges. Figures from the Federal Reserve Board’s Survey of Consumer Finances
(SCF) show that in 2004, 74.9% of families had credit cards, and of those families
only 58% had an outstanding balance at the time of the survey. The share of families
reporting that they usually paid off their credit card bills each month rose slightly

⁵ Data on household debt are from the Flow of Funds Accounts (FOF) published by the
Board of Governors of the Federal Reserve System. Home mortgage and consumer debt do
not account for all of the liabilities of the household sector. Debts of non-profit institutions
are consolidated into the household sector in the FOF. Other types of household debts not
classified as consumer loans include student loans, and borrowing against insurance
policies.
⁶ Karen E. Dynan and Donald L. Kohn, “The Rise in U.S. Household Indebtedness: Causes
and Consequences,” Finance and Economics Discussion Series Working Paper 2007-37,
Board of Governors of the Federal Reserve System, August 2007.
⁷ Board of Governors of the Federal Reserve System, Survey of Consumer Finances.
⁸ See Francesca Eugeni, Consumer debt and home equity borrowing, Federal Reserve Bank
of Chicago Economic Perspectives, vol. 17, March-April 1993, pp. 2-13.

from 55.4% to 55.7% between 2001 and 2004. That suggests that a substantial share
of the population uses credit cards more for convenience than as a source of credit.⁹
A study published by the Federal Reserve Board examined the contribution of
the convenience use of credit cards to the increase in consumer debt. The study
estimated that in 1992 convenience use accounted for about 6% of outstanding credit
card debt, and that by 2001 that proportion had risen to about 11%. The study also
estimated that if convenience use had not increased between 1992 and 2001, growth
in total credit card debt would have been one percentage point per year slower over
the period.¹⁰
In 2003, the four agencies that regulate the financial institutions that issue credit
cards announced new guidelines regarding minimum payments on outstanding credit
card balances.¹¹ Specifically, the agencies expected lenders to require minimum
payments for each outstanding balance sufficient to amortize the debt over a
“reasonable period of time.”¹² In response, many credit card companies began
increasing their minimum payment requirements in 2005.¹³ This development may
have slowed recent growth in credit card debt.
Another recent change may have influenced the willingness of households to
borrow. The 109^th Congress passed bankruptcy reform (P.L. 109-8), establishing a
means test for those filing for bankruptcy. The amount of debt relief available to
filers above specified income thresholds is now restricted.¹⁴
Alternative Debt Measures
Measures of household debt outstanding say little about how much of a burden
the debt is, or how much of a risk it poses to the financial health of the population.
One measure that may be useful in this regard is the level of debt relative to current
income.¹⁵ Whether a given level of debt is likely to pose financial risks depends on

⁹ See Peter S. Yoo, “Charging Up a Mountain of Debt: Accounting for the Growth of Credit
Card Debt,” Federal Reserve Bank of St. Louis Review, March/April 1997, pp. 3-13.
¹⁰ Kathleen W. Johnson, “Convenience or Necessity? Understanding the Recent Rise in
Credit Card Debt,” Finance and Economics Discussion Series, 2004-47, 28 pp.
¹¹ The four agencies are the Board of Governors of the Federal Reserve System, The Federal
Deposit Insurance Corporation, the Office of the Comptroller of the Currency, and the
Office of Thrift Supervision.
¹² See Federal Deposit Insurance Corporation Press Release PR-02-2003, January 8, 2003.
¹³ See CRS Report RS22352, Credit Card Minimum Payments, by Pauline Smale.
¹⁴ CRS Report RS22511, Preliminary Observations on the Impact of the Bankruptcy Abuse
Prevention and Consumer Protection Act of 2006 (P.L. 109-8), by Brian W. Cashell, Mark
Jickling, and Heather D. Negley.
¹⁵ Debt relative to expected future income might be more interesting, since that would give
a better idea of how much debt households felt they could afford. Such a measure is not
readily available. Those who expect their incomes to rise over time may be willing to take
(continued...)

how much income is available to cover its costs. Figure 2 shows household debt as
a percentage of disposable (aftertax) personal income.
Figure 2. Household Debt as a Percentage of Aftertax Income

^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
The ratio of debt to income is not sufficient to determine when, or if, households
are carrying too much debt for their own good, but changes in the ratio over time may
suggest relative degrees of risk faced by households. Figure 2 suggests that those
risks have been rising relatively rapidly since 2000. At the beginning of 2000, total
household debt represented roughly 86% of annual after tax income. At the end of

2007, that proportion was 126%.

That ratio is high relative to the recent past. But it is not out of line with the
ratios in some other developed countries. According to the Organisation for
¹⁵ (...continued)
on more debt than those who do not, even in cases where current incomes are equal.
Younger householders may be willing to hold more debt relative to current income if they
expect their earnings to rise over the course of their careers. Households may also borrow
less than they might otherwise because of uncertainty about future income streams.

Economic Co-operation and Development (OECD), household debt as a percentage
of disposable income in the United Kingdom has also been rising in recent years and
was 163% in 2006. In Canada the ratio in 2006 was 127%.
Income is not the only resource available to households against which to
measure their financial risks. If necessary, households can rely on other assets,
financial or real, to secure or pay off existing debts. While liquidating assets to cover
the costs of debt may help avoid more serious financial consequences, it is still a sign
of financial vulnerability.
Selling real assets, such as land, can take time and involves considerable
transaction costs. Selling financial assets takes less time, but it may be that they must
be sold at an inopportune moment. Nonetheless, the more assets a household has, the
more likely it is that it will be able or willing to take on additional debt, and the less
likely it is that it will default on those debts. Figure 3 shows the ratio of household
debt to household assets. This measure of assets includes both the market value of
financial and tangible assets. Tangible assets include equity in real estate, as well as
consumer durable goods, mainly automobiles.
Figure 3. Household Debt as a Percentage of Household Assets

^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
Between 1983 and the mid-1990s, the ratio of debt to assets tended to rise
modestly, as was the case with the debt to income ratio. But, unlike the debt-income

ratio, the ratio of debt to assets fell between the mid-1990s and 2000. Much of that
decline was attributable to the large increase in stock prices. Between 1994 and
2000, total household liabilities increased by 57.7%. Over the same period, total
household assets increased by 66.9%, and the market value of equity held directly or
indirectly by households increased by 167.9%. Between 2000 and 2002, stock prices
fell, contributing to an increase in the ratio of debt to assets. The stock market has
since recovered, but debt has continued growing more rapidly than assets. Total
household debt as a percentage of assets is now around 20%, which is higher than
it has been over the past 25 years.
The willingness, and ability, of households to borrow is affected by their wealth.
But the extent to which estimates of wealth, at any given time, are perceived to be
durable may also be an important consideration. In other words, if household wealth
rises because of an increase in the stock market, whether households are more willing
to borrow may depend on the extent to which stock market gains are perceived to be
permanent, and not the result of “irrational exuberance,” a phrase coined by former
Federal Reserve Board Chairman Alan Greenspan in 1996. More recently,
households apparently took on more mortgage debt than they might have otherwise,
in anticipation of continuing appreciation in house prices.
Income and wealth affect a household’s willingness and ability to take on
additional debt. But there are other important considerations as well. The burden a
debt places on a household’s finances is determined not just by the amount of the
debt, but also by the interest rate and the term of the loan. In the case of an
installment loan or a conventional mortgage, the payments are fixed at the time of the
loan and, unless the loan is refinanced, will not change over the term of the loan. In
the case of revolving credit, or a credit line, the burden of any borrowing depends on
the interest rate at the time.
Perhaps a better measure of the burden of debt on households is the ratio of
required payments on that debt relative to the income available to make those
payments. The staff of the Board of Governors of the Federal Reserve System
publishes estimates of the burden of debt service payments on households. The
measures are referred to as the debt service ratio and the financial obligations ratio.¹⁶
The debt service ratio is a measure of the minimum, or required, debt payments
relative to income. The financial obligations ratio is a measure of total recurring
obligations, whether it be debt service or auto lease payments, rent, homeowners
insurance and property taxes, all relative to income. Including rent recognizes that
just as changes in interest rates and the size of mortgage loans can affect a
homeowner’s liquidity, changes in rent may affect the ability of renters to take on
additional debt. Figure 4 shows the behavior of these two ratios since 1980.

¹⁶ Karen Dynan, et al., “Recent Changes to a Measure of U.S. Household Debt Service,”
Federal Reserve Bulletin, October 2003, pp. 417-426.

