Printer Friendly

Payment of household debts.

The indebtedness of U.S. households grew substantially in the decade that ended last year. The sum of home mortgages and consumer debt outstanding rose from $1.3 trillion at the end of 1980 to just under $3.4 trillion at year-end 1990. This increase averaged out to a rate of 10 percent per year, which was one-third again as large as the average growth in after-tax income over the same period.

The more rapid rise of debt than of income during the 1980s has raised concerns about the ability of individuals to meet their debt obligations comfortably. The recent slowdown in economic activity has intensified those concerns. Debt repayment problems can impair the profitability of lending institutions by generating higher loan-loss provisions and collection costs. Historically, few institutions have been seriously endangered by defaults on their portfolios of consumer or home mortgage loans, but lower profits in those areas could aggravate other problems. In a broader sense, higher delinquency and default rates could provoke a tightening of credit standards, curtailing the supply of credit and thereby damping consumption and housing activity. On the demand side, higher debt burdens could act as a drag on consumption because scheduled debt payments represent a competing claim on current income.

Various statistics on debt growth, personal bankruptcies, and loan delinquency rates provide some aggregate information on the extent to which households may be experiencing debt payment problems. These statistics, reviewed in the first part of this article, provide useful insights but have several shortcomings as well. Also, they contain no information about the concentration of debt problems among borrowers or about the individual characteristics of troubled debtors that can shed light on the extent and severity of repayment problems. To improve understanding of the financial condition of households, the second part of this article examines data on late payments by individual borrowers obtained from surveys of households conducted for the Federal Reserve by the Survey Research Center at the University of Michigan.


The financial health of the household sector is often gauged in rough fashion by comparing the total debts of the sector to its aggregate income or asset holdings. In addition, data on the proportion of loans that have payments past due provide more direct measures of repayment problems, while personal bankruptcy statistics indicate the frequency of the most severe problems-those in which debtors find the situation so extreme as to seek resolution from the courts.

Debt-burden Measures

The two major categories of household sector debt are home mortgage debt, which amounted to about $2.6 trillion at the end of 1990, and so-called consumer debt, which totaled about $800 billion at year-end. Consumer debt has two subcategories: installment (repayable in two or more payments) and noninstallment debt, with the installment component accounting for 90 percent of the total.

Loans are categorized as mortgages or as consumer debt more by their collateral than by the purpose of the loan. The mortgage debt of individuals consists of all borrowing secured by a lien on a home, including various types of "home equity loans" that are often used for purposes other than housing expenditure. (1) Consumer debt is often unsecured, like credit card debt and personal cash lending, or it may be collateralized by the goods it is used to finance, such as motor vehicles or household durables. While consumer credit is used predominately to purchase currently produced goods and services, no doubt it also finances some nonconsumption activities as well, like financial investments.

Ratios of household debts relative to income flows or asset holdings provide some indication of the burden that debts place on the resources available for

repaying them. Interpretation of such ratios is beset by many complexities, however. Perhaps the most frequently cited debt burden measure is the debt-to-income ratio; the first problem encountered in its use is deciding which types of debt to include in it. When the broadest measure of debt is used-mortgage plus total consumer debt-the debt-to-income ratio (with disposable personal income as the denominator) now exceeds 80 percent (table 1). With mortgage and consumer debt both expanding more rapidly than income during most of the 1980s, this ratio rose more than 15 percentage points between 1980 and 1990. On its face, this rise seems to indicate a substantial increase in the burden of debt, but that conclusion is not necessarily warranted.

Home mortgage debt, as noted, accounts for the largest part of the broad debt measure, and interpretation of a mortgage debt-to-income ratio is particularly murky. For instance, aside from those who own their home free and clear, virtually every household must make a regular outlay associated with their housing-either a mortgage payment or a rent assessment. The analytical usefulness of a debt-burden measure, including as it does one form of housing outlay (a mortgage) but not the other (rent obligation), seems at least questionable. An increase in the homeownership rate would boost the level of debt and thus the calculated debt burden (while reducing the amount of rent). Because of this offsetting but unmeasured decline in rent burden, interpreting a rise in debt burden attributable to an increase in homeownership rates in the same manner as a rise in debt burden from other causes seems inappropriate. In addition, home mortgages carry such lengthy maturities that only a small portion of mortgage debt comes due in a given year. The significance of including the entirety of mortgage debt outstanding in a measure of burden thus seems questionable too, especially if one's analytical interest is the constraint imposed by debt obligations on near-term consumption.

For such reasons, the debt-to-income ratio often excludes mortgage debt. The ratio frequently leaves out noninstallment debt too, in part because much of it is very short term in nature, such as charge accounts on which full payment is required by the due date or "bridge loans" used to facilitate real estate transactions. Noninstallment debt is a small component, however, and its inclusion or exclusion makes little practical difference.

In any case, the most commonly used debt-to-income measure has only consumer installment credit in its numerator. The surge of installment borrowing in the mid-1980s raised this measure of debt burden from about 14 percent at the outset of the expansion to a high of 19 percent late in 1989 (chart 1). This increase is often cited as evidence that consumers on the whole have become overextended, but any such conclusion must be tempered by several caveats. First, as with mortgage debt, only a part of the outstanding stock of installment debt is payable in the near term, making it a less-than-perfect indicator of current payment burden. This deficiency is not as acute for installment debt as for mortgage debt because installment loan maturities are much shorter than mortgage maturities, but it nevertheless hinders interpretation. Moreover, a steady lengthening of consumer loan maturities in recent years taints comparisons with earlier periods. (2)

Second, the installment-based measure, in common with other versions of the debt-to-income ratio, contains no information about the distribution of debt across households of differing income and asset profiles. The aggregate ratio thus gives no indication of whether a given buildup of debt is owed by people with the income, assets, or employment prospects to handle it comfortably, or by people of more limited resources who might be more vulnerable to payment difficulties. In addition, shifts in the content of consumer credit over time-such as the growing use of credit cards for "convenience credit"-make longer-term comparisons difficult, just as do changes in the maturity structure of consumer debt. (3) Finally, substitutions between consumer installment debt and other types of borrowing would also distort historical comparisons for any measure restricted to installment debt. The recent popularity of home equity lines of credit exemplifies such a substitution, and this substitution may have curtailed growth of installment debt.

So far, the discussion of debt-to-income measures has focused on outstanding debt rather than on debt payments. For many analytical purposes, a debt-service measure reflecting scheduled (or required minimum) payments of principal and interest would be a more relevant concept. Unfortunately, no comprehensive data series of repayment flows currently exists. Estimates can be made using data on outstanding debt, average maturities, and interest rates, but this approach requires making several arbitrary assumptions to fill data gaps and to supply certain behavioral parameters regarding the refinancing or prepayment of loans. Such estimates of repayments have limited analytical usefulness because of their indeterminate accuracy. Information on scheduled payments of debt was collected in the household surveys of late payments, however, enabling some discussion later in this article on the linkages between debt payment burden and late or missed payments.

Another caveat about debt-to-income ratios is that income is not the only source of funds from which households can make debt payments. Recognizing this fact, some observers also look at debt relative to the financial assets of households. At the aggregate level, assets of households are about four times the size of their liabilities, a ratio that declined only marginally during the 1980s. The net worth of the household sector-the difference between assets and liabilities-has grown from about $6 trillion in 1980 to $10 trillion at year-end 1990. Again, however, it must be recognized that the aggregate statistics relating assets and debts may be masking important distributional features that might alter-either favorably or unfavorably-an assessment of the household sector balance sheet.

Loan Delinquency Rates

The proportion of loans with payments past due offers another slant on the extent to which individuals may be overburdened with debt. A few such aggregate data series are available based on surveys of lending institutions. However, these aggregate delinquency rates provide no sense of whether those consumers with late payment problems form a small group of people who are chronic late payers or a larger group of individuals with transitory problems. The most widely referenced series are those produced by the American Bankers Association (for consumer loans at banks) and the Mortgage Bankers Association (for home mortgages at a cross-section of loan servicers).

Delinquency rates on consumer loans, as measured by the American Bankers Association (ABA), exhibit a pronounced cyclical pattern. The ABA series (shown in the top panel of chart 2) is a weighted average of eight separate series for specific types of loans, such as direct and indirect auto loans, personal loans, and loans for mobile homes and other products. Historically, this series has begun to rise a few months in advance of a recession period, has peaked some time during the recession, then has declined steadily into the subsequent economic recovery. (5)

Such a pattern of cyclicality is consistent with expectations. As economic activity slows approaching a recession, an increasing number of workers are being laid off, or their working hours are being reduced. It would not be surprising if the number of households falling behind on debt payments rose too. Also, with new borrowing typically curtailed at such times, the denominator of the delinquency rate increases more slowly, providing less of a downward pull on the rate. (Early in an expansion period, when borrowing is brisk, the calculated delinquency rate is held down by the large volume of loans coming on stream, very few of which become delinquent immediately.) Delinquencies peak and then begin to decline at some point during a recession for a number of reasons: As a recession unfolds, many borrowers have already begun taking steps to get their finances in order, and lenders have been applying stricter lending standards, resulting in loan portfolios of improving average quality. Moreover, as lenders write off uncollectable loans, removing them from portfolios, such loans cease to affect the delinquency rate. After several months, these various factors combine to drive the delinquency rate back down.

Mortgage delinquencies have generally risen during recessions as well, but their pattern appears tied much less directly to the business cycle. For instance, the proportion of home mortgages past due (Mortgage Bankers Association series in the bottom panel of chart 2) continued to rise for almost four years after the end of the 1981-82 recession. Over the next four years, however, mortgage delinquencies declined steadily, in contrast to the moderate uptrend in consumer loans past due. The relatively quick response of consumer delinquencies in contrast to the more delayed response of mortgage delinquencies to shifts in economic activity partly reflects the considerable differences in loan maturities for the two types of debt-consumer loans turn over much faster. If relaxed lending standards during expansion periods contribute to later increases in delinquency, then faster loan turnover would imply more rapid liquidation of risky loans. For example, the bulk of consumer loans made in 1978 would have been substantially paid down by 1981, and almost all would have been liquidated by 1983. In contrast, many mortgage loans made in 1978 would still have twenty years or more of scheduled life remaining in 1983. Thus any lower-quality mortgage loans that may have been extended in 1978 (under the presumably liberal credit standards of that year), were probably still on lenders' books in the mid-1980s, while any lower-quality consumer loans would have been retired.

Home mortgage delinquencies may also reflect the amount of equity that borrowers have in their homes. When home prices have been rising sharply, homeowner equity is likely to be substantial, and people may be less inclined to be delinquent on their mortgages-at least for periods longer than thirty days. Because equity may be substantial even during recessions, depending on the strength of previous home price trends, aggregate delinquency rates on mortgages may exhibit only minimal changes in response to current economic activity. In the present economic downturn, which has been preceded by unusual weakness in house prices in many areas, the mortgage delinquency rate may exhibit greater sensitivity to income and employment developments. Through the fourth quarter of last year, however, mortgage delinquencies remained near ten-year lows.

Bankruptcy Statistics

Individual borrowers who lack any reasonable prospect of being able to repay their debts according to agreement have, in bankruptcy, a process for dissolving those debts, in full or in part, under court protection from collection efforts of creditors. Only a few types of obligations are ineligible for bankruptcy, such as federal tax obligations and child support payments, or any debt incurred by fraud or, in the judgement of the bankruptcy referee, in contemplation of declaring bankruptcy. (6)

Chapter 7 of the Bankruptcy Act provides for dissolving debts in full, and Chapter 13 provides for a partial repayment plan administered by the court-the so-called wage-earner plan. Under either type of bankruptcy, secured creditors are still permitted to recover collateral from bankrupts, and any assets above certain exemption levels are liquidated and distributed among the creditors. Historically, the social stigma attached to bankruptcy, the relatively limited asset exemptions under the laws of many states, and the difficulty of obtaining credit after bankruptcy all served to make this remedy truly one of last resort for troubled debtors. Even so, vigorous collection efforts by creditors, particularly through the garnishment of wages, often propelled people into bankruptcy court.

Limitations on garnishment and some other refinements of federal bankruptcy law in 1969 curtailed the use of bankruptcy in the 1970s. However, further major revisions to the law ten years later-especially the overriding of state asset exemptions with more liberal federal provisions-made bankruptcy a more attractive alternative for dealing with debt payment problems. In addition, though the development is hard to document, bankruptcy now seems to carry less of a stigma than it once did, and some evidence suggests that credit may be easier to obtain after bankruptcy than it once was. (7)

Historical trends in personal bankruptcies, shown adjusted for population growth in chart 3, reveal a prominent cyclical pattern. Most periods of sharply rising bankruptcy coincide with periods of economic contraction, although the extremely sharp rise in the expansion years of 1985 and 1986 and the continued strong increases over the next four years seem strikingly out of step with past patterns. On its face, this development suggests that the sizable expansion of debt in the 1980s has substantially weakened the financial condition of the household sector. However, because of the legal changes noted above that made bankruptcy a more attractive option to troubled debtors, the recent surge cannot be taken as an unqualified sign that the incidence of severe debt payment problems is mounting rapidly. The rise in bankruptcies may be reflecting to a significant degree a shift in the way households choose to respond to debt problems rather than an increase in debt problems per se.


The Federal Reserve Board has for many years sponsored surveys of consumers to gather information about their overall financial situation and about their use of specific financial services, including various types of debt instruments. As noted earlier, while some information is available regarding trends in consumer debt payments in the aggregate, less is known about the payment behavior of individual households. In particular, relatively little is known about how widespread or how severe payment problems may be among households. Also little documentation exists on what consumers do when they fall behind in their payments or what types of actions creditors pursue in cases of late payment. Surveys sponsored by the Federal Reserve in late 1990 and early 1991 provide some information on these questions.

Payment Behavior

The consumer surveys indicate that, overall, 85 percent of all households had an outstanding debt obligation or access to a line of credit under a credit card plan when interviewed (table 2). (8) Among all households, only 3 percent had home mortgages exclusively; 45 percent had only consumer credit, and 38 percent had both outstanding mortgage and consumer debt.

The vast majority of indebted households reported no problems meeting their debt payment obligations on time during the twelve months preceding the survey (table 3). (9) Specifically, 86 percent of the indebted households reported that they met or exceeded all of their scheduled debt payment obligations. As indicated by the survey, 42 percent of these households made payments that were larger or more frequent than scheduled on at least one of their outstanding obligations. These households may have been attempting to retire their debts ahead of schedule or, in some cases, the larger payments may have involved a refinancing.

Among the various types of debt owed by consumers, "other mortgages," primarily home equity loans, had the best overall payment performance. This finding is consistent with the extremely low delinquency rates reported by commercial banks on their outstanding home equity lines of credit and, to a lesser degree, on their other types of home equity loans.

Despite the overall satisfactory payment performance of the majority of indebted households, a significant minority, about 14 percent of those surveyed, reported falling behind on at least one of their scheduled debt payments. A small fraction of these late payers-roughly 12 percent-reported both that they fell behind in their payments and that, at some point during the year, they made larger or more frequent payments than scheduled. In some cases, consumers may have done so to catch up on earlier missed payments.

Households reported falling behind in their payments most frequently on vehicle loans and other types of non-credit-card installment debt (table 3). These two categories, along with credit cards, also had the highest incidence of consumers who reported falling more than thirty days behind in their payments. Of the households reporting payment problems, relatively few with home mortgages or vehicle loans let these debts get as much as sixty days in arrears (table 3). Delinquency of this duration is more likely to raise the possibility of foreclosure or repossession, actions most consumers would like to avoid.

Characteristics of Consumers with Payment Problems

Not all types of households are equally likely to report falling behind on their payment obligations. Consumer survey data provide an opportunity to profile the household characteristics (such as income, marital status, and age) that may be associated with late payment problems. The relationships observed, however, do not directly reflect the creditworthiness of persons with given characteristics because the debtor segment of the population has already passed through a credit-screening process designed to weed out the riskiest applicants. To be precise, the data can show which factors are associated with missed payments, given the credit standards prevailing in the marketplace, but not which factors, before the fact, are associated with default risk. (10)

Strong correlations exist between payment problems and housing tenure, marital status, and debt-service burdens (table 4). (11) Differences in payment behavior also show up with respect to household income and age, although the differences are greatest at the extremes of the household groups. The combined results for debtors in the lowest two income quintiles, for example, show that they are about twice as likely to have been late or missed a scheduled payment as are households in the highest income group. A similar relationship holds with respect to the age of the household head: Households headed by persons under thirty-five years of age are nearly four times as likely to report payment problems as are those headed by an individual at least fifty-five years of age. As might be expected, debt-service burdens-measured by the ratios of scheduled monthly payments to monthly income-are positively related to late payment problems. Households in the group with the highest ratios of debt payment to income are more than four times as likely to be late or to miss payments as households in the group with the lowest payment burdens.

While the relationships described above appear straightforward, their interpretation is not quite so clear. For example, although households with younger heads tend to miss or be late on their debt payments more often than their older counterparts, this finding may reflect the fact that younger persons also tend to have lower-paying jobs. Thus the extent to which age and income differences independently affect payment behavior is left unsettled. For this reason, a multivariate analytical framework was used to assess the likelihood that a household either repaid its debts as scheduled or fell behind or missed one or more payments. (12) However, because the survey did not collect some relevant information, such as data on asset holdings, the analysis cannot be as complete as one might wish.

The multivariate analysis suggests that payment problems are most strongly related to debt-service burdens, educational attainment, the number of children under eighteen years of age in a household, and marital status. Higher debt-service burdens were positively related to late payment problems, as was the number of young children in a household, while separated or divorced heads of households were also significantly more likely to report payment problems than either married couples or other single households. In considering all factors simultaneously through this approach, no statistically significant relationship was found between the incidence of payment problems and either the age of the household head or housing-tenure status. In general, the level of household income was also not found to be a good predictor of payment performance, except for those in the highest income quintile. The analysis found that compared with households in the lowest-income quintile, the highest-income families were much less likely to fall behind in their payments.

Severity of Payment Problems

Summary statistics, such as those presented in the previous sections, provide a broad perspective on the recent performance of consumers in paying their debts. These statistics, however, do not convey information, except in the most general sense, about the severity of payment problems among indebted households.

Households who reported falling behind in paying their debts were asked about the number of payments that were more than thirty days late and whether any of these payments were as many as sixty days late. All respondents reporting loans more than thirty days in arrears were also asked why they fell behind, what they did about it, and what actions creditors took in response to the late or missed payments.

Overall, the survey found that 9 percent of all indebted households, which is two-thirds of households reporting late payments, fell behind more than thirty days on one or more of their debt obligations in the year preceding the survey (table 3). The survey further found, however, that relatively few households fell sixty or more days behind in their payments. Only 3 percent of all indebted households (or roughly 20 percent with any instance of late or missed payments) reported at least one payment sixty or more days delinquent during the preceding twelve months.

A second way to measure the severity of payment problems is to examine the distribution of late payers by the number of payments on which they fell behind by more than thirty days during the previous year (table 5). Among the households who had at least one payment more than thirty days late, most reported only a limited number of instances in which payments were behind by this much. Eighty-two percent of the households that had at least one payment late by more than thirty days were this late on three or fewer of their payments during the year preceding the interview. At the other end of the spectrum, 7 percent of the households who were late at least once by more than thirty days reported ten or more such instances. Overall, the mean and median number of payments more than thirty days past due for those with such late payments were 2.9 and 2.0 respectively. If having more than three payments past due for more than thirty days in a twelve-month period is considered serious," then about 18 percent of all late payers, or roughly 3 percent of all debtors, could be said to have had a serious debt payment problem.

Reasons for Late Payment

Of those consumers who have had at least one payment more than thirty days late, most indicated that the main reasons for their difficulties were that they became overextended by taking on too much debt or they experienced an unforeseen change in their employment or health status. Overall, 55 percent of the families experiencing payment problems indicated that they became overextended; 6 percent experienced medical-related problems; and 24 percent either lost their jobs, were not working, or had suffered a cutback in the number of hours worked. (13) A fairly small group of households (roughly 14 percent) reported that they became delinquent either because they were on vacation or forgot to mail their payment.

Consumer Response to Payment Difficulties

Individuals facing debt payment problems respond in a variety of ways. Nearly 40 percent of those who were more than thirty days behind in their payments reported that they caught up on their delinquencies the next month or paid "when they were able." Other households delinquent in their payments cut back on other types of spending, took second jobs, worked longer hours, sold various items to raise funds, or borrowed or received gifts from relatives or friends. About 22 percent of late payers reported that they called their creditor about the problem; in some cases, the terms of the loans were extended, in others new loans were obtained. Some late payers, about 12 percent, said that they took no action in response to their late payments.

Creditor Responses to Late Payments

The survey asked households that were more than thirty days late in making their loan payments what actions creditors took in response. The most common responses mentioned, 71 percent of the total, were to call the borrower, send a letter, or send additional billing notices. (14 Other actions taken by creditors were to suggest new payment plans, notify credit reporting agencies, and cancel lines of credit. Only one household reported the repossession of an item securing a loan. In about 19 percent of the cases, respondents reported that creditors took no significant action in response to the late payments.


Some insights into the aggregate delinquency statistics developed from lender reports may be drawn from the results of the household survey, although some conceptual differences prevent simple comparisons between the two types of data. For example, one difference is that the aggregate statistics from industry sources reflect the proportion of loans delinquent at a particular time, while the household survey identifies borrowers who had been delinquent at any point in the twelve months preceding the survey date, regardless of their current status. Another difference is that the industry statistics usually apply to a specific type of lender, while the survey of households asked that late payments to any type of lender be reported.

Despite these differences, comparisons between the two types of delinquency statistics can be informative. For example, the higher that the household-reported delinquency rate is for a given type of loan relative to an industry-reported rate, the more widely dispersed late-payment behavior is among the debtor population. If a household-reported delinquency rate (covering twelve months) is close in magnitude to a corresponding industry-reported rate (for a given point in time), that suggests that industry-reported delinquencies reflect a relatively small set of chronically delinquent debtors. On the other hand, if the household-reported rate markedly exceeds the rate reported by lenders, that suggests that a larger group of individuals is occasionally delinquent. That is, it would imply that a borrower misses a payment one month, but subsequently gets back and stays on schedule and is replaced in the delinquent category by another short-term delinquent the next month.

For consumer loans, exclusive of credit cards, the ABA reported an aggregate thirty-day delinquency rate of about 2.5 percent on average during 1990, as indicated in chart 2. From the household survey, about 8.7 percent of those having such debts (combining observations for vehicle loans and for other instaLlment debt in table 3) had been delinquent some time during the previous twelve months. The relationship between the ABA and household survey rates suggests a moderate degree of rotation from period to period among those identified as delinquent. (If no one were late more than once a year or for as long as sixty days, the survey-based rate should be about twelve times the aggregate rate.)

Information from the household survey about the number of instances of delinquency also sheds light on the severity of delinquencies in the aggregate. As shown in table 5, several households in the survey reported multiple instances of late payments. Among those who were thirty days late at least once, 18 percent said they were that late on more than three occasions. From the distribution in table 5, it can be calculated that these frequently late debtors accounted for roughly half of all instances of late payment among surveyed households. When this approximation is applied to the ABA'S installment loan delinquency rate of around 2.5 percent, it appears that perhaps 1.25 percent of the installment loans on banks' books at a given time are owed by persons with a chronic late-payment problem. The actual figure is likely to be lower, however, because the classification of individuals as chronic late payers was based on the number of missed payments for any type of loan a person had. The proportion of debtors with multiple late payments for a single category of loans would be at least somewhat smaller.

On balance, the household survey data suggest that a substantial proportion of the loans that are past due at a point in time do not reflect serious payment problems, but ones that will be rectified within a reasonably short period. This conclusion is based on the relatively few survey respondents that reported payments as much as sixty days late, the small fraction of borrowers with several instances of late payment, and the statements of respondents concerning what actions they took after missing loan payments. In that the survey was conducted during a period of economic growth, the question may be raised as to whether these patterns would hold during an economic downswing. The answer cannot be known with certainty. It should be noted, however, that aggregate debt burdens were relatively high in the period surveyed, and the economy, while not in recession, was sluggish, which suggests some broad applicability for the results reported here.


To obtain information on consumer debts and recent household experience with repayment problems, the Federal Reserve Board developed questions that were included in the September and November 1990 and January 1991 Survey of Consumer Attitudes conducted by the Survey Research Center of the University of Michigan. Interviews were conducted by telephone, with telephone numbers chosen from a cluster sample of residential numbers. The sample was chosen to be broadly representative of the four major regions-northeast, North Central, South, and West-in proportion to their populations (Alaska and Hawaii were not included). For each telephone number drawn, a randomly selected adult from the family was the respondent.

The survey defines the family as any group of persons living together who are related by marriage, blood, or adoption, and any individual living alone or with persons to whom the individual is not related. The head of the family is defined as the individual living alone, the male of the married couple, or the adult in a family with more than one person and only one adult. Generally, when there is no married couple and more than one adult, the head is the person most familiar with the family's finances, or the one closest to age 45. Adults are persons aged 18 years or more.

Together the surveys sampled 1,534 families, 668 of whom were homeowners with outstanding mortgage debt. Six hundred and fifty-six additional famIlies had outstanding consumer credit but no mortgage debt. Altogether, 187 of the 1,331 indebted families reported falling behind in at least one of their scheduled loan payments in the twelve months before their interview. The survey data have been weighted to be representative of the population, thereby correcting for differences among families in the probability of their being selected as survey respondents. Estimates of population characteristics derived from samples are subject to errors based on the degree to which the sample differs from the general population. Table A. I indicates the sampling error for proportions derived from samples of different sizes.

1. For an analysis of the home equity loan market and how such loans are used, see Glenn B. Canner, Thomas A. Durkin, and Charles A. Luckett, "Home Equity Lending," Federal Reserve Bulletin, vol. 75 (May 1989), pp. 333-44.

2. For example, between 1980 and 1990, the average maturity on a new car loan at the major auto finance companies lengthened from forty-five to nearly fifty-five months. By reducing the rate of scheduled repayment, this lengthening of maturity has raised the stock of debt at any given time relative to the amount currently due and payable.

3. "Convenience credit" refers to funds that are borrowed on a short-term basis to facilitate transactions and are repaid in full within the billing cycle. Insofar as credit card accounts used for convenience frequently have positive balances on the last day of the month, reflecting charge activity occuring after the end of the last billing cycle, the amounts of debt outstanding reported by lenders would include convenience credit.

4. For a comprehensive examination of the distributional changes in consumer debt among U.S. households in recent years, see Robert B. Avery, Gregory E. Elliehausen, and Arthur B. Kennickell, "Changes in Consumer Installment Debt: Evidence From the 1983 and 1986 Surveys of Consumer Finances," Federal Reserve Bulletin, vol. 73 (October 1987), pp. 761-78.

5. Commercial banks (with deposits of at least 300 million) have been reporting consumer loan delinquency rates on their Report of Condition since 1982. In recent quarters, an average delinquency rate based on these reports has risen somewhat faster than the ABA delinquency rate. However, given the relatively recent origin of this series, little can be said of its behavior during different phases of the business cycle.

6. For a more complete discussion of the bankruptcy process and trends in the number of filings, see Charles A. Luckett, "Personal Bankruptcies," Federal Reserve Bulletin, vol. 74 (September 1988), pp. 591-603.

7. Research at the Credit Research Center of Purdue University indicates that a substantial proportion of persons declaring bankruptcy in recent years has been able to obtain credit fairly soon after completing the process. In some cases, a bankrupt may retain a credit card that had no outstanding balance at the time of the bankruptcy (and therefore need not have been listed among the filer's debts), which is then still available for use after bankruptcy has been declared. See Michael Staten, "The Availability of Credit to Consumers After Personal Bankruptcy," Working Paper (Purdue University, Krannert Graduate School of Management, Credit Research Center, 1991).

8. This figure of the proportion of indebted households exceeds similar estimates from other surveys sponsored by the Federal Reserve primarily because households having credit cards, but reporting no outstanding balance after their last payment, were categorized here as indebted households even if they had no other types of debt. This distinction was made because most of these households had in all likelihood used at least one of their credit cards during the preceding twelve months and consequently could have missed or been late in a payment during this period. Surveys, such as the 1983 Survey of Consumer Finances, have found that few households have credit cards and never use them. See Glenn B. Canner and Anthony W. Cyrnak, "Recent Developments in Credit Card Holding and Use Patterns Among U.S. Families," Journal of Retail Banking, vol. 7 (Fall 1985), pp. 63-74.

9. Every indebted household participating in the survey was asked the following question with respect to each of its outstanding loans. "During the past twelve months, were all the (type of debt) payments made the way they were scheduled, did you get behind on any of the payments, or did you make payments that were larger or more frequent than scheduled?"

10. Of course, lenders might also respond to higher-risk seekers of credit by modifying downpayment or collateral requirements or by charging higher interest rates commensurate with the risk assumed in extending the credit. To the extent that riskier applicants are served with higher-cost credit rather than by being refused credit, the variables associated with ex ante default risk would more likely correlate with payment performance as well.

11. Debt-service burdens are measured by the estimated monthly ratio of total debt payments to income.

12. The specific multivariate technique employed was the logit model. This technique is one in a family of econometric models that may be used to estimate statistical relationships when the dependent variable takes on a limited number of values.

13. Included in the medical-related category are situations in which insurance failed to cover medical expenses.

14. Some of the letters were reminders, others threatened specific action, still others were notifications of late charges. TABLE 1 OMITTED
 2. Debt status of U. S. households (1)
 Percentage distribution
Type of debt owed Percent
None 15
Mortage only 3
Consumer only 45
Both mortgage and consumer 38
 Total 100
Households with mortgage debt 41
Households with any debt 85
 1. Here and in the following tables, data have been weighted to ensure
the representativeness of die sample.
 2. Details may not add to 100 percent because of rounding.
Surveys of Consumer Attitudes, September and November 1990,
and January 1991, Survey Research Center, University of Michigan.
 4. Proportion of indebted households with payment
difficulties, by demographic characteristics
and type of outstanding loan (1)
 Demographic Any type Mortgage Consumer
 characteristic of debt debt debt
Household income
 Lowest 16 20 17
 Second 19 15 16
 Third 15 10 13
 Fourth 14 8 12
 Highest 9 6 7
 Education of household
 11th grade or less 14 13 13
 High school graduate
 or some college 16 10 14
 College graduate 13 9 11
 Age of household head
 Less than 25 13 13 13
 25-34 20 11 17
 35-4 18 11 15
 45-54 16 12 12
 55 and over 5 2 5
 Marital status
 Married 14 9 12
 Never married 13 6 12
 Widowed 4 2 3
 Divorced, separated 27 15 23
 Housing tenure status
 Own 12 9 10
 Rent 20 n.a. 20
 ratio (thirds) (2)
 Lowest 6 5 5
 Second 12 8 9
 Highest 26 15 24
 All debtors 14 9 12
 1. Households categorized as having payment difficulties are those who
reported having missed or been late in their debt payments in the preceding
 2. Three groups of equal size were determined for each debt category
separately. The figures shown are the proportion of each size group that fell
behind in their debt payments.
 n.a. Not applicable.
 Surveys of Consumer Attitudes, September and November 1990,
and January 1991, Survey Research Center, University of Michigan.
 5. Severity of late payment problems
 Percentage distribution
 Percentage of late
 Number of late payments payers with at least
 more than thirty days past due one payment more than
 thirty days past due
 1 34.9
 2 26.6
 3 20.6
 4 1.8
 5 2.8
 6-9 6.8
 10 or more 6.5
 Total 100.0
 Percentage of all indebted households that
 had at least one payment more than
 thirty days past due 9.2
 Number of payments more than thirty days
 past due
 mean 2.9
 Median 2.0
 SOURCE: Surveys of Consumer Attitudes, September and November 1990,
and January 1991, Survey Research Center, University of Michigan.
 A.1. Approximate sampling errors of survey results,
 by size of sample(1)
Percentage points
 Survey result Size of sample
 (percent) 100 200 1,500
 50 10.5 6.2 3.2
 30 or 70 9.6 5.7 2.9
 20 or 80 8.4 4.9 2.6
 10 or 90 6.3 3.7 1.9
 5 or 95 4.6 2.7 1.4
 1. The figures in this table represent two standard errors. Hence, for most
items, the chances are 95 in 100 that the value being estimated lies within
a range equal to the reported percentages, plus or minus the sampling error.
COPYRIGHT 1991 Board of Governors of the Federal Reserve System
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Middleton, Nellie D.
Publication:Federal Reserve Bulletin
Date:Apr 1, 1991
Previous Article:Electronic transfer of government benefits.
Next Article:Treasury and Federal Reserve foreign exchange operations.

Related Articles
Statement by Lawrence B. Lindsey.
Household sector borrowing and the burden of debt.
Painless tax deferral when selling a subsidiary.
Statement by Janet L. Yellen, Member, Board of Governors of the Federal Reserve System, before the Subcommittee on Financial Institutions and...
Statement by Lawrence B. Lindsey, Member, Board of Governors of the Federal Reserve System, before the Committee on Banking and Financial Services,...
The United States Flow of Funds Accounts and Their Uses.
Recent changes to a measure of U.S. household debt service.
The consumption effects associated with filing for personal bankruptcy.
Recent developments in the credit card market and the financial obligations ratio.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters