America's Debt Safety-Net

doi: 10.1111/j.1467-9299.2012.02094.x

AMERICA’S DEBT SAFETY-NET

JOHNNA MONTGOMERIE

This article evaluates the rapid escalation of American household debt in relation to the changingdynamics of liberal welfare capitalism. Starting with the outcome of rising household debt levelsduring the credit and asset bubble of 2001–7 it argues that the failure of asset-based welfare andthe inability of households to move beyond their historical dependence on earned income madeindebtedness essential to household social participation and protection. It examines the uniquerelationship of young adults (households headed by persons under age 35) and senior citizens(households headed by persons over age 65) to the liberal welfare regime, in particular the ways inwhich these relationships were shaped by the 2001–7 credit and asset bubble. By using a frameworkin which debt is analyzed as a claim against income, alongside other costs of social participationand daily living, we see the impact of the credit and asset boom on both young adults and seniorcitizens: growing indebtedness and financial insecurity.

INTRODUCTION

Successive debates about the form, content, and purpose of American welfare statereform have outlined the way in which different political traditions and institutionsshape policy measures. Some emphasize structural pressures, such as globalization orthe ageing population (Schwartz 2007), as key factors creating a permanent austerity thatchallenges the survival of the post-war welfare state (Pierson 2007). Others emphasizeideational change brought about by neoliberalism or managerialism as instigators ofwelfare state reform (Clarke and Newman 1997; Squires 1990). Esping-Andersen’s (1990)initial outline of the American liberal welfare state regime highlights inter-linkagesbetween employment policy, labour markets, and the welfare system. However, long-term cumulative changes in the American labour market regime and successive effortsat welfare reform have so transformed the American liberal welfare regime that it barelyresembles Esping-Andersen’s initial typology. The welfare regime to emerge over the past20 years has been described as privatized Keynesianism because it creates economic stability,effective demand, and welfare provision through housing and capital markets rather thanthrough employment and production (Crouch 2009); or ‘finance-led growth regime’because easy credit offered American households asset-based wealth gains in the form ofhousing and portfolio investments to compensate for income loss due to stagnating wagegrowth (Boyer 2000). It has been widely acknowledged that rising household debt was anintegral part of these transformations (Langley 2008; Montgomerie 2009; Seabrooke 2009).Not just in the United States, the ‘Anglo-liberal’ growth model could be generalized acrossAnglophone countries, like the United Kingdom and Ireland, that increasingly relied onconsumer-led growth financed with private debt, but also supported by high levels ofpublic expenditure (Hay 2011).

This article evaluates the rapid escalation of American household debt in relation tothe changing dynamics of liberal welfare capitalism. Starting with the outcome of risinghousehold indebtedness during the credit and asset bubble of 2001–7, this trend is assessedin relation to transformations within the welfare regime. Although the credit and asset

Johnna Montgomerie is at ESRC Centre for Research on Socio-Cultural Change (CRESC), University of Manchester,UK.

Public Administration 2013© 2013 Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden,MA 02148, USA.

2 JOHNNA MONTGOMERIE

bubble occurred ‘outside’ the logic and processes of welfare reform, it was nonethelesslinked through households’ daily use of debt to cope with these changes. Household debtis typically evaluated with reference to changes in financial markets, either innovationsor changing regulations. This leaves the obvious gap of understanding household debtwith reference to the transformations the household sector experienced over the past twodecades. Rising indebtedness may have been an unplanned outcome of the welfare regimechanges, but this indebtedness is as much about government household provisioning andlabour market practice as about the transformations in financial markets.

To provide an alternative account of the US liberal welfare regime during the 2001–7boom years, this article puts debt-led consumption at its core, conceptualizing how thefailure of asset-based welfare and the inability of households to move beyond theirhistorical dependence on earned income made indebtedness essential to household socialparticipation and protection.

Here we look at two specific subpopulations: young adults (households headed bypersons under age 35) and senior citizens (households headed by persons over age65). These two groups have a unique relationship to the liberal welfare regime, whichaims to support economic stability by smoothing out fluctuations in household incomeand/or consumption, particularly for the most financially insecure. Through direct incometransfers to the retired or unemployed and the funding of higher education and skillstraining, liberal welfare capitalism serves as a vital buttress to economic growth andstability. Under this welfare regime, young adults and senior citizens are a residual socialstratum usually deemed politically worthy of state support. Their links to specific forms ofstate support such as pensions, healthcare, and education provide a basis for consideringthe outcomes of changes in the liberal welfare regime.

Attempts to evaluate the intergenerational dynamics of transformation in the liberalwelfare regime are most often framed in terms of conflict between competing birthcohorts. Usually this arises from an emphasis on changing demographic trends, namelythe relatively numerous Baby Boomers compared to the decreasing birth rates of recentdecades, and how this interplays with economic assumptions about the income andwealth life-cycle. Whether it is private pension funds, social security, public financing ofgovernment services, or the recent housing boom, intergenerational dynamics are seen interms of the winners and losers (Hamil-Luker 2001; Kohli 2002; Mortensen and Seabrooke2008). This is especially so when we consider debates about the future of welfare services:‘the young’ must pay today unsure if the same services will be available to them in thefuture, while ‘the old’ benefit today without considering the fiscal implications. Alongsidethe intergenerational conflicts over welfare provisions, this article explores importantsimilarities between age groups in terms of the impact of the credit and asset boom,namely that many young adults and senior citizens experienced growing indebtednessand financial insecurity.

This is accomplished by illustrating these households’ experience of the last two decadesthrough descriptive statistics. Since it is impossible to mediate between ‘individual choice’and ‘structural logic’ in such a short article, a more modest effort is made to pragmaticallyassess the impact of the 2001–7 credit and asset bubble on the average young adult andsenior citizen household. As these households shape the form and content of the boom,bust, and faltering recovery, how they cope with the profound transformations in the USeconomy is relevant. The explicit focus is the severity of indebtedness resulting from thebubble and the ensuing financial insecurity. The timing of the credit and asset bubblesuggests that ‘stage of life’ is as important as the political and economic times one lives in.

Public Administration 2013© 2013 Blackwell Publishing Ltd.

AMERICA’S DEBT SAFETY-NET 3

DEBT AS PANACEA

America’s debt safety-net is the product of decades of sustained reforms to the liberalwelfare regime. Combined with the 2001–7 credit and asset bubble, it has created ahousehold sector now plagued by financial insecurity. Many households borrowedheavily to acquire assets but also to cope with the multiple demands of daily life: buyinga home, purchasing an automobile, getting a university degree, and buying ‘big-ticket’items, as well as everyday expenses. An uncomfortable reality is emerging in the UnitedStates in which the cost of social participation is beyond the income of most households.For example, in 2004, one-third of households reported using credit cards to cover basicliving expenses during four out of twelve months on average (Wheary and Draut 2005, p.11). Moreover, households regularly use credit as a form of social protection: to pay billsduring unemployment, to cover emergencies and unforeseen events (like car accidentsor illness), and to fund healthcare expenses and life in retirement (see DeNavas-Waltet al. 2003; Doty et al. 2005; Sullivan 2008). In this way, debt acts as a safety-net for manyhouseholds because borrowing is a necessity, not an option (Draut 2006).

Acceptance that increasing income inequality is a major ‘fault line’ inhibiting USeconomic stability is increasingly mitigated by the argument that household consumptionlevels is the variable that really matters (Ragan 2010, p. 12). The new ‘consumer welfare’paradigm claims that the redistribution of living standards is more important than theredistribution of income (Krueger and Perri 2006). Household consumption becomes a keybenchmark for assessing material well-being and a proxy measure of living standards;as such, access to credit becomes an important redistributive tool (Wilkinson 2009).Importantly, this complements decades of reforms seeking to replace public-fundedcash welfare programmes with welfare benefits that contribute to asset accumulationand are financed through indirect methods, such as credit subsidies and tax deductions(Sherraden 1991). This ‘asset-based welfare’ model seeks to enable households to providetheir own financial security by investing in portfolio funds and/or property markets,largely through tax deductions on home loans and portfolio investments. Households canaccess a combination of up-front tax deductions for portfolio investments contributions,years of tax-deferred growth, and eventual taxation at relatively low rates (such asduring retirement). Tax deductions are also available for interest paid on debts securedby a principal residence or a second home, the first $1 million of debt used to acquire,construct, or substantially improve a residence, as well as the first $100,000 of homeequity debt, regardless of its purpose. Also, when selling property, $250,000 ($500,000 fora married couple) can be excluded from capital gains, provided the owner used it as aprimary residence for two of the five years prior to sale.

This myriad of tax deductions translated into significant loss of tax revenue and,as such, promoted asset-based welfare at the expense of the social welfare provisions,hastening retrenchment of the ‘old’ welfare regime. Castles (1997) noted the ‘really bigtrade-off’ between homeownership and welfare provision in the American context beforethe housing bubble occurred. By making the ‘home’ simultaneously a dwelling, a store ofwealth, and a reserve of cash (equity withdrawal), it could redistribute wealth over thelife-cycle and smooth consumption over time; and, as such, mimic many of the primaryfunctions of the liberal welfare state (Kemeny 2005). Also, homeowners are more sensitiveto tax increases than renters; thus promoting homeownership puts significant politicallimits on welfare spending, creating ‘a new conservative politics’ where households arehostile to tax increases to fund state services (Schwartz 2008). The politically powerful ‘low



tax’ constituency in the USA remains resolute in defence of these deductions, despite theirlink with the housing/financial crisis and in the face of ongoing fiscal deficits. Support fortax breaks exacerbates welfare retrenchment by starving the government of tax revenues.

Ultimately, the failings of the current welfare regime created America’s debt safety-net.During the 2001–7 boom households were encouraged to invest in property and portfoliofunds; doing this in the midst of a credit and asset bubble too often required households toborrow heavily to realize (speculative) gains decades later. Tax incentives perpetuated thebubble as deductions on home loans meant that the more one borrowed the bigger one’stax break. According to the Joint Center on Housing Studies (JCHS) 2010 report on TheState of the Nation’s Housing, ‘total mortgage debt exploded from under $6 trillion in 1999to over $10 trillion in 2009 (real dollars), but when the bubble burst it drove home equitydown from its $14.5 trillion peak in 2005 to $6.3 trillion in 2009, wiping out more thanhalf of all housing wealth’ (JCHS 2010, p. 12). Similarly, the more income was diverted toportfolio investments, the larger the tax breaks (with increasing asset prices requiring everlarger contributions to realize returns). This created a curious situation where householdsbecame reliant on the tax relief of borrowing and investing, irrespective of whether theyrealized meaningful returns.

A key failing of the contemporary welfare paradigm is the belief that householdscan transcend the limits of their earned income, namely by investing in property orportfolio funds. Wages and salaries remain the single biggest source of income for mostAmerican households (Froud et al. 2010, p. 152). Longstanding policy commitmentsto promote non-inflationary growth through flexible labour markets and active labourmarket policies (i.e. welfare reform) translated into growing employment insecurity,stagnating income growth and declining non-wage benefits for most households (Kirshner2001; Temple 2000). Simultaneously, America’s corporate sector embarked on years ofpermanent restructuring by shedding jobs, pushing hard against wage growth andsignificantly reducing or eliminating non-wage benefits, like healthcare and pensions, fornon-management workers (Cutler and Waine 2001). In the face of this booming asset,markets did not improve household income gains: median household income droppedfrom $52,400 in 2000 to $49,800 in 2008; even at their last cyclical peak in 2007, realmedian incomes were 1.2 per cent below 2000 levels (JCHS 2010, p. 3). Instead rising assetprices and cheap credit allowed household paper wealth to appear healthy; in fact thiswealth-effect prompted even more borrowing and investment.

In 2007, this process reversed and gains were lost faster than they accrued. From 1999 to2009, real household wealth slid from $503,500 in 1999 to $486,600 in 2009, dropping over$17 trillion in just two years from 2006 to 2008 (JCHS 2010, p. 3). This is because during the2000–7 boom, homeowners cashed-out an astounding $1.2 trillion from refinancing primeconventional first-lien mortgages, and when home prices declined it left 11.2 million(roughly one-quarter) of homeowners with mortgages underwater on their loans (JCHS2010, p. 3). Such a pronounced drop in asset values only exacerbated the financial straincaused by the larger debt levels required to acquire the assets in the first place. At thevery least, asset-based welfare provided households with inadequate social protection; atworst, it increased financial insecurity.

AGE AND INSECURITY

The contemporary welfare regime expects households to strategically allocate their (stag-nating) earnings to buy their homes, build their investments portfolios, fund consumption,



protect against uncertainty, and create wealth for future financial security. For senior cit-izens this usually means borrowing against the equity in their home; for young adultsit means regularly borrowing to invest in education or a home, while simultaneouslydiverting income into a pension or mutual fund that does not guarantee future returns.As a result both groups must regularly rely on debt to fund social participation. Too oftenthe reality of rising household debt levels is dismissed or simply depoliticized with thestrategic deployment of household portfolio theory, and, for the young and the old, thelife-cycle permanent income hypothesis. Like all economic theories of human behaviour,the core belief is that individuals under conditions of scarcity allocate resources to max-imize utility. As such, individuals can choose whether to save for retirement or take outa loan, and decisions are based on information, preferences, and the assumption that allother things are equal (Guiso et al. 2002).

Household portfolio theory constructs the household balance sheet in terms of stocks(debts and assets) and flows (income and interest payments). For instance, during the2001–7 bubble it was widely believed that rising debts were offset by rising asset prices(in property and equity markets) and minimal income flows were mitigated by lowinterest rates. Comparing income to interest flows became the foundation of mortgageunderwriting standards. Fannie Mae, Freddie Mac, and Veteran Affairs applied the‘28/36 split’ criterion as its benchmark for its mortgages, where a maximum of 28 percent of income should be dedicated to mortgage and housing-related payments and nomore than 36 per cent of income dedicated to all debt payments. This framework wasconsidered sufficient to gauge a household ability to safely taken on a mortgage. What thefinancial crisis and housing market crash revealed was the progressive deterioration ofunderwriting standards among leading banks, mortgage companies, and the GovernmentSponsored Enterprises, for both prime and subprime mortgages. Relying on the householdportfolio framework (to formulate benchmarks or inform policy) obscured the deleteriousimpact of lowering down-payment thresholds, negative amortization loans, interest-onlymortgages, and adjustable-rate mortgages, and most importantly the residential housingbubble. This is because income was compared to interest payments, not total loan value.Now we see that residential property prices can decline rapidly and precipitously whilethe stock of debt does not adjust downward. Household portfolio theory not only does notconsider the long-term implications of debt servitude, it also obscures the practicalitiesof paying down the entire stock of outstanding debts while (supposedly) building thenecessary asset-base for present and future financial security.

Life-cycle theory assumes a balance between income, assets, savings, and debt changesacross an adult’s lifetime. According to this model, the very young will have low incomes,limited savings and assets, and will borrow relatively more; as individuals acquirelonger employment histories their income, savings, and assets will increase; and uponretirement, they enter a phase of dis-saving in which assets and savings are depleted toreplace employment income. Problematically, this assumes stable economic growth, lowunemployment, stable working careers, and a numerical balance between birth cohorts.This conceptual framework impacts how we understand the relationship of young adultsand senior citizens to the liberal welfare state (Slesnick 2001); most importantly bydismissing rising debt levels for young adults and senior citizens as predicable outcomesof the life-cycle model and failing to account for the impracticalities of building a sufficientasset base to live on after retirement.

These assumptions, concepts, and theories obfuscate the realities of rising householddebt and potential failures of asset-based welfare. Here household finances are assessed



in terms of sources and claims on income. For the vast majority of households, incomeis overwhelmingly from wages and salaries. Financial insecurity therefore derives fromthe many peremptory claims on these limited earnings. Additional sources of incomefor most households include government transfer payments and interest/profits fromvarious capital assets. Income data from the Survey of Consumer Finances (SCF) includewages, self-employed business income, interest, dividends, capital gains, food stamps andother government support, pension income, social security, alimony, and other support.Therefore, the asset-side of the balance sheet is represented insomuch as it contributesto household income. Similarly, total debt outstanding is the sum total of ‘mortgagedebt’ – all loans, mortgage, or home equity secured against the primary residence – and‘consumer debt’ – the sum of outstanding totals for SCF categories: credit card, educationloans, all instalment loans, all lines of credit, and vehicle loans. Since credit is alreadydefined as deferred future income (because buying on credit provides the use of a house,goods, and services before they are paid for), framing debt as a claim against current andfuture income is not a major conceptual leap. Also, as debt repayments are made in thepresent and the entire outstanding balances must presumably be paid off sometime in thefuture, the total size of a household’s debt is a significant claim against present and futureincome.

Debt is framed as a claim against income, alongside all other costs of social participationand daily living. This approach seeks to highlight the political importance of indebtednessas an indicator of financial insecurity. Grouping income sources together and comparingthem to total claims against that income provides a unique picture of indebtedness.By comparing (a) total debt outstanding as proportion of income, (b) repayment costsas proportion of total income, and (c) the impact of the repayment burden followingsimulated 10 and 25 per cent falls in gross annual income, we get a basic picture of therelative financial (in)security faced by young adults and senior citizens.

Importantly, this approach eschews the ‘allocation’ paradigm as central to under-standing household finances. This paradigm reduces financial decisions to an expressionof calculative or preferential choice. All too often, household debts are understood interms of the functional purpose of loans and the borrowing decisions derived from theseassumptions. For example, when ranking debt instruments by risk profile, secured loans(mortgages or home equity loans) rank topmost, because they are linked to an appreci-ating asset and the interest payments are deductable against income; whereas unsecuredloans (credit cards and other forms of consumer lending) rank lowest because they arehigh cost, not asset-linked, and interest payments are not tax deductable. Between thesetwo categories the typology is less clear: auto loans are linked to a depreciating asset,while education loans, often depicted as borrowing to invest in an ‘intangible’ asset, are infact non-revolving consumer loans. This hierarchy allows us to assume that consolidatingconsumer debts – whose interest payments are not tax deductable – into tax deductiblehome equity loans is ‘rational’, even advantageous. This ignores the equally plausiblescenario of accelerating indebtedness as other high-cost credit sources balloon becausedebt consolidation does not alleviate the underlying strain on household budgets.

While home equity loans may appear a ‘rational’ means of reallocating debt moretax-efficiently, in the current financial crisis and economic malaise it appears to havecontributed to growing financial insecurity. US homeowners cashed out an astounding$1.2 trillion when refinancing prime conventional first-lien mortgages between 2003 and2007 (JCHS 2010, p. 30). Some of these funds went directly to retire credit card and othernon-mortgage debt, and some substituted for auto loans and other forms of consumer



borrowing that might have otherwise occurred. Another interesting example is creditcard debt: for the three major US credit card issuers, 58 per cent of all receivables areover five years old (Froud et al. 2010, p. 157), suggesting that many households use thisvery expensive form of debt for long-term borrowing rather than short-term cash flowmanagement or big-ticket purchases; more problematically, borrowers are not paying offtheir debts. This suggests significant diminishing returns when transferring between debtinstruments that are not adequately considered, especially if households are taking onlower quality, higher cost debt simply to stay afloat.

IN DEBT TO GET AHEAD

High debt levels in the early stages of working life are often accepted uncriticallyas a necessity to begin building assets and long-term wealth holdings. Growth inoverall debt levels since 2001, especially compared to income levels, suggests thatindebtedness is threatening – not aiding – long-term financial security. Part of theproblem is contextualizing young people’s borrowing over their lifetime, where debtis seen as necessary and temporary for expected wealth gain later in life. Many seeindebtedness as a necessary ‘risk’, but the sheer volume of debt now required for socialparticipation surely undermines this common assumption.

This section interrogates the widespread belief that getting a university degree and/orhouse is a good investment, by evaluating young adult households with high schoolversus college degrees as well as homeowners versus renters. By observing the generaltrend and comparing these subpopulations, we see that young adults living through thecredit and asset bubble had a different experience from their predecessors. Buying a homeand/or getting a university education, considered standard middle-class entitlementsand cornerstones of liberal welfare capitalism, now require young adults to borrow atlevels that may undercut the assumed returns. Current debt levels of average under-35households suggest that high debt levels early in working life may intensify financialinsecurity, especially if income, savings, and assets never exceed cumulative debt, orif servicing costs creates a sustained drain on income. As such, indebtedness is not atemporary stage but a lifelong necessity.

Figure 1 illustrates the average total debt and income levels of all under-35 households,showing the deleterious impact of the 2001–7 bubble on financial stability. Throughoutthe 1990s, total debt closely tracked income at 100.1 per cent of income in 1992 and 122per cent in 1998. Over the decade, total outstanding debt increased by approximately$14,000 and income by $5,800; by contrast, in the 2000s debts increased just over $44,000compared to a $7,500 increase in annual pre-tax income, reaching 171 per cent of incomein 2007.

High debt levels result not only from investing in future assets like a home; in arecent survey 45 per cent of under-35s reported using credit cards in the previous yearto pay basic living expenses such as rent, mortgage, groceries, and utilities (Whearyand Draut 2005, p. 3). This is reflected in increased repayment costs from 1992 to 2007,rising from 17 to 20 per cent of pre-tax income. Falling nominal interest rates, assumedby household portfolio theory to mitigate rising debt, did not alleviate the repaymentburden. On the contrary, borrowing grew so much (partly encouraged by low interestrates) that it increased the proportion of income diverted to debt repayment, counteractingthe assumed benefit of lower interest rates. Young adults already face unique financialinsecurity because they comprise the bulk of temporary or contract workers, making them



All Under-35 families

$30,354$35,899

$44,343

$52,328

$71,986

$88,554

$0$10,000$20,000$30,000$40,000$50,000$60,000$70,000$80,000$90,000

$100,000

1992 1995 1998 2001 2004 2007

ConsumerDebt

MortgageDebt

Income

FIGURE 1 Young adults’ average outstanding debts and incomeSource: Survey of Consumer Finances, all families with head under 35, n = 4010 (1992), 4429 (1995),4149 (1998), 4030 (2001), 3769 (2004), 3510 (2007). For further details see Appendix.

vulnerable to job loss. Rising indebtedness compounds these problems, because under-35sare less able to cope with relatively minor drops in income. A 10 per cent drop in 2007pre-tax income levels, for example resulting from a pay cut or reduction in non-wagebenefits, increases the debt repayment burden from 20 to 22 per cent of household income.A more substantial 25 per cent drop, approximating a job loss or temporary lay-off for lessthan three months, would increase the repayment burden to 26 per cent of income. Thisshows the importance of contextualizing the affordability of debt if rising indebtednessmeans that households cannot withstand even minor drops in income without the addeddanger of insolvency.

Educational attainment is considered the single most important factor in determiningpotential lifetime earnings in the United States (Breen and Jonsson 2005) and the mostimportant step a young adult can take to ensure higher long-term financial security(Machin and Vignoles 2004). However, this claim is based on lifetime earnings data forthe past 30 years, and may not be the most accurate predictor of the next 30 years.Additionally, the cultural importance of a university degree as a basic middle-classentitlement or a ticket to the middle-class for the poor provides political justification forborrowing heavily to attain it. But the level of debt now required to complete universityand the lower incomes of university graduates due to ‘education inflation’ (more college-educated workers devalues the income gains achieved by the qualification) tend toundermine assumptions about the future benefits of higher educational attainment.

The long period of retrenchment in US government funding and support for highereducation arguably began in 1992, with the introduction of the unsubsidized federalstudent loan programme. This policy was justified by claiming that education is aninvestment; since the students are the primary beneficiaries of higher incomes fromeducation, they should bear the additional costs (Baum and O’Malley 2002). Since 1992,costs of education have risen sharply: from 2000 to 2010, average tuition and fees for afour-year degree at a public university rocketed 72 per cent, compared to 39 per cent inthe previous decade, while equivalent costs at private institutions increased 35 per centin the 2000s and 30 per cent in the 1990s (College Board 2010a). Government bursary



programmes, such as the 1965 GI Bill and the Higher Education Act that guaranteeaffordable university education for all who qualify, have not kept pace with theseescalating educational costs (Demos 2007). Instead, credit has become the main vehiclefor access to a university education. Federal loans now comprise the bulk of student aid:$65.8 billion compared to $41.5 billion for all federal grant and aid programmes combinedin 2009/10 (College Board 2010b).

Access to credit has obscured the more insidious problem of rising costs inhibitingaccess to higher education, even though these trends have undercut bachelor degreeattainment in the United States (ACSFA 2001, 2002). The 2002 National Student LoanSurvey revealed that over 70 per cent of students agreed that student loans were very orextremely important in allowing access to education after high school, while 72 per centsaid student loans were very or extremely important in allowing them to pursue graduatestudies (Baum and O’Malley 2002). In 2008, 68 per cent of students graduating with four-year public university degrees had debts averaging $20,300; and 74 per cent graduatingwith four-year private university degrees had debts averaging $26,200 (NPSAS 2009).This is because, in addition to education loans, university students are actively targetedfor a plethora of credit products. Credit card lenders regard students as more profitablesince they tend to hold revolving balances longer (Kara et al. 1994; Manning 2001).Students not only use credit cards for consumer purchases: 27 per cent use them towardsundergraduate education costs, such as tuition or books (Baum and O’Malley 2002).

Table 1 compares the income, debt, and repayment levels of under-35 householdswith a high school diploma and a college degree. As expected, compared to high schoolgraduates, college graduates have much higher incomes, but their much higher debt levelspartly offset perceived gains. In 2007, average total debt outstanding was 164 per cent ofincome for high school graduates and 192 per cent for college graduates. Again, 2001 is asignificant turning point.

For high school graduates over the 1990s, average income rose approximately $4,500while average total debt outstanding grew $8,200, nearly twice as fast, with debt toincome reaching 104 per cent in 1998. In the 2000s, income increased at a similar rate(approximately $4,700), whereas debts over the same period surged by $30,000. Debtrepayment costs remained steady at around 16–17 per cent of income from 1992 to2004, but then jumped to 23 per cent in 2007 because of rapid escalation in overall debt

TABLE 1 Young adults’ average income, debts outstanding, repayment for high school and collegegraduates

High school diploma/GED College degree

Income Total debtsoutstanding

Annualrepayment

Income Total debtsoutstanding

Annualrepayment

1992 $25,619 $23,068 $4,228 $42,594 $50,050 $8,0601995 $27,162 $26,909 $4,775 $42,560 $57,752 $8,0341998 $30,099 $31,280 $5,247 $50,822 $61,614 $7,8822001 $35,563 $36,137 $5,958 $67,736 $86,958 $11,6122004 $35,026 $45,361 $5,881 $63,947 $117,813 $12,0482007 $40,300 $66,088 $9,216 $77,844 $149,389 $15,059

Source: Survey of Consumer Finances, selected families with head under-35; with high school diploma, n=1318(1992), 1516 (1995), 1294 (1998), 1358 (2001), 1201 (2004), 1110 (2007); with college degree, n=1429 (1992), 1429(1995), 1399 (1998), 1220 (2001), 1273 (2004), 1175 (2007).



outstanding; again suggesting that high debt levels counteracted perceived borrowinggains during the post-2001 low interest rate period.

College graduates show a similar trend. Income gains in the 2000s ($10,100) werecomparable to the 1990s ($8,200), whereas average total debt outstanding increased$11,500 during the 1990s and an astronomical $62,500 during the 2000s. Average incomefor under-35 college graduates increased by $35,000 over the two decades while averagedebt increased $100,000; these figures seem to undermine the assumed benefits of getting acollege degree. Importantly this assumption is based on calculations from income gains ofthe Baby Boomer generation and overlooks the indebtedness resulting from a universityeducation. Nevertheless, high school graduates fare worse regarding income growth: 57per cent compared to 83 per cent for college graduates over the two decades from 1992. Auniversity education therefore still ensures higher income and faster income growth, butfails to protect against the long-term repercussions of the debt required to obtain a degree.

The debt repayment burden of college students tells a somewhat different story. In1992, average repayments were 19 per cent of average income, and steadily declined to15.5 per cent in 1998, then rose again to 19.5 per cent of pre-tax income by 2007. Thissuggests that college graduates had the same repayment burden at the height of the2001–7 bubble as in the depths of the recession in the early 1990s, hardly an endorsementof the benefits of borrowing to ‘invest’, and somewhat troubling as these households nowface the post-bubble recession.

Carrying such high debt levels during school adds additional financial pressure aftergraduation, when loan repayments begin. To consider how debt levels might impactyoung adult households in the current recession: in 2007, average debt repayments ofhigh school graduates were 23 per cent of average income. A minor 10 per cent drop inincome would push debt repayments to 25.5 per cent; while a 25 per cent drop woulddrive repayment costs to 30.5 per cent of pre-tax income levels. Compare this to collegegraduates where average repayments were 19 per cent in 2007: repayments after a minoror a major drop would reach 21.5 and 26 per cent of average income, respectively.The coping potential of college graduates appears better than that of their high schoolcounterparts for servicing the debt.

The overwhelming importance of homeownership to American households, as theironly major asset and increasingly their key source of social insurance and protection,means that getting on the housing ladder is critical to establishing financial independenceand security for young adults (Arnett 1998; Kennedy 2004). Rapidly ascending propertyprices, especially in major metropolitan areas where most young adults live and mostwell-paid work resides, raises the bar for entry into the housing market. The problem ofaffordability was addressed by easier access to credit. Young adults needed to borrowincreasing multiples of income to own a home because incomes did not keep pace withthe escalation in house prices. These levels of leveraged investment, combined with otherforms of borrowing by young families, suggest that homeownership may compoundrather than prevent long-term financial insecurity.

In fact, easy credit conditions in the 2000s only modestly increased homeownership ratesfor young adults. According to the SCF, homeownership among the ‘under-35s’ increaseda meagre 4 per cent from 1992 to 2007. This suggests that homeownership did not extendto the previously excluded, but rather that the problem of access and affordability createdby the property bubble was mainly solved by excessive leverage. Table 2 points to theambiguities of leveraged homeownership compared to renting. Homeowners have higherincome than renters, but also considerably higher debt levels. In 1992, total homeowner



TABLE 2 Young adults’ average income, debts outstanding, repayment for owners and renters

Own Rent

Income Total debtoutstanding

Annualrepayment


Annualrepayment

1992 $43,673 $67,003 $10,531 $22,540 $8,999 $1,9771995 $42,516 $75,587 $10,807 $24,345 $11,684 $2,3721998 $53,830 $87,576 $12,424 $24,834 $16,796 $2,2702001 $64,708 $107,955 $14,995 $30,575 $15,443 $2,7382004 $65,083 $146,470 $15,967 $30,935 $18,986 $2,3932007 $76,505 $188,443 $21,129 $34,753 $20,077 $2,719

Source: Survey of Consumer Finances, selected families with head under-35; own, n=1465 (1992), 1806 (1995),1335 (1998), 1416 (2001), 1401 (2004), 1355 (2007); rent, n=2545 (1992), 2623 (1995), 2814 (1998), 2614 (2001), 2368(2004), 2155 (2007).

debt averaged 153 per cent of annual income; by 2007 this had reached 246 per cent.By comparison, renter debt-to-income grew from 40 per cent in 1992 to 58 per cent in2007. Not only do homeowners have mortgage loans, they also have higher consumerdebt levels than renters, potentially due to the additional costs associated with owninga home and/or the cost of servicing mortgage debts from take-home income, requiringgreater recourse to consumer credit facilities for daily expenses (Sharpe 1997). Of courserenters still pay living costs, which are partially subsumed under mortgage payments forhomeowners, but this does not negate the clear financial pressures homeownership putson young adult households.

Another way of seeing the potential problems of borrowing heavily to buy a house,compared to renting, is the annual cost of servicing debts as a claim against income:renters appear to do better than homeowners when it comes to servicing their debts. In2007, the average annual repayment for homeowners was $21,129, which is 28 per centof average income, and $2,719 for renters, which is 8 per cent of average income. Withthese starting points it is no surprise that renters would also be better able to cope withpotential reductions in income levels as a result of the current recession: a 10 per centdrop in average income for renters would only increase debt repayment to 9 per cent; a25 per cent drop would push repayments to a manageable 10.5 per cent of income. Thiscompares to homeowners, where a minor (10 per cent) drop in average income wouldpush average debt repayments to 31 per cent of pre-tax income and a major drop to astaggering 37 per cent.

For young homeowners unable to withstand even a minor drop in income, higher debtto buy a home is likely to undermine rather than bolster financial security. The financialcrisis exacerbated this risk as young adult homeowners, having bought into the housingmarket during the boom only to have the market collapse, were much more likely to havetheir home equity completely wiped out (JCHS 2010, p. 12).

BORROWING TO LIVE

Individuals who retire from the workforce are expected to deplete their savings and wealthholdings. Social security was created in 1935 to protect elderly households with insufficientsavings and wealth holdings to fund retirement. Alongside the disabled, widows, andorphans, they received regular government cash transfers aimed at protecting these



vulnerable groups from poverty and destitution. Today social security is one of the largestexpenditures in the US federal budget, and thus a regular target for fiscal retrenchment.Faced with an ageing population and a declining birth rate, successive US administrationshave attempted to ‘plug the fiscal gap’ by capping social security payouts, and have evenchanged the Consumer Price Index (CPI) methodology to save $1 trillion attributed tothe ‘over-indexing’ of social security payments (Boskin et al. 1996, p. 15). Meanwhile, thelargest components of expenditure in senior citizen households – healthcare, prescriptiondrugs, housing, fuel, and food – undergo price increases greater than the all-items CPI usedto index social security payments (FARS 2010; Purcell 2006). Social security is a key sourceof income in most senior citizen households: although estimates vary, approximately 84per cent of over-65 households receive social security benefits, while 40 per cent claimsocial security as their largest income source (AARP 2006). As a regular target of callsfor reform to ‘entitlement’ programmes, what happens with social security is of hugenational political and economic significance.

Efforts to shift retirement from dependence on government social security towards a‘new rentier’ lifestyle of living off investments has had limited success. Even after thelong boom in asset prices, the average household still relied much more on social securityand earnings than on pension or asset income: income for over-65 households in 2008was derived 37 per cent from social security, 30 per cent from earnings, 19 per cent frompensions, and 13 per cent from assets (FARS 2010). The political and economic vision ofasset-based welfare appears to overlook the inability of many middle- and low-incomehouseholds to build a sufficient asset portfolio even if they wish to. Froud et al. (2010)highlight two important considerations: whether a specific generation has a good or bad30 years on the stock market, and whether individuals sell their fund when prices arehigh or low. ‘In every year of the 2000s, the value of the required ‘‘half of median’’ fundwas near or above $500,000 and at the peak in 2003 the required fund was more than $2million’ (p. 161). This simulation shows that the size of fund required during an assetbubble is so large that no middle- to low-income household could invest enough incometo realize it.

Moreover, regular market downturns and interdependent asset classes make massinvestment strategies highly unreliable as a form of social protection. For instance,market-indexed investment plans were widely seen as ‘safe’, but when markets declinedtogether, as in 2001 and 2007, these funds made significant losses. In the wake of thedot.com bubble, household wealth dropped from $55 trillion in 1999 to $49 trillion in 2002;in 2006 real aggregate household wealth soared to nearly $69 trillion, only to plummetagain in 2008 to $51 trillion (JCHS 2010, p. 12). With wealth still lower in 2008 than in1999, we can see that mass investment in the midst of a credit and asset bubble madethe household sector worse off. This is because asset price hazards make the timing ofportfolio investment important, but households have limited control over which years (oreven decades) they buy and sell their portfolios.

The over-65 households are a diverse group: some are very wealthy but most surviveon meagre incomes; some retire before 65 and others continue working and earningfrom choice or necessity. Rather than follow convention and create subpopulations ofhouseholds that receive social security or a social security income supplement (SSI)versus those with private pensions and/or property and investment income, this sectionwill compare all over-65 families to those with mortgage holdings and to the numericalmajority of over-65 households earning below $40,000 a year. This highlights the key



TABLE 3 Senior citizens (over-65s)’ average income, outstanding debt and repayment for all families andfamilies with mortgage holdings

All families With mortgage holdings


Repayment Income Total debtoutstanding

Repayment

1992 $26,593 $7,902 $1,715 $29,646 $41,285 $8,6781995 $32,569 $8,136 $1,786 $37,198 $39,731 $8,0011998 $38,153 $12,382 $2,755 $43,397 $58,298 $11,4452001 $47,419 $17,064 $3,346 $52,975 $71,331 $13,0082004 $50,179 $30,189 $4,105 $55,849 $83,743 $11,3112007 $68,792 $36,023 $5,441 $78,467 $117,897 $16,130

Source: Survey of Consumer Finances, households with head over 65; all families, n = 4520 (1992), 4705 (1995),4515 (1998), 4540 (2001), 4401 (2004), 4800 (2007); with holdings, n = 749 (1992), 890 (1995), 922 (1998), 963 (2001),1134 (2004), 1372 (2007).

failing of asset-based welfare: inadequate social protection because rising indebtednesscreates financial insecurity.

On the left of table 3, ‘all families’ presents the average income, debt, and repayments of‘over-65’ households. It suggests a relatively benign picture of rising indebtedness, withaverage income remaining above average total debt outstanding, income growing fasterthan debt, and debt around 30 per cent of income until 2004, when it jumps to 60 percent, then falls to 53 per cent in 2007. Debt repayments fluctuate only from 6.5 per cent in1992 to 7.9 per cent in 2007, and remain manageable in the face of 10 per cent and 25 percent simulated drops in average 2007 income, at 8.8 and 10.55 per cent, respectively. Inpart, this relatively healthy picture results from over-sampling of wealthy households inthe SCF, which tend to be older (Getter 2007); and because most senior citizens own theirhomes outright, mortgage debt is relatively low across the whole population. Yet evenwithin this broad picture we see a nearly fivefold increase in average total debt post-2001,leaping $19,000 during the 2000s after increasing only $4,500 through the 1990s.

The right side of table 3 isolates over-65 households that borrowed against their primaryresidence or took out a mortgage to buy property. This is relevant because residential realestate is the largest single asset class held by over-65 households (Apgar and Di 2006).Smaller asset holdings are represented in the income figures insomuch as they actuallycreate meaningful income gains, because, I contend, the size of the asset portfolio mattersless than the income derived from it. Presenting senior citizen households in this wayspotlights the compounding effect of the credit and asset bubble on financial insecurity.

Since most over-65s are ‘house rich and cash poor’, greater access to equity withdrawalis seen as an efficient way of smoothing life-cycle consumption and allocating scarcehousehold resources. Indeed, this is a key tenet of asset-based welfare initiatives. Amidthe 2001–7 property bubble, senior citizens, like many other families, used their homesas proverbial ATMs by cashing out the higher value of their property. The subsequentdrop in home equity since 2007 is astounding: ‘aggregate real home equity has not beenthis low since 1985 when there were far fewer homeowners than today. In addition, theamount of home equity and the amount of mortgage debt outstanding essentially flippedin just one decade: mortgage debt climbed from 65 per cent of home equity in 2000 to163 per cent in 2009’ (JCHS 2010, p. 13). The costs of repaying these loans now createadditional and often very substantial claims against fixed incomes.



Observing only over-65 households with loans secured against their primary residencesubstantially changes the picture of financial security. Throughout the 1990s, average debtgrew around $17,000, average income increased by $13,700, and debt-to-income remainedaround 140 per cent. By contrast, in the 2000s average debt increased $46,500, averageincome grew far less ($25,500), and debt-to-income reached 150 per cent in 2007. Manyassume that increased borrowing was simply a ‘rational’ response to lower interest rates,booming property markets, and tax advantages: the average repayment as a proportionof income did decline from 29 per cent in 1992 to 20.5 per cent in 2007. However, thisoverlooks the reality that such rapid increases in debt levels can counteract the perceivedgains of lower borrowing costs, especially in the wake of the financial and propertymarket collapse.

By and large senior citizens live on fixed incomes and have little recourse to paidemployment; a 10 per cent drop in income pushes average repayments to 23 per centof income, while a more substantial 25 per cent drop would push repayments to 28per cent. It appears, therefore, that borrowing after 65 creates financial insecurity, withaverage annual debt repayments for families with mortgage holdings three times higherthan average senior households. This trend is well recognized in the retail credit industrywhere senior citizens are now a key demographic, as their fixed incomes make themmore likely to revolve balances and stay in repayment status longer (Lichtenstein et al.2003; Mathur and Moschis 1994). In fact, the retail credit industry has developed newdebt collection strategies to target the emotional vulnerabilities of elderly people, suchas persistent phone calls and debt collection notices (Punch 2004). Pressure to makerepayments on outstanding debt is used to sell additional third-party products such assecond mortgages or viaticals – selling one’s life insurance policy for cash (Vincentini andJacques 2004).

The problem of growing indebtedness among the elderly becomes clearer when weobserve those earning under $40,000 a year – the majority of over-65 households.According to the SCF, a third of over-65s earn below $20,000 and a further 25–30 percent (depending on the year) earn $20,000–39,999. Figure 2 combines both categories toillustrate how nearly 60 per cent of senior citizen households experienced the credit andasset bubble. As the core demographic welfare spending seeks to support – and doubtlessprime targets for new asset-based welfare initiatives – their experience provides theclearest example of its failings.

Throughout the 1990s, average debt increased by $3,300 and income by $2,200; bycomparison, over the following decade debt increased $13,000 against a meagre $4,000growth in average income, reaching 232 per cent of income in 2007. Like other groups,lower-income seniors experienced a steady reduction in debt repayment obligation as aproportion of income, from 56 per cent in 1992 to 36 per cent in 2007, largely because oflower interest rates. Nevertheless, the cost of servicing a much larger debt still createdadditional claims against largely fixed incomes, with a 10 per cent or 25 per cent dropin income raising repayments to 40.5 or 49 per cent of average annual pre-tax income,respectively.

Growing evidence indicates that senior citizens living on modest incomes borrowheavily for daily living because incomes have not kept pace with the rising living costsunique to over-65s (Hurst and Willen 2007). Rising healthcare costs in particular haveadversely affected senior citizens (EBRI 2008). Many already retired households havehad to accept successive rounds of corporate restructuring to reduce ‘legacy costs’: from1988 to 2003, the percentage of large employers offering medical coverage to retired



$27,359

$20,346

$30,690

$39,377

$46,861$52,390

$0

$10,000

$20,000

$30,000

$40,000

$50,000

$60,000

1992 1995 1998 2001 2004 2007

Over-65 Earn less than $40,000with holdings

Consumer debt

Mortgagedebt

Income

FIGURE 2 Senior citizens’ average income and outstanding debts for households that own home and earningless than $40,000Source: Survey of Consumer Finances, selected households with head over-65, with mortgageholdings and earning less than $40,000 (Bulletin Categories 1 and 2), n = 171 (1992), 194 (1995), 213(1998), 236 (2001), 316 (2004), 333 (2007).

employees dropped from 66 to 38 per cent (McGhee and Draut 2006, p. 5). As employer-sponsored benefits decline, seniors become increasingly reliant on state support for basicliving costs. Yet for lower-income seniors, gaps in Medicare coverage mean that, withoutmedical insurance, they must contribute up to a third of their income to healthcare-relatedexpenses (Public Policy Institute 2003). Most often these expenses are for prescriptiondrugs which, for those covered under Medicare, still average $860 a year in out-of-pocketexpenses (Zeldin and Rukavina 2007). Gaps in healthcare coverage mean many seniorsface rising medical expenses at a time when they have more frequent and serious healthproblems.

CONCLUSION

This article assesses US household indebtedness to reveal the underlying financial inse-curity plaguing many households. Household debt is mostly evaluated in terms oftransformation in financial markets, for instance the impact of new practices like creditscoring or securitization but also predatory lending. Also rising debt levels are explainedin purely functional terms with reference to prevailing macroeconomic conditions: lowinterest rates and excess liquidity increased the supply of credit, causing debt levels toincrease. The aim here was to evaluate household indebtedness as more than a financialtransaction conditioned by economic conditions. To do so, this article highlights thecumulative effects of transformations in the American liberal welfare regime, in particularhow the homeownership–welfare trade-off was shaped by the credit and asset bubble andcontributed to young adults and senior citizens using debt as a safety-net.

The contemporary welfare paradigm (wrongly) assumed that adequate social protectionwould naturally result from asset-based welfare and using easy credit, rather than risingwages and salaries, to maintain living standards. After 2001 many over-65 householdswere still reliant on social security, but were also borrowing against existing equity in theirhome, or more problematically, against the inflated value of their home. Young adults



became indebted by borrowing for social participation: to buy homes, get an education,supplement lacking employer-sponsored benefits such as medical costs, and create astopgap during unemployment. Evidence presented here indicates that the recent creditbubble has made young adults and senior citizens more financially insecure.

These findings are relevant insomuch as they can contribute to ongoing policy debatesand provide a picture of why economic recovery is failing. The 2007 financial crisisexposed the politics of abandonment that has shaped the liberal welfare regime overthe past two decades: the old and the young are left to cope with the long period ofeconomic restructuring as well as protracted asset and credit bubble through unsustainableborrowing. Yet, how households are meant to cope with a financial crisis, prolongedrecession, and faltering recovery without going further into debt is by no means clear.

ACKNOWLEDGEMENTS

The author would like to thank Ian Bruff, Simone Scherger, Mikko Kuisma, and threeanonymous referees for their helpful comments and advice.

REFERENCESAARP (American Association for Retired Persons). 2006. Fact Sheet: Sources of Income for Older Persons. Washington, DC: American

Association for Retired Persons Policy Research Institute.ACSFA (Advisory Committee on Student Financial Assistance). 2001. Access Denied: Restoring the Nation’s Commitment to Equal

Educational Opportunity. Washington, DC: Committee on Heath, Education, Labor, and Pensions.ACSFA (Advisory Committee on Student Financial Assistance). 2002. Empty Promises: The Myth of College Access in America.

Washington, DC: Committee on Health, Education, Labor, and Pensions.Apgar, W.C. and Z.X. Di. 2006. ‘Housing Wealth and Retirement Savings: Enhancing Financial Security for Older Americans’,

in Gordon L. Clark, Alicia Haydock Munnell and Michael Orszag (eds), Oxford Handbook of Pensions and Retirement Income.Oxford: Oxford University Press, pp. 618–37.

Arnett, J.J. 1998. ‘Learning to Stand Alone: The Contemporary American Transition to Adulthood in Cultural and HistoricalContext’, Human Development, 41, 5/6, 295–315.

Baum, S. and M. O’Malley. 2002 . College on Credit: How Borrowers Perceive Their Education Debt. Results of the 2002 National StudentLoan Survey. Washington, DC: Nellie Mae.

Boskin, M.J., E. Dulberger, R. Gordon, Z. Griliches and D. Jorgenson. 1996. Toward a More Accurate Measure of the Cost of Living.Final Report to the Senate Finance Committee. Washington, DC.

Boyer, R. 2000. ‘Is a Finance-Led Growth Regime a Viable Alternative to Fordism? A Preliminary Analysis’, Economy and Society,29, 1, 111–45.

Breen, R. and J.O. Jonsson. 2005. ‘Inequality of Opportunity in Comparative Perspective: Recent Research on EducationalAttainment and Social Mobility’, Annual Review of Sociology, 31, 1, 223–43.

Castles, F.G. 1997. ‘The Really Big Trade-Off: Home Ownership and the Welfare State in the New World and the Old’, ActaPolitica, 32, 4, 1–19.

Clarke, J. and J. Newman. 1997. The Managerial State: Power, Politics and Ideology in the Remaking of Social Welfare. London: Sage.College Board. 2010a. Trends in College Pricing. New York: College Board Advocacy and Policy Center.College Board. 2010b. Trends in Student Aid. New York: College Board Advocacy and Policy Center.Crouch, C. 2009. ‘Privatised Keynesianism: An Unacknowledged Policy Regime’, British Journal of Politics and International

Relations, 11, 2, 382–99.Cutler, T. and B. Waine. 2001. ‘Social Insecurity and the Retreat from Social Democracy: Occupational Welfare in the Long Boom

and Financialization’, Review of International Political Economy, 8, 1, 96–118.Demos. 2007. Higher and Higher Education: Trends in Access, Affordability, and Debt. New York: Young Adult Economics Series.

Demos.DeNavas-Walt, C., B.D. Proctor and R.J. Mills. 2003. Income, Poverty, and Health Insurance Coverage in the United States. Current

Population Reports. Washington, DC: Census Bureau.Doty, M.M., J.N. Edwards and A.L. Holmgren. 2005. ‘Seeing Red: Americans Driven into Debt by Medical Bills. Results from a

National Survey’, Issue Brief Commonwealth Fund, 837, 1–12.Draut, T. 2006. Strapped: Why America’s 20- and 30-Somethings Can’t Get Ahead. New York: Doubleday.EBRI (Employment Benefit Research Institute). 2008. The 2008 Retirement Confidence Survey. Issue Brief No. 316. Washington, DC:

EBRI.



Esping-Andersen, G. 1990. The Three Worlds of Welfare Capitalism. Cambridge: Polity Press.FARS (Forum on Aging-Related Statistics). 2010. Older Americans Update 2010: Key Indicators of Well-Being. Washington, DC:

Federal Interagency Forum on Aging-Related Statistics.Froud, J., S. Johal, J. Montgomerie and K. Williams. 2010. ‘Escaping the Tyranny of Earned Income? The Failure of Finance as

Social Innovation’, New Political Economy, 15, 1, 147–64.Getter, D.E. 2007. ‘Using the Survey of Consumer Finances: Some Methodological Considerations and Issues: Sins, Sinners,

Data, and Details: Commentary’, Journal of Consumer Affairs, 41, 2, 216–18.Guiso, L., M. Haliassos and T. Jappelli. 2002. ‘Introduction’, in L. Guiso, M. Haliassos and T. Jappelli (eds), Household Portfolios.

Cambridge, MA: MIT Press.Hamil-Luker, J. 2001. ‘The Prospects of Age War: Inequality between and within Age Groups’, Social Science Research, 30, 3,

386–400.Hay, C. 2011. ‘Pathology Without Crisis? The Strange Demise of the Anglo-Liberal Growth Model’, Government and Opposition,

46, 1, 1–31.Hurst, E. and P. Willen. 2007. ‘Social Security and Unsecured Debt’, Journal of Public Economics, 91, 7–8, 1273–97.JCHS (Joint Center on Housing Studies). 2010. The State of the Nation’s Housing. Cambridge, MA: Harvard University.Kara, A., E. Kaynak and O. Kucukemiroglu. 1994. ‘Credit Card Development Strategies for the Youth Market: The Use of

Conjoint Analysis’, International Journal of Bank Marketing, 12, 6, 30–48.Kemeny, J. 2005. ‘The Really Big Trade-Off between Home Ownership and Welfare: Castles’ Evaluation of the 1980 Thesis, and

a Reformulation 25 Years on’, Housing, Theory and Society, 22, 2, 59–75.Kennedy, D.G. 2004. ‘House Arrest’, American Demographics, 24, 4, 13–16.Kirshner, J. 2001. ‘The Political Economy of Low Inflation’, Journal of Economic Surveys, 15, 1, 41–70.Kohli, M. 2002. ‘Generational Equity is More than Pension Finance’, Zeitschrift Fur Gerontologie Und Geriatrie, 35, 2, 129–38.Krueger, D. and F. Perri. 2006. ‘Does Income Inequality Lead to Consumption Inequality? Evidence and Theory’, Review of

Economic Studies, 73, 1, 163–93.Langley, P. (2008). The Everyday Life of Global Finance: Saving and Borrowing in Anglo-America. Oxford: Oxford University Press.Lichtenstein, J., Y. Chen, K. Wu, S. Verma, J. Gist and J. Tumer. 2003. ‘Retirement Income Security, Wealth, and Debt Burden

among Baby Boomers and Vulnerable Populations: Research Results’, Gerontologist, 43, 80–9.Machin, S. and A. Vignoles. 2004. ‘Educational Inequality: The Widening Socio-Economic Gap’, Fiscal Studies, 25, 2, 107–28.Manning, R.D. 2001. Credit Card Nation: The Consequences of America’s Addiction to Credit. Washington, DC: Basic Books, pp. 159–94.Mathur, A. and G.P. Moschis. 1994. ‘Use of Credit Cards by Older Americans’, Journal of Services Marketing, 8, 1, 27–39.McGhee, H.C. and T. Draut. 2006. Retiring in the Red: The Growth of Debt among Older Americans. Borrowing to Make Ends Meet.

New York: Demos.Montgomerie, J. 2009. ‘The Pursuit of Past Happiness? Middle-Class Indebtedness and American Financialisation’, New Political

Economy, 14, 1, 1–24.Mortensen, J.L. and L. Seabrooke 2008. ‘Housing as Social Right or Means to Wealth? The Politics of Property Booms in Australia

and Denmark’, Comparative European Politics, 6, 2, 305–24.NPSAS (National Postsecondary Student Aid Study). 2009. Results: National Postsecondary Student Aid Study. Washington, DC:

US Department of Education.Pierson, P. 2007. ‘Investigating the Welfare State at Century’s End’, in P. Pierson (ed.), The New Politics of the Welfare State. Oxford:

Oxford University Press, pp. 1–16.Public Policy Institute. 2003. Out-of-Pocket Spending on Health Care by Medicare Beneficiaries Aged 65 and Older. Washington, DC:

Data Digest.Punch, L. 2004. ‘Older Debtors, New Problems’, Credit Card Management, 17, 5, 26.Purcell, P. 2006. Topics in Aging: Income and Poverty Among Older Americans. Washington, DC: Congressional Research Service.Ragan, R.G. 2010. Fault Lines: How Hidden Fractures Still Threaten the World Economy. Princeton, NJ: Princeton University Press.Schwartz, H. 2007. ‘Round Up the Usual Suspects! Globalization, Domestic Politics, and Welfare State Change’, in P. Pierson

(ed.), The New Politics of the Welfare State. Oxford: Oxford University Press, pp. 17–44.Schwartz, H. 2008. ‘Housing, Global Finance, and American Hegemony: Building Conservative Politics One Brick at a Time’,

Comparative European Politics, 6, 2, 262–84.Seabrooke, L. 2009. ‘Embedded Liberalism is Dead, Long Live Embedded Liberalism: National Welfare Concerns and Interna-

tional Policy Responses to the Sub-Prime Crisis’, New Political Economy, 15, 1, 51–70.Sharpe, D.L. 1997. ‘Credit and Debt Portfolios of the Young and Old’, Consumer Interests Annual, 43, 210–12.Sherraden, M. 1991. Assets for the Poor: A New American Welfare Policy. London and New York: M.E. Sharpe.Slesnick, D. 2001. Consumption and Social Welfare: Living Standards and their Distribution in the United States. Cambridge: Cambridge

University Press.Squires, P. 1990. Anti-Social Policy: Welfare, Ideology and the Disciplinary State. London: Harvester Wheatsheaf.



Sullivan, J.X. 2008. ‘Borrowing during Unemployment: Unsecured Debt as a Safety Net’, Journal of Human Resources, 43, 2,383–412.

Temple, J. 2000. ‘Inflation and Growth: Stories Short and Tall’, Journal of Economic Surveys, 14, 4, 395–426.Vincentini, C. and P. Jacques. 2004. ‘Seniors Get Ready to Charge’, MarketFacts Quarterly, 23, 1, 64.Wheary, J. and T. Draut. 2005. The Plastic Safety Net: The Reality Behind Debt in America. New York: Demos and The Center for

Responsible Lending.Wilkinson, W. 2009. ‘Thinking Clearly about Economic Inequality’, Policy Analysis, July, 640, 1–28.Zeldin, C. and M. Rukavina. 2007. Borrowing to Stay Healthy: How Credit Card Debt is Related to Medical Expenses. New York:

Demos and The Access Project.

APPENDIX: NOTE ON SURVEY OF CONSUMER FINANCES

The SCF employs a dual-frame design, including an area-probability and a list component.The area-probability (AP) sample provides robust coverage of the nation and good repre-sentation of behaviours that are broadly distributed in the population. About two-thirdsof the ultimately completed cases derive from this sample. The other set of survey caseswas selected as a list sample from tax data by the Statistics of Income Division of theInternal Revenue Service (SOI) and disproportionately selects the wealthy. Because theSCF sample is not an equal-probability design, weightings play a critical role in interpret-ing the survey data and are intended to compensate for unequal probabilities of selectionin the original design and for unit non-response (failure to obtain an interview). See:Kennickell, A.B. and R.L. Woodburn. 1999. ‘Consistent Weight Design for the 1989, 1992,and 1995 SCFs and the Distribution of Wealth’, Review of Income and Wealth, 45, 2, 193–215.


Documents

America's Debt Safety-Net