Figure 4. Debt Service and Financial Obligation Ratios

^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
These measures of payments relative to income indicate that for a brief period
following the economic contraction of 1990-1991, household financial obligations
tended to decline relative to income. They also show that since then, obligations
have been rising steadily. As was the case with the debt-to-asset ratios, these
measures are higher now than they have been for most of the period shown. It also
seems clear that most of the variations in total minimum payments relative to income
have been due to changes in debt service. According to the Office of Federal
Housing Enterprise Oversight (OFHEO) house prices increased by about 50%
between 2001 and the first quarter of 2008. Over the same period, however, the ratio
of the value of household real estate to mortgage debt fell, according to data
published by the Federal Reserve Board.
Financial obligations ratios are available separately for renters and homeowners.
These data are shown in Figure 5. The ratio for renters is significantly higher than
it is for homeowners. Renters tend to spend a higher share of their income on both
housing and consumer debt payments. Between the early 1990s and 2001, the ratio
for renters rose much more rapidly than the homeowner ratio. One reason for that

was that renters’ incomes increased much more slowly than homeowners’ incomes.¹⁷
Since then, the gap between the two groups has narrowed.
Figure 5. Homeowner and Renter Financial Obligation Ratios

^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
Consumer debt comes in two forms, revolving and non-revolving. Non-
revolving debt includes fixed-term loans such as those for automobiles. Revolving
debt includes credit card debt and other lines of credit. Figure 6 shows the
components of consumer debt as a percentage of aftertax income.
¹⁷ Ibid. It may be worth noting that renters are typically younger than homeowners. Data
from the 2000 Census show that 40% of renter householders were under the age of 35. That
figure for homeowners was 13%. In 2006 the median income of renters was about half that
of homeowners.

Figure 6. Consumer Debt as a Percentage of Aftertax Income

^Sources^:^D^e^part^m^eⁿ^t^of^C^o^m^m^e^{rce, Bu}^reau^of^Ecoⁿ^o^mⁱ^c^Aⁿ^a^l^y^sⁱ^s^;^{Board of}^G^o^v^e^rn^ors^of^t^h^e
^F^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
Non-revolving debt fell between the mid-1980s and the early 1990s. That
reflected both a decline in interest rates, and an increase in the length of maturity of
automobile loans. Since then, non-revolving debt has risen steadily. Revolving debt,
until 2001, had been rising relative to income. Since then, that ratio has fallen
slightly. Some of the increase in revolving debt in the 1980s and 1990s was due, in
part, to the increased number of credit cards in circulation. The increase in revolving
debt is also attributable to the increased use of credit cards as a substitute for cash.
Although it has declined slightly over the past few years, total consumer debt relative
to income is still higher than it has been over much of the past 25 years.
A study by the Federal Reserve found an additional reason for a rise in credit
card payments relative to income.¹⁸ Credit scoring has become widespread and
allows lenders to charge interest for consumer loans based on the risk characteristics
of each borrower. That makes it possible to offer loans to relatively higher risk
borrowers, typically those at the lower end of the income distribution, that might not
have had access to credit previously. Relatively less risky borrowers may, at the
¹⁸ Kathleen W. Johnson, “Recent Developments in the Credit Card Market and the Financial
Obligations Ratio,” Federal Reserve Bulletin, Autumn 2005, pp. 473-486.

same time, have experienced reduced interest costs, which, in theory, increased the
amount of credit they were able to afford.
The Demographics of Household Debt
The significance to the overall economy of a given level of household debt
varies depending on how that debt is distributed. For example, an increase in the
total amount of debt might signal an increase in the risk of defaults, or of a
prospective cutback in consumer spending that might lead to a slowdown in
economic growth. But if that increase in debt is attributable to increased borrowing
by relatively well-off households, it might be of less concern than if it were due to
increased borrowing by those households with fewer resources. Other things being
equal, an increase in the indebtedness of households might pose increased risks for
the economic outlook, but there is no one level of indebtedness above which an
economic downturn becomes “likely.”
Table 1 shows the percentage of families that have debt, of any kind, in selected¹⁹
income percentiles for 1998, 2001, and 2004. The data indicate that there is a
tendency for the share of families that have debt to rise with income. Table 1 does
not provide evidence of any significant trend over time, although the largest increase
in the proportion of households with any kind of debt was in the lowest 20% of the
income distribution.
Table 1. Share of Households Holding Debt of Any Kind in 1998,
2001, and 2004, by Income Percentile
Income Percentile199820012004
All families74.175.176.4
less than 20%47.349.352.6
20% to 39.9%66.870.269.8
40% to 59.9%79.982.184.0
60% to 79.9%87.385.686.6
80% to 89.9%89.691.492.0
90% to 100%88.185.386.3
^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
The burden of debt, and the risks it may pose to households depend, however,
not on whether households hold debt, but on the cost of financing that debt relative
to their financial resources. Table 2 presents the ratio of debt payments to income,
by income percentile. These statistics are a cross section of the data shown in the

¹⁹ These data are from the Survey of Consumer Finances, published by the Federal Reserve
Board. This is a triennial survey. Data for 2007 may be available sometime in 2009.

lower line of Figure 4. The numbers suggest that debt burdens are similar across the
income distribution with the notable exception of those families in the top 10%.
Table 2. Ratio of Debt Payments to Income
Income 1989 1992 1995 1998 2001 2004
Percentile
All families12.614.414.114.912.714.4
less than 20%15.316.418.918.916.218.1
20% to 39.9%12.615.717.316.715.516.8
40% to 59.9%16.316.315.418.617.419.4
60% to 79.9%16.816.718.019.216.718.5
80% to 89.9%16.115.616.716.717.017.3
90% to 100% 8.111.4 9.610.5 7.8 9.5
^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.
These data indicate that in 2001, a year of economic contraction, the ratio of
debt payments to income fell for most income classes, but that overall the trend has
been for it to rise.
One measure of financial distress is the share of families at least sixty days late
in making debt payments. Table 3 shows these data, again by income percentile.
Not surprisingly, a larger percentage of families in the lower income classes have
been late making payments than have those families with higher incomes. It is also
noteworthy that more and more households in the middle of the distribution have had
trouble making payments on time.
Table 3. Families With Payment 60 Days or More Late
Income 1989 1992 1995 1998 2001 2004
Percentile
All families7.36.07.18.17.08.9
less than 20%17.911.09.913.113.316.5
20% to 39.9%12.39.310.712.711.613.5
40% to 59.9%4.96.98.79.58.110.2
60% to 79.9%6.04.46.45.83.97.2
80% to 89.9%1.11.82.83.82.62.3
90% to 100%2.41.01.01.81.30.3
^Source:^{Board of}^G^o^v^e^rn^ors^of^t^h^{e F}^e^deral^R^e^s^e^rv^{e S}^y^s^t^em^.

Household Debt as an Economic Indicator
The general concern expressed with rising levels of household debt is that at
some point this will lead to a reduction in consumer spending and thus initiate, or
magnify, a slowdown in the rate of economic growth. In theory, this could happen
in several different ways. Once above a given level of debt, any change in consumer
confidence about the near-term economic outlook may trigger a cutback in spending,
and an unwillingness to take on additional debt.
It may also be that above certain debt levels households are more sensitive to
fluctuations in interest rates. The debt burden measures discussed above depend on
the level of debt, the maturity of those debts, the interest cost of the debt and
household income. A change in any one of those variables can affect the related
burden of the debt.
The debt burden also depends on income. To the extent that income
expectations are realized, the debt burden may not be a problem. But an unexpected
drop in income, or in wealth for that matter, may raise the debt burden above what
households intended and motivate them to cut back on either any additional
borrowing they may have planned, or on their spending.
What for borrowers is a burden, is income for lenders. As long as borrowers are
able to meet their debt obligations, these payments are simply a transfer, or exchange,
of income. What may matter more than the level of debt is the use to which the
borrowed money is put. Borrowing may finance either consumption or investment
spending. In the case of consumption, the borrowing is a shift of resources from the
future to the present. In other words, a claim on future income finances an increase
in current consumption. Households that borrow to increase consumption spending
can enjoy a higher standard of living now than they otherwise could, but will be²⁰
worse off in the future if they must reduce their consumption to pay off their debt.
If a household makes its debt payments on time, then the shift in resources
involves only that household. It is consuming more now at the expense of its own
future consumption. But if that household defaults on its debt, then its consumption
will have risen at the expense of the lender’s future consumption. In this case, the
lender may be inclined to reduce his current consumption to make up for the loss in
wealth.
In cases where a household borrows to finance the purchase of durable goods
such as household furnishings or an automobile, payments on the debt usually
coincide with the life of the good, more or less, so that rather than shifting future
income to present uses, payments are made as the good is consumed (e.g., as the
automobile depreciates). If the good purchased still has value after the loan is paid

²⁰ There is also the possibility that an investment fails to yield enough income, or utility, to
offset its finance costs, or that its price falls below the amount borrowed to pay for it.

off then to that extent the payments actually represented saving.²¹ If households
default on this type of loan, then the lender may be able to cover his losses to the
extent of the remaining value of the good, and the transfer of resources is limited.
In the case of borrowing to finance the purchase of a home, a large component
of the debt payments represents saving. By paying off mortgage debt, households
acquire an asset. If borrowers default on mortgage debt, the lender is protected from
loss by the value of the asset, and the borrower may even have accumulated some
equity in the house; thus, there is little transfer of wealth either from borrower to
lender or lender to borrower. As with all loan defaults, however, there may be
significant administrative costs. To the extent that the borrower has built up equity
in the house, there may have been a shift in resources from the present to the future.²²
This is saving.
Saving is defined as income less consumption. If households borrow heavily
to increase consumption spending they may have a negative saving rate. In fact, for
the past several years the average U.S. household saving rate has been very close to
zero, meaning that, in the aggregate, households are consuming nearly all of their
current income.
In some respects, the saving rate may be a more meaningful economic measure
than debt.²³ An important determinant of the long run rate of economic growth is the
national rate of saving and investment. The more out of current income that is saved,
the more there is available to invest and add to the domestic stock of capital. A
larger capital stock makes workers more productive and raises incomes and living
standards. To some extent growth of debt represents dis-saving. Where saving is a
shift of current income to future uses, accumulation of debt, at least some of it,
represents a shift of future income to current uses. But not all household debt has
that effect. If households increase their borrowing to finance additional consumption,
it will be reflected as a decline in the saving rate. If instead, they reduce their
outstanding consumer loans, that will be reflected as an increase in the saving rate.
If households increase borrowing to finance the purchase of a home, that will also be
reflected as an increase in the saving rate.²⁴ Changes in the saving rate are thus

²¹ There are two sources for saving data in the United States. The Department of Commerce,
Bureau of Economic Analysis publishes the National Income and Product Accounts, and the
Board of Governors of the Federal Reserve Board publishes the Flow of Funds Accounts.
In the Flow of Fund Accounts measure of saving, automobile purchases by households are
counted as investment and hence are also reflected in household saving. In the National
Income and Product Accounts, only business purchases of automobiles are counted as
investment.
²² In some cases, households may not have accumulated enough equity to cover the costs of
foreclosure. In cases where households are at risk of foreclosure, they may sell the house
themselves and pay off the loan in order to avoid the additional costs associated with
foreclosure.
²³ For the most recent saving rate data, see CRS Report RS21480, Saving Rates in the United
States: Calculation and Comparison, by Brian W. Cashell.
²⁴ In the case of automobile purchases, only the Flow of Funds measure will reflect the
(continued...)

probably a better measure of these intertemporal shifts than are changes in
outstanding debt.
Are measures of household debt helpful in assessing the economic outlook?
Intuitively, it might seem that households’ sensitivity to interest rate changes would
increase with their debt load. An increase in interest rates would normally raise their
debt service payments and induce them to curtail other spending.
Most studies of the relationship between household debt and economic growth
suggest that rising debt has not been a threat to economic growth.²⁵ Instead of being
a harbinger of economic hard times, rising household debt has been found to be
associated with a growing economy. Changes in consumer debt tend to be a leading
indicator of consumer spending, and thus of overall economic growth. One reason
for this may be that increases in consumer borrowing are an indication of confidence
in the economy, both on the part of borrowers and lenders.²⁶

²⁴ (...continued)
increase in saving.
²⁵ See C. Alan Garner, “Can Measures of the Consumer Debt Burden Reliably Predict an
Economic Downturn?,” Federal Reserve Bank of Kansas City, Economic Review, Fourth
Quarter 1996, pp. 63-76.
²⁶ See Dean M. Maki, The Growth of Consumer Credit and the Household Debt Service
Burden, Board of Governors of the Federal Reserve System, February 2000, 26 pp.