Printer Friendly

Analyzing consumer spending and debt.

Analyzing Consumer Spending and Debt

THE OFFICIAL TROUGH in the last recession was November, 1982. However, months before that date, consumers had already indicated that they would be the first "spenders" out of the gate, initializing what would become an expansion of record length. Followers of consumer surveys, such as those conducted regularly by the Survey Research Center at the University of Michigan, were "told" by the survey results that the economy was about to turn around.

This paper reviews the recent performance of the Michigan survey data as leading indicators of changes in consumer spending, saving and debt use. A methodology is presented that utilizes the underlying cross-section data to develop insight into the recent growth in consumer debt and to evaluate its "quality" (e.g., the ability of consumers to repay the obligations owed).


The Survey Research Center at the University of Michigan has been polling consumers on economic issues since the pioneering work of George Katona in the 1950s. Katona developed the now-famous Index of Consumer Sentiment, a composite of ten questions that now have been incorporated into the government's set of leading indicators. Figure 1 shows the history of the Index since 1978. The Index clearly signaled the coming problems of 1980-82 and turned up quickly as the recovery began. At historic record-high levels in 1984, the Index has moved in a very narrow but high range since the beginning of the recovery. Although a modest downward trend appears in the data, no indication is apparent of pending economic problems in the near term such as that preceding 1980.

With so many factors affecting each individuals financial well-being, they are simply asked if they are better off financially than twelve months ago and whether or not they expect to be better off financially twelve months after the survey. Figure 2 shows the net(1) percent reporting that they are better off financially and the net percent of families expecting to be better off financially twelve months later. Once again, these two series (included in the Index) showed substantial distress prior to the 1980-82 period, and show only the slightest hints of any problem now.

Consumers' opinions about the course of the economy exhibit similar patterns (Figure 3). The net percent of families reporting that the economy is better now than twelve months ago is rising. However, the net percent expecting the economy to be better in twelve months has been falling. This question asks for a prediction of the change in the economy. It is not surprising that, as the expansion gets longer and longer, and employment higher and higher, fewer and fewer people feel that the economy can get any better. However, the net percentage expecting "good times" in the economy is rising, suggesting that although improvement is not seen in the cards, more consumers expect that we can hold the level of prosperity attained for at least twelve months more.

Deteriorating economic conditions, high interest rates, and inflation clearly led consumers to feel that the 1978-82 period was not a good time to be buying cars, durables and houses (Figure 4). This attitude began to change in 1983, with the proportions of families reporting that "now is a good time to buy" rising to record high levels for houses and durables. The auto question is sensitive to the presence of "dealer incentive plans", and housing, not surprisingly, is sensitive to swings in mortgage rates. Although the assessment of current buying conditions has started to slip, all three indexes remain at very strong levels. Buying in advance of anticipated price increases was popular in the late 1970s, and shows only modest signs of reviving (Figure 5). The current position indicates little concern for a resurgence of inflation in the near term on the part of most consumers.

High interest rates thoroughly discourage house buying (Figure 6, right scale). Nearly 20 percent of all families are now concerned that rates are too high to make buying a home a good idea. Interest rate variations have far less impact on autos and even less on durables (Figure 6, left scale). All three series indicate a modest rise in concern about higher interest rates and the associated postponement of purchases.

Figure 7 makes it clear that consumers are slipping out of their "buying mood," expressing increasing reluctance to use savings or incur new debt to finance discretionary spending. Overall, the data indicate that the consumer is still prepared to provide floor support for the expansion, at least through the end of 1989. Although the consumer mood could change precipitously in response to a real "shock," it now appears that optimism is fading only slightly and at a slow pace. Spending can be anticipated to stay the course.


Since 1975, consumer credit (excluding mortgage obligations) has grown from $160 billion to nearly $670 billion, an increase of over 300 percent. More spectacular has been the rise in revolving credit obligations, increasing from $14 billion in 1975 to $180 billion, a gain of over 1,200 percent in less than fifteen years. Aggregate debt growth was obviously driven by inflation (the same car costs more to buy), rising real aggregate income (more purchases, more to finance) and rising employment. (Consumers with their first job typically wish to buy more than current income receipts can finance. They use credit and borrow against expected future earnings.)

Record-high employment, expressed as a percentage of those over sixteen years of age, has contributed substantially to the growth in aggregate debt. The most important qualifying characteristic for credit use is "employment" - evidence of the ability and capacity to repay the amouts borrowed. With record employment, creditors are seeing record numbers of applicants. Even in the absence of "aggressive" (lower) credit standards, the number of consumers qualifying for credit would be very high in this economic environment. Aggressive marketing strategies have further swelled the number and percentage of acceptances, producing unusually large numbers of new credit users.

The Ratio of Instalment Debt to

Disposable Income

The ratio of debt owed to consumer disposable income has also risen dramatically, reaching nearly 19 percent in 1986 and stabilizing at that historically record-high level. This increase cannot be explained by inflation, which affects both the numerator and the denominator of this popular analytical ratio. Nor does rapid employment growth account for this dramatic rise, because new employees add debt to the numerator and income to the denominator.(2)

A number of very important regulatory and behavioral changes have permanently increased this ratio:

1. Revolving credit has grown to nearly 30 percent of

total consumer credit. However, more than half of

these balances are repaid each month with no finance

charge paid, i.e., this is not credit in the eyes

of the user, but rather an economical substitution

of the use of credit cards for cash and writing


2. Deregulation has made it possible for lenders to

serve economically many customers previously denied

credit due to low legal rate ceilings. With

higher rates, more people qualify for credit, adding

debt to the numerator but not adding income to

the denominator. Aggressive marketing of credit

has a similar impact on the ratio.

3. The incidence of credit use has risen, especially

among consumers in the top 20 percent of the income

distribution. Again, this adds credit to the

numerator while leaving the denominator unchanged.

4. With high rates of interest available on savings,

consumers are less willing to use savings to make

purchases, using debt instead. This shift also raises

the numerator while not impacting the denominator,

but keeps the "quick ratio" for debtors at

healthy levels.

Revolving Credit

The growth in the use of revolving credit has been dramatic. The 1,200 percent growth in revolving credit outstandings since 1975 can be attributed primarily to the attractiveness of this new payment device. In fact, the credit card is really a "financial service" card, providing safety, identification, bill consolidation, reduced check fees, less time-consuming transactions, good records, access to financial service systems and, if needed, credit with very flexible terms and no application. Indeed more than one-fourth of the retail credit outstanding balances and half of the bank card outstanding balances are repaid each month with no finance charge - in short, a substitute for money (cash and checks). Since 1975, the bank credit card share of revolving credit has risen from less than half to nearly 65 percent, while retail revolving credit balances have risen in step with sales.

However, all of these balances are included in the consumer credit statistics. Although these balances are "at risk", in a behavioral sense, the balances that do not constitute true extensions of credit should be excluded from the consumer credit statistics if the ratio of debt owed to disposable income is to be meaningfully compared to its past values. If half of bank card outstandings and 25 percent of all other revolving credit obligations were not counted, the credit figure would be lowered by approximately $75 billion and the ratio of debt to disposable income lowered from 18.6 percent to 6.5 percent, a number quite in line with historical experience.

Cross-Section Evidence on Consumer

Credit Use and Quality

Analysis of cross-section studies spanning twenty years indicates that consumer credit use on an individual basis is very stable. Although recessions and expansions have some impact on the statistics, the ratio of debt to income remained basically unchanged between 1963 and 1983. A substantial increase occurred in the percentage of consumers using debt in each income quintile, particularly among consumers in the top end of the income distribution where usage rose 13 percentage points, or 25 percent (Table 1). The dollar amounts of credit involved are substantially higher for this group as well.

Table : Table 1 Percent of Households with Instalment Debt and Average Ratio of Outstanding Debt to Income(1)
 Approximate 1967 1970 1977 1983(2)
Income Quintile % Avg % Avg % Avg % Avg
Lowest 25% .24 23% .22 31% .30 34% .25
Second 43 .22 47 .18 49 .18 56 .16
Third 59 .17 61 .16 64 .16 61 .15
Fourth 60 .15 67 .11 66 .13 67 .15
Highest 53 .12 47 .10 58 .11 69 .12
All Households 48% .17 49% .15 50% .16 57% .16

(1)For households with instalment debt, individual debt to before-tax income ratios were computed and the mean and median of these taken. The ratio of liquid asset holdings to outstanding debt was computed for each household with debt and the median of these was also taken. (2)Contrary to earlier data, the 1983 measure of consumer instalment debt includes all reported credit card debt. Source: 1960, 1965, 1967, 1970 Surveys of Consumer Finances, University

of Michigan Survey Research Center; 1977 Consumer

Credit Survey and 1983 Survey of Consumer Finances, Board

of Governors of the Federal Reserve System.

Table 2 presents approximate shares of credit owned by the five income groups. The highest income consumers added ten points to their share during the twenty year period studied. More recent data just released by the Federal Reserve apparently confirm these findings.(5,9) [Tabular Data Omitted]

From these data, the total change in real debt during the twenty-year interval can be calculated. Then, by using appropriate combinations of the 1963 and 1983 data, the impact of changes in various factors can be calculated. The total growth in instalment debt for the entire sample and for specified age groups is calculated as follows:

[Households.sub.83] x Proportion with [Debt.sub.83] x [Avg. Real Debt.sub.83] - [Households.sub.63] x Proportion with [Debt.sub.63] x Avg. Real [Debt.sub.63] = Total Change in Instalment Debt From 1963 to 1983

As shown in Table 3, this computation produces a total change of $113 million (in real terms) during the period. Although this figure is not easily reconciled with the aggregate debt statistics, it does represent a consistent base from which estimates of the contributions of various factors to debt growth can be computed. The contribution of each factor will be expressed in terms of the percentage of the total amount of debt growth that can be attributed to it.

Table : Table 3

Instalment Debt Growth Between 1963 & 1983
 (Real 1982 dollars - Millions)
 Age Consumer Debt Total % Percent
Distribution 1963 1983 Change Change of Change
Under 25 $ 7066 $ 11773 $ 4708 67% 4%
25 to 34 15945 54462 38523 242 34
35 to 44 19261 49700 30439 158 27
45 to 54 18181 36448 18268 100 16
55 to 64 8426 27317 18892 224 17
65 and over 1776 3953 2178 123 2
 ALL $70655 $183653 $113003 160% 100%

Using this framework, it is possible to examine the "partial" effect of a number of factors impacting on the credit statistics, including: (1) a higher frequency of debt use ("widening"); (2) a more intensive use of debt leading to higher individual debt-to-income ratios ("deepening"); (3) a shift in the age distribution; (4) population growth; (5) income growth, and even inflation, although calculations done here are in constant dollars.

To estimate the impact of a rise in the propensity to use debt, other factors held constant, the percentage of each age group using debt is allowed to change. Except for those under the age of twenty-five, more consumers in each age group became debt users. The overall proportion of households who use consumer credit also increased by 8 percentage points. But, it is assumed that the new credit users behave like the existing users in each age group. Applying the change in the propensity to use debt to the 1963 credit data indicates that approximately 13 percent of the total growth in credit was due to a "widening" of debt use (Table 4).(3) [Tabular Data Omitted]

Making a series of similar assumptions, the following approximate "causal" shares were assigned to the various factors affecting the growth in real consumer credit outstanding between 1963 and 1983:
 Selected Changes Affecting Credit Share
 A higher proportion of credit users 13%
 More leverage 7
 Population growth 33
 Higher real per capita income 22
 Changes in the age distribution -1

These calculations are only approximate and not exhaustive, but they do provide useful estimates of the relative impact of various factors identified as contributing to the rapid growth in debt. [1,2] More recent analyses conducted by the Board of Governors of the Federal Reserve confirm that these relationships remained stable through 1985, and there is little reason to expect that they would have changed in the past three or four years.[5,8,9]


If the historically established "leads" hold up, the Michigan survey data indicate that consumers are prepared to see the economy through the end of 1989 with no major disruptions. Housing will weaken, as fewer consumers feel that now is a good time to buy a house. The main concern is high interest rates, although the demographics are beginning to move adversely for housing. Only the use of variable rate contracts, which have kept average effective mortgage rates roughly constant for the past twelve months, has kept the housing market from softening even further.

Consumer credit is often analyzed independently of the spending that generates debt in the consumer portfolio. In this sense, the most important fact to keep in mind is that consumers incur debt in order to buy goods and services. Typically, when debt use is high, spending is high as well (and savings as defined by the National Income accounts is low because durable goods are assumed "consumed" in the period purchased). A large stock of debt usually occurs with a large stock of durables. The more consumers have accumulated (the larger the stock of durables in the consumer portfolio), the less likely they are to buy more. However, they will continue to pay down the obligations they have incurred while taking on much less new debt. This development rebuilds liquidity and rebalances the consumer portfolio.(4)

Consumers have been on a "spending spree" for seven years. They are comfortably "stocked up". New spending can be easily curtailed without seriously compromising living standards, because existing stocks of goods will continue to provide acceptable levels of service. This "discretion", in combination with the fact that employment is at record-high levels, makes consumer spending very vulnerable to bad economic news. Declines in employment will leave many debtors with no income to support their credit obligations. Any shock to the economic system will be magnified by these types of problems.(5)

Consumers are becoming more reluctant to use savings to make major purchases or to incur new debt. This factor translates into a slower spending and slowdown in the rate of credit growth. Only unexpectedly high employment gains could offset the pending gradual slowing in consumption.


(1)The "net percent" is calculated as the percentage of families giving a "favorable" answer to the question less the percentage giving an "unfavorable" answer.

(2)Unless one assumes or has evidence that the rate at which new hires add to the numerator is sufficiently higher than the rate at which their jobs raise the income figure in the denominator.

(3)The total change in debt calculated in Table 4 due to changes in the frequency of debt using holding the population, age distribution and debt-to-income ratios constant at 1963 levels was divided by the estimated total change in debt from Table 3.

(4)During the 1980-82 period, consumers reduced spending and repaid debt, driving the ratio of debt to disposable income down to levels observed in the 1960s. When the recovery began, these holdings of liquid assets were deemed excessive for improving economic conditions. "Excess" holdings of liquid assets were subsequently converted into durable goods in the portfolio.

(5)Recall that interest payments are consuming a huge share of corporate cash flow and that many debt-dependent mergers and acquisitions have occurred when the economy is at or near its peak for this expansion. Just as consumers who have jobs today may look creditworthy, so these firms may also encounter substantial difficulty in meeting credit obligations if the economy slows.


[1.] Dunkelberg, W. and Worden, D., "The Quality of

Consumer Credit", unpublished manuscript,

School of Business and Management, Temple

University, 1987. [2.] Dunkelberg, W. and Worden, D., "Consumer Credit

in the U.S.: Structural and Cyclical Determinants",

unpublished manuscript, School of Business

and Management, Temple University, 1987. [3.] Dunkelberg, W.C. and others, Consumer Credit in

the 1970s, unpublished manuscript, August,

1980; available from the School of Business and

Management, Temple University. [4.] Enthoven, Alain, "The Growth of Instalment Credit

and the Future of Prosperity", American Economic

Review, December, 1957; pp. 913-29. [5.] Luckett, Charles A. and J.D. August, "The Growth

of Consumer Debt," Federal Reserve Bulletin 71

(June 1985): 389-402. [6.] Pearce, Douglas K., "Rising Household Debt in Perspective,"

Federal Reserve Bank of Kansas City

Economic Review, (July/August 1985): 3-17. [7.] Steinberg, Edward I., "Consumer Credit, 1960-80,"

Survey of Current Business, 61 (February 1981):

14-18. [8.] Tapscott, Tracy R., "Consumer Installment Credit,

1980-1985," Survey of Current Business, (August

1985): 12-16. [9.] Wilson, J.F., E.M. Fogler, J.L. Freund, and G.E.

van der Ven, "Major Borrowing and Lending

Trends in the U.S. Economy, 1981-1985," Federal

Reserve Bulletin, (August 1986): 511-524.

PHOTO : Figure 1 Index of Consumer Sentiment (University of Michigan)

PHOTO : Figure 2 Personal Financial Situation (% "Better" - % "Worse")

PHOTO : Figure 3 Outlook for the Economy (% Favorable - % Unfavorable)

PHOTO : Figure 4 Buying Conditions (Net % "Good Time" Univ. of Mich)

PHOTO : Figure 5 Expected Prices and Buying Conditions (Good Time - Prices Will Rise)

PHOTO : Figure 6 Interest Rates and Buying Conditions (Bad Time - Rates High)

PHOTO : Figure 7 Opinions About Use of Savings & Credit (OK to Use for Major Purchases)

(*)William C. Dunkelberg is Dean, School of Business and Management, Temple University, Philadelphia, PA.
COPYRIGHT 1989 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Dunkelberg, William C.
Publication:Business Economics
Date:Jul 1, 1989
Previous Article:The saving solution.
Next Article:Consumers to play a supportive, not a leading role.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters