Consumer Credit and the Politics of Market Access Gunnar Trumbull by jizhen1947


									Consumer Credit and the Politics of Market Access

Gunnar Trumbull

November 2009


        From 1980, consumer debt levels rose across most of the advanced industrialized

countries. In the United States, non-mortgage household debt increased by half, from 17% to

26% of annual disposable income. In France, which had started from a much lower base,

household non-mortgage debt increased six-fold. Still, French household borrowing remained

50% below the level in the United States. Consumer credit default rates were correspondingly

lower in France. In 2005, 2% of consumer credit accounts in France were in default (over 90

days late), compared to 6% in the United States. 1 This extraordinary growth in household debt

testifies to a revolution in consumer lending. From a market in which capital had been rationed,

in both France and the United States, credit had become abundant. Even relatively poor

            Cédric Houdré, L’endettement des ménages début 2004,” Insee Première 1131, April

households were able to gain access to credit. For middle class households, there seemed to be

virtually no limit on the credit they could access. At the same time, the French were persistently

more frugal in their use of credit than their American counterparts. To understand why American

households borrowed so heavily in this period, I compare the postwar trajectory of consumer

credit markets in the US and in France.

Figure 1. Household debt as a share of disposable income, 1980-2005. 2

                             non-mortgage debt                            total debt
                  1980          1991         2005         1980           1991          2005
US                17%           22%          26%          70%            98%           135%
France            3%            8%           18%          30%            40%           64%

         Explanations of differences in household credit use, both over time and across countries,

have tended to emphasize the distinctive nature of, and changes in, supply and demand for credit.

The most common set of explanations attributes differences in credit supply to the perceived

riskiness of consumer borrowers. The bulk of economic accounts of consumer borrowing

therefore focus on institutions intended to manage borrower risk, including credit rating

agencies, secure collateral, and bankruptcy. 3 They emphasize the need to manage non-payments,

and the role that institutional context—including property rights and information sharing—can

play in helping lenders to detect consumers’ ability and willingness to pay. A second common

explanation has focused on cultural determinants of demand. French families were portrayed as

             Federal Reserve Board, Bureau of Economic Analysis.
             Japelli and Pagano, etc.
culturally thrifty. American families were seen to have sacrificed fiscal responsibility in the

interest of materialism. 4 Culture has also been invoked to explain changes over time. In such

accounts, cultural norms about credit are perceived to have changed, with recent generations

more accepting of credit.

       I argue that national patterns of consumer credit use are best understood as reflections of

policy accommodations over social policy goals in the two countries. In both cases, access to

credit was closely tied to concerns about public welfare. How they understood that connection

was nearly opposite. In France, credit was a threat to working class welfare. From the 1950s to

the early 1980s, administrative restrictions on credit price and volume reflected a concern that

consumer credit would crowd out critical industry investment. By the late 1980s, rising concerns

about overindebtedness of French households led policymakers to see credit as a potential social

trap that threatened permanent social and economic exclusion. In the United States, “third-way”

policies of the Bill Clinton administration embraced liberal access to credit as a channel for

social mobility. As housing markets boomed in the late 1990s, their bet seemed to be justified.

First-time home buyers, who tended to hold less equity in their houses, experienced dramatic

profits due to high levels of leverage. With newfound wealth in their homes, they were able to

roll over credit card debt into loans secured on home equity. The Financial crisis of 2008 was, in

significant degree, the decisive defeat of third-way economic policy.

Theories of Consumer Credit

       The economic theory of consumer credit has its roots in arguments made by two

influential postwar economists. The first, Franco Modigliani, proposed that individual decisions

           Juliet Schor, The Overspent American (New York: Harper, 1999).
about consumption, saving, and borrowing were dictated by changes in income over the course

of an individual’s lifespan. 5 The core insight of his ‘lifecycle’ savings hypothesis was grounded

in the distinction between consumption and utility. Whereas consumption measured the money

you spent, utility represented the benefit you gained from it. Modigliani’s core assumption,

borrowed from the utilitarian philosophers of the 19th century, was that ‘marginal’ utility

decreased as an individual’s consumption increased. This amounted to the seemingly reasonable

claim that a dollar spent by a rich man generated less utility for him personally than a dollar

spent by a poor man. Jeremy Bentham had famously used the assumption of declining marginal

utility to justify a redistributive welfare state, on the grounds that it would maximize social

utility. Modigliani’s proposal was that individuals would opt to redistribute their purchasing

power across different periods of their lives, through savings and borrowing, in order to

maximize their lifetime utility. The Modigliani model seemed to provide a rationale for

consumer debt that was unrelated to economic investment, since individuals might borrow in

expectation of higher future income.

       The second piece in the puzzle was introduced by Joseph Stiglitz in a 1981 article with

Andrew Weiss on credit rationing. 6 In the absence of additional information about the intention

of borrowers, Stiglitz and Weiss proposed, lenders could not distinguish between borrowers that

           Franco Modigliani and Robert Brumberg, “Utility analysis and the consumption

function: An interpretation of cross-section data,” in Kurihara, ed., Post-Keynesian Economics

(New Brunswick, NJ: Rutgers University Press, 1954).
           Joseph E. Stiglitz and Andrew Weiss, “Credit Rationing in Markets with Imperfect

Information,” American Economic Review 71/3 (1981): 393-410.
accepted to pay a higher interest rate because the loan was genuinely risky and those who

accepted to pay a higher interest rate because they did not intend to repay the loan at all. The

credit rationing thesis was an extension of George Akerloff’s 1976 account of the ‘market for

lemons’ in used car markets. Akerloff had argued that when customers could not detect hidden

flaws in used cars, they would pay less, and those lower prices would induce sellers to offer only

cars with flaws. Without better information, the fear of lemons became a self-fulfilling prophecy.

Extended to credit, the market failure thesis suggested that without additional information about

borrower motivations, price alone would not be sufficient to create an efficient market, and many

deserving but risky borrowers would not be able to secure loans.

       Taken together, the Modigliani life-cycle theory and the Stiglitz-Weiss theory of credit

rationing provided a theoretical account of patterns of credit use. As lenders adopted new and

sophisticated credit scoring technologies in the 1980s and 1990s, they were able to overcome the

adverse selection problems that Stiglitz and Weiss suggested would lead to credit rationing. 7 The

life-cycle smoothing theory gave policymakers a reason to allow household debt to expand.

Rising household debt levels could be justified on the grounds that increased access to credit

would improve consumption smoothing and thereby raise private household utility. 8 Even more

important for policymakers, higher household debt seemed to promise increased economic

growth. If credit constraints had previously made consumption excessively sensitive to current

income, then consumers were not borrowing as much as would be optimal given their total

           Darryl E. Getter, “Consumer Credit Risk and Pricing,” Journal of Consumer Affairs

40/1 (2006), pp 41-63.
           Japelli and Pagano, 2009.
resources, including assets and future income. This implied that greater access to credit would

raise consumption, and thereby output. 9 More household debt would boost GDP growth, and this

would benefit both individuals and the whole economy.

       It is a measure of the influence of their theories that both Modigliani and Stiglitz would

be awarded the Nobel Prize for their work on consumer credit. So compelling was this argument

that by 2005, the IMF was urging the core European countries to pursue financial liberalization

in order to increase household debt. They estimated that financial liberalization could increase

the EU’s GDP by 0.5% to 1% per year. Stiglitz himself became a key framer of ‘third way’

economic policies during his tenure as Clinton’s chairman of the Council of Economic Advisors.

At the same time, consumer credit was proving to be highly profitable for the financial sector.

Politically, pro-credit policies of the 1990s were grounded in a cross-class coalition driven by

financial sector profitability on the right and ‘third-way’ consumer credit advocates on the left.

The Stiglitz-Weiss/Modigliani framework provided the theoretical cement that held the two sides


       The second common explanation for growing levels of debt attributes the trend to

changing cultural norms about credit. In these accounts, consumer debt rose because in the 1990s

because consumers became more comfortable with credit. Typical of this sentiment is a writer

for the Wall Street Journal in 2000: “In a nation that once saved all it could, and resisted taking

on debt, Americans no longer just borrow money to buy houses. They load up on debt to

            Tullio Jappelli and Marco Pagano, “Capital Market Imperfections,” American Economic

Review 79/5 (1989): 1088-1105.
purchase stock. To go on vacation. To pay the electricity bill. To buy groceries.” 10 This idea of a

generational change in perceptions of consumer credit has been a theme of consumer credit

studies since at least the beginning of the century. In 1958, the vice-president of Security-First

National Bank of Los Angeles warned: “The doctrine of thrift – the idea that individuals should

save first in order to buy luxury goods, had deep roots in this country…. This concept died hard,

but it did fall.”11 Thirty years before that, one could hear the same analysis being voiced. In

1929, Samuel Grafton wrote of the then emerging credit selling plans: “Once there was shame in

not buying for cash…And once there was great shame in borrowing. Now all of that is dead

gone.” 12

        Data on household use of credit show that most people’s parents bought on credit for

most of the past century. Surveys conducted in the 1950s show that about half of Americans at

any time reported using installment credit. In 2005, about 60% of Americans reported having

non-mortgage debt. (See below.) The key difference is that credit users in the 1990s and 2000s

repaid their debt more slowly. Given the constant of household debt, why have observers

             Gregory Zuckerman, “Borrowing binge in U.S. sparks concern Debt is at a record level

and those who might least afford it have been accumulating most,” Wall Street Journal,

        July 5, 2000.
             D.Z. Albright, Vice-President, Security-First national Bank of Los Angeles, “The Role

of Consumer Credit,” 1957 Consumer Credit Symposium, New York: Consolidated Reporting

Company, 1958, p 53.
             Samuel Grafton, “Need, Gold, and Blood,” The North American Review 227/3 (March

1929), p 342.
consistently claimed that their parents shunned credit? We may tend to perceive our parents as

fiscally prudent because of the demographics of credit. People have always borrowed more

heavily when they were young, starting their careers, and having children. By the time children

have grown and are starting their own families, their parents are typically relying less, or not at

all, on credit. A second reason is almost certainly psychological. The idea that consumer credit

has become accepted in society has probably never been fully true. Surveys of consumer credit

consistently show that consumer’s words and deeds are at odds. In his survey of poor consumers

in Harlem, for example, David Caplovitz finds that attitudes about credit and credit practice

diverged dramatically. Although 59% of respondents said buying on credit was a bad idea (and

only 15% said it was a good idea), only 21% of families reported only ever paying cash for

consumer durables they purchased. 13 It may be that people attribute to their parents their norms

about credit use that they themselves hold, but that they are not able to live up to in practice.

       Against these economic and cultural arguments, I propose that consumer credit markets

are best understood as the outcome of an evolving cross-class coalition focused on welfare goals,

rather than as an evolving effort to reach efficient market outcomes through regulations that

manage market failures. The same sort of analysis is likely to be pertinent to other kinds of

markets. Some of these, like health care, public lotteries, and insurance, are self-evidently tied to

broader welfare outcomes. Even seemingly ‘natural’ markets like agricultural products and

manufactured consumer goods can be constituted by cross-class coalitions that have as their goal

specific welfare outcomes rather than the pursuit of efficiently functioning markets. For markets

of this kind, we may understand more about both regulatory variation, and likely future

            David Caplovitz, The Poor Pay More (New York: Free Press, 1967), pp 95-97.
regulatory directions, if we approach them as variants on welfare policy rather than regulatory

patches to fix specific market failures.

       Why did American politicians encourage credit extension when their French counterparts

did not? National patterns of credit use have been dictated by the way in which the political left

and right have linked household credit to questions of welfare and redistribution. In the United

States, the supply of consumer credit has been supported by an evolving cross-class welfare

coalition that, for idiosyncratic reasons, did not exist in other countries. The groundwork for

abundant consumer credit was laid in the 1910s and 1920s, when credit access was supported by

the left as a benefit to the emerging working class and by the right as an alternative to expansive

welfare of the kind that was spreading in Europe. In the 1950s and 1960s, consumer credit was

supported as an element of the postwar productivity coalition, in which the left supported credit

to the working class and the right encouraged credit as a bulwark against communism. By the

1970s, credit market access became a consensus issue of equal rights campaigns by women and

blacks. In the 1990s, consumer credit had again been re-imagined, this time as a middle-ground

‘third way’ policy that would both empower the poor and drive returns high returns on financial


       The evolving political coalitions that sustained support for consumer credit over time also

legitimated shifting public discourses about the rationale for consumer credit. Consumers in the

1910s and 20s were relying on credit as a form of insurance in case of injury, unemployment,

birth, or any other shock to income or expenses. During the 1950s and 1960s, consumers,

stereotyped as returning soldiers, were borrowing to furnish their newly acquired houses. This

sort of borrowing was perceived to create a virtuous circle, in which greater demand drove

manufacturing productivity and higher wages. During the 1970s, credit became a symbol of
economic equality for women and blacks. It also helped middle-class Americans to hedge against

the rising costs of everyday goods that was driven by the OPEC oil embargo. By the 1990s,

consumers were borrowing to maintain a growing standard of living in the context of stagnant

real wages.

       In France, politicians on the left and right shared a reluctance to accept liberal access to

consumer credit even once consumer lending became profitable. As in the United States, their

approach to consumer credit markets was grounded in a welfare-like cross-class coalition. But,

unlike in the United States, French politicians emphasized the risks: that excessive consumer

credit would crowd out industrial investment, that interest charges would reduce purchasing

power, and that over-indebted consumers could become permanently excluded from the benefits

of social and economic citizenship. More pragmatically, banks that had failed to enter the

consumer lending market in the 1980s raised little objection to restrictive policies on credit rating

and usury, since they wese would apply to consumer finance companies without a significant

impact on the banking system.

Credit Use in France and America

       For most of the postwar period, Americans have been heavy users of consumer credit. In

2007, three quarters of families held consumer debt in the form of mortgages, installment loans,

and credit card balances. Outstanding credit card debt averaged $10,637 per person, spread

across 5.5 bank and retail cards; outstanding installment debt averaged $21,000, primarily in the

form of automobile and educational loans. 14 Taken together, all non-mortgage consumer credit

            Nielson Report, April 2009; Experian marketing insight snapshot, March 2009; Federal

Reserve Board, Survey of Consumer Finance.
was equal to 25% of household disposable income. If we include equity extracted from homes

that was used for consumption (and to pay off consumer debt), outstanding consumer debt as a

share of disposable income rose to 33%. Servicing all of this debt was imposing an ever higher

burden on the family budget. For 15% of families, monthly debt service payments accounted for

over 40% of their disposable income. The burden of debt service was especially high for the

poor. Among families in the bottom 20% of income, over a quarter faced monthly credit

payments that exceeded 40% of their disposable income. 15 How did American households get so

far into debt?

       If American households were spendthrifts, French households were frugal. Between 1955

and 1985, American households consistently carried consumer (non-mortgage) debt equal to

15%-20% of their disposable income. During the same period, French household debt rose

gradually from 1% to 3% of disposable income. From the mid-1980s, household debt levels rose

dramatically across the advanced industrialized countries. Yet French debt levels remained

persistently low, even relative to other European countries. By 2005, French household debt had

not yet caught up with the level of debt carried by American households in 1955.

            Brian K. Bucks, Arthur B. Kennickell, Traci L. Mach, and Kevin B. Moor3e, “Changes

in U.S. Family Finances from 2004 to 2007: Evidence from the Survey of Consumer Finances,”

Federal Reserve Bulletin, February 2009, pp A37-A50.
Figure 2. Household non-mortgage debt in France and United States, 1945-2005 (share of
disposable household income)


                                                                         US (with extracted equity)




                     1945   1950   1955   1960   1965   1970   1975   1980     1985   1990   1995   2000   2005

       Notes: Home equity extraction contribution to consumer spending and debt reduction

estimated at 80% of total home equity extraction, assuming that 20% of extracted equity was

shifted to other assets. Based on Kennedy and Greenspan, 2008. France experienced no

significant home equity extraction.

       This broad pattern of credit use in France and the US is reflected in the share of consumer

durable sold on credit. A survey by the Federal Reserve Board in 1949 found that half of new

cars, refrigerators, and TVs purchased in the United States were sold on credit. 16 As wartime

restrictions on credit were eased in the early 1950s, consumer credit grew as a share of all

purchases. In the 1950s, 85% of new cars and 80% of new TVs were sold on credit. 1718 Credit

            CHA, UFB, “Notes sur le financement des ventes à credit aux Etats-Unis,” December

1952, pp 1-2.
            Michel Drancourt, Une force inconnue: le credit (Paris: Hachette, 1961), p 26.
sales represented 60% or all durable goods sales, and a third of all retail sales. 19 In France, credit

sales as a share of total retail sales were significantly lower. In 1954, 19% of new car sales were

on credit, and 70% of TVs. 20 By 1960, the credit share of car sales rose, to 26%, and for

televisions fell, to 50%. 21 By 1965, 40% of new cars were being sold on credit. 22 Compared to

the US, where two thirds of all cars were being sold on credit in the 1950s, in France the share

seems to have been about one third. André Malterre, in his study of credit sales in the US and

France, found credit accounted for roughly 60% of all durable goods sales in the United States in

the mid-1950s, including cars, refrigerators, TVs and radios. In France, only about 30% of such

items were sold on credit. 23 By the 1990s, revolving credit accounts had made it difficult to

distinguish which particular products were being sold on credit. Nevertheless, the broad cross-

             Union federale de la consommation, Bulletin Mensuel d’Information 29 (1955), p 48.
             François des Aulnoyes,” Le credit à la consummation,” Combat January 30, 1954 ;

Michel Drancourt, Une force inconnue: le credit (Paris: Hachette, 1961), p 26.
             Marius Allègre, “Étude du credit a la consummation,” Conseil Économique, March 4,

1954, p 269.
             “Le Credit a la consummation en France,” Agence Economique et financière, June 9,

             Gerard Constant, "La Vente à crédit," Les Cahiers Français, April 1965, p 18.
             André Malterre, “Problème du credit à la consommation,” Journal Officiel 20, 11

August 1961, pp 770.
national pattern of credit sales seems to have been the same. In 1995, 22% of all household

goods were being sold on credit in the United States, compared to 9% in the France. 24

       The French clearly relied less on consumer credit than their American counterparts, but

this seems not to have reflected a specific objection to buying on credit. When the founders of

the Carte d’Or credit card surveyed the French in 1965 to assess openness to a genuine credit

card, 60% said that they would appreciate access to credit. 25 A survey conducted in 1969 found

that three of four workers at Renault viewed consumer credit favorably. 26 By the early 1990s,

attitudes had hardly changed. A survey in 1992 finds that 74% of French were favorable to

credit; 24% are opposed credit use. Similar surveys were conducted in the US in the 1950s, and

they revealed greater apprehension. A survey in 1954 found that 50% of American households

thought installment buying was a good idea, versus 37% who opposed it. 27 A similar survey of

farm-operator families in 1959 found that 54% thought it was acceptable to buy household

equipment on credit. 28 While it would be a mistake to place too much weight on such surveys, it

            Hubert Balaguy, Le crédit a la consommation en France (Paris: Presses Universitaires

de Paris, 1996), p 22.
            Maurice Roy, "La guerre des cartes est déclarée," L’Express, March 4, 1968.
            Seidman 2004, p 264.
            Jan Logeman, “Different Paths to Mass Consumption: Consumer Credit in the United

States and West Germany during the 1950s and 1960s,” Journal of Social History 41/3 (2008), p

            RLDM, Box 152, folder 32, Robert L. D. Morse, “Credit and its Use: Attitudes and

Practices of Kansas Farm-Operator Families, 1955,” Draft article, May 14, 1959.
does at least seem clear that the French were not strongly culturally averse to using consumer


          The French lag also seems not to reflect any particular technical or innovative deficiency.

In reality, the cross-border diffusion of consumer credit innovation has been surprisingly quick.

If many of the early consumer credit leaders in the US were students of the European small loan

institutions, by the end of World War II, the flow of small credit innovation and diffusion had

reversed. Bankers around the world were visiting the US in the 1950s to track developments in

the small loans and credit card sector. French bankers made repeated study trips of this kind in

the 1950s and 1960s. Credit cards were especially subject to the smooth international transfer of

technology and ideas. Many of the early payment cards were specifically targeted at businessmen

who traveled. In 1947, the International Air Transport Association launched the Universal Air

Travel Plan (UATP), an international credit card accepted by 72 airlines. 29 The early travel and

leisure cards, Diners’ Club and American Express, were also specifically international in their

orientation. Diners’ Club was launched in 1950 in the United States; its French-owned affiliate

opened for business just four years later, in 1954. The American Express card launched in 1958

in the United States and in 1961 in France. American banks began experimenting with revolving

credit in 1958 and 1959; France saw the first experiments with revolving credit in 1962, directly

inspired by the American model. When computer systems emerged that could automate account

tracking and billing, the French moved quickly to adopt them. In the United States, JC Penney

               BdF CNC, 1427200301, box 318, Grande Bretagne, Reglementation des ventes à

crédit, février 1955 – décembre 1968, Correspondence, Sharman Wright, Bank of England,

December 31, 1954.
purchased four IBM 1400 machines in 1962 to process their credit accounts; in France, Cetelem

installed two of the same computers the following year. As one prominent French lender would

later acknowledge, nearly every credit innovation in France had been directly borrowed from

lender innovations in the United States.

       If the French were willing to use credit and their consumer finance companies had the

capabilities to offer it, why did borrowing rates remain so low? The answer hinges on a

combination of business context and regulatory response. Until the mid-1980s, the sector was

barely profitable. At the same time, national industrial policy focused available funds on

investment rather than consumption. Whereas American lenders were willing to make largely

unprofitable loans to consumers, France’s lenders were not. Even France’s dedicated finance

houses required cross-subsidies from retailers and manufacturers to survive. One side effect of

this low profitability was that they delayed launching revolving credit facilities until the mid-

1980s, when a confluence of financial liberalization the discovery of profitability in consumer

lending made such plans financially feasible.

       The late move into revolving loans helps to explain some, but not all, of the continued

low credit extension under liberalization in the late 1980s and 1990s. French households were

borrowing more, but not too much more. The reason had to do with a series of regulatory

responses that focused on consumer over-indebtedness as a threat to social and economic

inclusion. Supported by a cross-class coalition that included the Catholic right and labor left,

credit markets were hemmed in by new restrictions on usury, the collection of positive credit

data, the provision of ‘free’ credit, and a provision for discharge under bankruptcy. Unlike in the

United States, where credit access was seen as a tool for social mobility, excessive consumer
debt was seen in France as having the potentially marginalize an entire segment of the


A Supply-Side Revolution

       Observers who trace the emergence of the ‘hyperliquid’ consumer credit market typically

point to five revolutions: securitization, risk scoring, credit reference bureaus, risk-based pricing,

and low monthly payments. The first three changes are either not distinctive to the United States,

or have strong analogs in France. The last two are predominantly American practices that do help

to explain differences in lending outcomes that we observe between the two countries.

       First, much emphasis has been placed on the role of securitization of consumer credit as a

tool for reducing the cost and increasing the volume of capital available for lending. In the

United States, the first credit card accounts receivables were securitized in 1986. This practice

has mainly been limited to the United States and, later, Britain. But European lenders also found

ways to directly tap the markets for inexpensive capital. In 1984, Cetelem made its first

convertible bond offering. By including an option to purchase stock, it helped Cetelem to tap

Europe’s deep corporate bond markets at favorable interest rates. Second, observers have pointed

to the role of advanced risk analysis and credit scoring in reducing loan losses and increasing the

range of potential borrowers. But these technologies were not unique to the US market. Credit

scoring technology that emerged with the computer revolution of the 1970s spread rapidly to the

largest consumer lenders around the world. In the US, Fair Isaac developed the first consumer

credit scoring system in 1970. Cetelem in France adopted a similar statistical scoring technique

borrowed from the United States in 1974.
       Third, much has been written about the role of credit reference bureaus in reducing the

risks of consumer borrowing. 30 One might imagine that some of the difference in credit

extension we see between France and the United States is attributable to the availability of

accurate credit data. But this problem is not as great as it would theoretically seem, for three

reasons. In the absence of positive credit data, French lenders found other means of assessing

creditworthiness. They relied initially on a strategy of indirect risk assessment by marketing

credit products to customers who had already successfully repaid sales loans agreed to by

retailers. Credit reference bureaus had long played an important were important role in the US

credit market, but that was primarily because the US market was so highly fragmented. In the

1950s and 1960s, the US had tens of thousands of consumer lenders. France by contrast had only

a small number of lenders, including less than a hundred registered consumer finance companies,

and the largest of these had since 1974 shared black list data with each other. In general,

repayment risk seems to have been relatively easy to manage among the many challenges of

making cost-effective consumer loans.

       The final two changes in lender practice are genuinely distinctive to the United States,

and likely played important roles in the high level of consumer debt in that country. The first of

these is risk-based pricing. Traditionally, consumer lenders pooled the risk of their borrowers, so

that the impact of good and bad credit risks would offset each other. The beauty of risk pooling

was in the power of large numbers: even if a lender did not know the risk of any individual, he

knew a lot about the likelihood of repayment in a properly-defined population. In this way, credit

            See Tullio Jappelli and Marco Pagano, “Information Sharing in Credit Markets,”

Journal of Finance 48/5 (1993), 1693-1718.
served as insurance. As with other forms of insurance at the time, risk pooling provided a

transfer from lower risk to higher risk individuals. A pair of innovations caused this system to

break down. The first was a set of new information technologies that allowed individual risk

assessment. If one assumed that individuals were different in systematic ways, instead of

homogenous members of a group, then each individual represented a different level of risk and

should therefore pay a different risk premium. Suddenly, risky borrowers would be given access

to credit, but that access would come at a higher price. Critics of this approach noted that

charging higher prices could accentuate risk, making risk scoring a self-fulfilling prophecy.

       The last revolution in lending practice that remained distinctive to the United States was

the negative amortization account. 31 For American lenders, the business logic of reducing

required monthly payments had long been understood. First, lenders knew that lower monthly

repayments reduced loan defaults. Second, in a world in which creditors where competing for

customers, it made little sense to lose a good customer simply because they had paid off their

loan. In 2000, for example, the cost of acquiring a new credit card customer in the United States

was roughly $50. That placed a premium on customers you already had, especially if they were

proven re-payers. Moving to a negative amortization minimum repayment rate was an easy way

of retaining reliable customers.

       Financial data from Cetelem’s Historical Archives allow us to trace how the costs of

lending changed over time. Figure 3 below breaks down the total interest charged on a loan into

its cost components. Note first the dominant role of administration and interest payments, and the

relatively low cost of provisions for nonpayment. The significant increase in the cost of capital

            Citibank’s Andrew Kahr claims to have invented the idea of the 2% minimum credit card payment.
from 1960 to 1986 were almost entirely offset by lower administration costs and a reduction in

government taxes.

Figure 3. Cost components of Cetelem loans, 1960-1992.

                                 1960 32          1973 33          1986 34          1992 35
Consumer lending rate            18.7%            19.2%            18.8%            17.6%
Breakdown of costs:
Administration                7.0%             7.0%                5.1%             3.8%
Interest                      6.0%             9.4%                10.3%            8.8%
Risk                          0.7%             0.8%                1%               1.9%
Tax                           4.5%             1.0%                1.2%             1.0%
Profit                        0.8%             1.0%                1.2%             2.1%
Source: Cetelem Historical Archive, data compiled by author.

       These data on credit costs at Cetelem show that the single most important factor in rising

profitability was the reduction in administrative costs. From 1973 to 1992, the costs of

administering a lone were cut in half. Computerization was a core part of the reason. As

Cetelem’s president, Pierre Boucher, acknowledged: “Computers…constitute the hard core of

our strategy. 36 In December 1962, Cetelem installed 2 cutting-edge IBM 1401 machines. In 1973

it is replaced with an IBM 370/145, which was 40 times more powerful than the two 1401

            "Des Tarifs pratiqués dans la vente a crédit des biens de consommation durables,"

Revue du Centre d’Information et d’Etude du Crédit, no. 3 (October 1962); Michel Renault, “La

Vente a Crédit s’acclimate en France,” Le Monde, November 29, 1962.
            BNPP CHA, “La finance et la trésorerie,” à Découvert, no. 9, June 1973, p 6.
            CHA, Pascal Bonnet, “Tenir nos taux,” Nous, no. 2, June-July 1987, pp 10-11.
            CHA, Nous Avons 40 Ans, April 1993, p 25.
            CHA, Correspondence from Pierre Boucher, director general, October 4, 1982. (D01)
machines combined. In 1984, they upgraded to an IBM 3081, again the lastest computer

technology. The computers allowed them to take advantage of telecommunications. In 1974,

Cetelem installed an internal data network called Transpac that connected all of the roughly 200

Cetelem offices and agencies. In 1980, they begin making real-time approvals over the network,

and by 1981 all approvals were made over Transpac. In 1983, loan approval times were brought

below one minute. By 1984, 3,000 affiliated retailers were connected directly to the Cetelem

network via France’s early public teletex network, Minitel.

         US lenders were equally aggressive in embracing technology. From the early 1970s, JC

Penney’s management made substantial investments to improve operations and lower costs. In

1974 they leased 7,000 sales credit communications terminals and eleven System 4000

computers from TRW. In 1975 they completed a nationwide Interconnect System that enabled

instant credit authorization from any of the stores. In 1977 they deployed new statistical risk-

evaluation systems at each of their 22 regional credit offices. These systems were expected to

increase credit application approval rates by 18%. 37 Automated IBM remittance processing

equipment was installed in 1978. With the help of greater automation, JC Penney’s credit

operations made a profit for the first time ever in 1975. 38 After a downturn corresponding to the

high inflation of the second oil shock, profits returned. See Figure 4 below.

              “Report to Credit Management,” JC Penney Management Report 4/8 (September

              JCPCR, Credit: Credit Growth Campaign, 1975, Steve Kernkraut, “Update: Net cost of

Credit,” Internal document, Credit department, September 28, 1975, pp 1, 8.
Figure 4. JC Penney Charge Card Operations, Net Earnings (1969-1982).

                  $ millions




                  1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982



       Ironically, it was this profitability that would spell the end for retail credit. Retailers too

small to have their own brand card simply accepted the main bank cards. Retailers that valued

their own-brand credit cards contracted out the service to consumer finance companies like

HBSC and GE Capital. Retailers with profitable in-house credit programs were made offers that

they couldn’t refuse. In 1985, JC Penney acquired a bank, the First National Bank of Harrington,

Delaware, and renamed it the JC Penney National Bank. 39 In 1999, GE Capital purchased JC

Penney’s credit business. In 1986, Sears launched the general purpose Discover Card, with no

fees and a 1% refund. With the addition of insurance and real estate services, financial services

            JCPCR, JCPenney Credit: 40th Anniversyary Booklet, 1998.
accounted for the lion’s share of Sear’s earnings between 1986 and 1991. 40 The Discover

network was ultimately acquired by Morgan Stanley in 1997. Retailers’ share of the US

consumer credit market fell from 12% in 1971 to 8.9% in 1981. 41

       By the 1980s, France’s lenders were also thinking differently about their industry. The

high inflation period of the late 1970s-early 1980s in particular had proved formative. Through

the steep fluctuations in interest rate that the inflationary shock demanded, lenders learned that

consumer demand for credit was highly insensitive to price. Not only did consumers not seem to

care about nominal interest rates, they also were seemingly indifferent to real interest rates. The

first lesson came from the run-up in inflation in the late 1970s. After bouts of significant

inflation during the 1970s, inflation in France briefly touched 20% in mid-1981. In order to

maintain reasonable real interest rates, consumer finance companies pushed their consumer loan

rates to almost 30%. At first they didn’t know how borrowers would respond. It had long been

assumed that 25% represented a psychological ceiling for borrowers. Yet as this barrier was

passed, the head of the Association of French Finance Establishments (APEF) recalled

concluding that ”30% per year could be adopted without significant consequences…” 42 As the

crisis deepened, loan applications actually increased.

            Ronald D. Michman and Alan J. Greco, Retailing Triumphs and Blunders: Victims of

Competition in the New Age of Marketing Mangement (Westport, Conn.: Greenwood

Publishing, 1995), p 39.
            JCPCR, JCP Credit – 10 year strategic Plan, 1983-1993, Part 1, JCPenney Credit

Division, Stategic Plan, April 1983, p A-1.
            Le Matin, June 29, 1981.
Figure 5. Nominal and real consumer lending rates in France, 1979-1986 43







                   1979    1980    1981       1982     1983      1984   1985      1986
                    Consumer borrowing rate          Inflation      Real interest rate

       The second lesson came with the decline in inflation in the early 1980s. It was during this

period that lenders found that consumers were also insensitive to real interest rates (i.e., nominal

interest minus inflation). As inflation fell, lenders lowered their interest rates more slowly—

recurring bouts of inflation in the 1970s had taught them to be cautious. The result was a

growing wedge between consumer interest charges and inflation. Between 1983 and 1986,

average real interest rates for consumer credit in France rose from 10% to 15%. Yet demand for

credit continued to rise dramatically. By 1986, lenders who had observed this phenomenon drew

the obvious conclusion. The head of one consumer finance company explained: “Consumers pay

almost no attention to the absolute level of the rate. They are interested mainly in the monthly

payment…No study has ever shown the least correlation between a rise in the cost of credit and a

            Philipppe Manière, "Les constructeurs auto "bradent" leurs prêts," Quotidien de Paris,

April 30, 1986.
reduction in consumption.” 44 For Cetelem, the insight was decisive for the company’s strategy.

In 1986, CEO Pierre Boucher sent a note to the entire company declaring: “The interest rate is

not a decisive element in households' decisions to take a short-term loan.” 45 Cetelem would no

longer be the lowest cost lender, as it had in the past. They would instead focus on product

features that consumers did seem to value.

Liberalization in the United States

       These changes in the logic of lending were accompanied by government policies that, in

the US case, liberalized credit access. These reforms were driven primarily by a vision of

competitive discipline on consumer lenders. The first blow to usury caps came from a court case,

the 1978 Marquette National Bank vs. First of Omaha Supreme Court decision that applied state-

of-origin usury rules to out-of-state credit card lenders. In the case, Marquette National Bank of

Minnesota had filed claim against First of Omaha for charging 18%, which was permitted under

Nebraska state law, to its credit card customers in Minnesota, where usury caps limited interest

rates to 12%. Marquette at the time was charging its customers 12% interest, plus a $10 fee to

cover expenses. Apart from the differences in interest rate and fees, the services were the same:

revolving credit cards that were both affiliated with the BankAmericard (Visa) network. The

defendants argued that the National Bank Act (National Bank Act, Rev.Stat. § 5197, as amended,

12 U.S.C. § 85) preempted state usury laws (Minn.Stat. § 48.185 (1978)) for states in which they

were not located. The relevant passage of the law, Section 85, read: “Any association may take,

receive, reserve, and charge on any loan or discount made, or upon any notes, bills of exchange,

            Le Matin, June 29, 1981.
            CHA, Pierre Boucher, internal memo, 1996.
or other evidences of debt, interest at the rate allowed by the laws of the State, Territory, or

District where the bank is located.” Since the Nebraska bank operated no branch banks in

Minnesota, the court found they were not located in Minnesota. Justice William Brennan noted:

“Minnesota residents were always free to visit Nebraska and receive loans in that State.” He

argues that the intent of the original 1864 National Bank Act was to create a “national banking

system”; any change in section 85 should therefore be made by Congress, not judicial

interpretation. 46

        Suddenly, national banks enjoyed an unprecedented advantage in competition against

retailers and state banks. Both federal and state regulators responded quickly to eliminate the

advantage. A provision added to the 1980 Depository Institutions Deregulation and Monetary

control Act (PL 96-221) extended the Marquette state-of-origin policy to state banks insured by

the Federal Deposit Insurance Corporation (Section 501). In order to untie the hands of retail

lenders and local finance companies that competed with banks, most states changed their usury

policies. By the end of 1981, 45 states had either raised their usury rate ceilings for consumer

loans or eliminated them all together. For states that retained usury laws, smaller local lenders

found that they could partner with out-of-state banks to provide loans that exceeded local interest

rate caps. Tax preparers began working with banks to make tax refund anticipation loans

(RALs). These RALs had APRs of 150% to 300%, and were used especially by low-income

recipients of the Earned Income Tax Credit. Pay-day lenders and auto title lenders, whose loans

             U.S. Supreme Court, Marquette Nat. Bank v. First of Omaha Svc. Corp., 439 U.S. 299

(1978), No. 77-1265,

        439 U.S. 299.
ranged up to 390%, also partnered with out-of-state federally chartered and FDIC banks to

exempt themselves from state usury laws. 47 The Consumer Federation of America lamented:

“The result is a set of usury laws that should make a finance company proud.” 48

       Marquette v. First of Omaha left open a range of questions about the scope of state

regulatory authority over national banks that would be settled only during the 1990s. In 1996, the

Supreme Court heard the appeal of Smiley v. Citibank, in which a New Jersey man had argued

that the late fees charged by Citibank violated the state law restricting late fees. Banks, which in

the debates leading up to the 1968 truth in lending legislation had argued that fees should be

treated separately from interest charges, now argued that they were equivalent to interest and

thus covered by the Marquette v. First of Omaha precedent. When the Supreme Court found in

favor of Citibank, credit card use of late fees expanded, from an average late fee of $12 in 1996

to $29 in 2002. 49 Similar cases explored the ability of state regulators to dictate minimum

repayment rates. In 2000, the California legislature passed legislation requiring credit card

issuers to include a warning on their statement that paying only the minimum balance would

increase their principal. In ABA v. Lockyer (2002) the American Bankers Association sued the

Attorney General of California, alleging that the law infringed on a different provision of the

            Michael S. Barr, “Banking the Poor,” Yale Journal of Regulation 21 (2004), pp 161-

            RLDM, Box 212, folder 24, Letter from the Consumer Federation of America to the

Senate Banking Committee, June 17, 1982.
            Furletti, “Credit Card Pricing Developments and Their Disclosure,” Fed Discussion

Paper, January 2003.
National Bank Act that provides for “all such incidental powers as should be necessary to carry

out the business of banking.” 50 The ABA argued that cost of printing such a warning made the

law “overly burdensome” on banks, and was therefore preempted by the NBA itself.

           In 2004 the Office of the Comptroller of the Currency issued new rules that preempted

states regulation of a wide range of nationally chartered bank activities, including use of non-

interest charges, credit account management, mandatory disclosures, and interest rates and fees. 51

By the time of the 2004 rules, it had become clear that federal preemption of state bank

regulations was being treated expansively. Mark Furletti writes: “The agency [OCC] essentially

declared that states have little or no authority to impose any consumer-protection-oriented

regulation on nationally chartered banks and that any such regulation is the province of federal

law.” 52

Liberalization in France

           Between 1984 and 1987, the French financial sector was dramatically liberalized. This

included the elimination of ‘encadrement de crédit’, the privatization of previously nationalized

banks, and a shift in authority over banks from the finance ministry to the competition ministry.

For both banks and finance companies, liberalization would change their approach to consumer

lending. For banks, liberalization would initiate a period of rapid expansion into the personal

loan business, followed by failure and retrenchment. For consumer finance companies,

                National Bank Act, Section 24 (seven).
                Mark Furletti, “The Debate Over the National Bank Act and the Preemption of State

Efforts to Regulate Credit Cards,” Temple Law Review 77 (2004).
                Furletti, “Credit Card Pricing,” p 16.
liberalization opened up the market for revolving credit accounts of the kind that American

consumers had long enjoyed.

       France’s banks had traditionally avoided consumer credit. The few times when they had

moved into the area, as in the early 1970s, they had failed. Close administrative guidance from

the Banque de France had also made it difficult for banks to compete in the sector. As recently as

1982, the governor of the Banque de France had written to all of the banks asking that they

extend no more personal loans. 53 This sort of stop-go administrative oversight made it difficult to

justify the significant investment large-scale consumer lending would require. More importantly,

France’s postwar indicative planning had relied on banks to finance its projects, and banks

became dependent on these investments for their profits. Financial liberalization changed this.

Firms in France began to seek non-bank sources of capital. They increasingly began to look to

the stock market, to factoring, and to self-financing through retained earnings. 54 Bank corporate

lending fell from 159 billion francs in 1982 to 109 billion in 1983, corresponding to a steep rise

in interest rates. Yet, even as interest rates began to fall off, corporate borrowing continued to

decline. By 1986, outstanding corporate loans were down to 75 billion francs.

            François Gomez, «La vente à crédit aux particuliers, » Gestion et Technique Bancaires

451 (June 1985), p 597.
            Richard Deeg and Mary O’Sullivan, “The Financialization of Europe,” paper presented

at the Annual Meeting of the American Political Science Association, Philadelphia, 2006.
       Liberalization and the decline in corporate lending pushed banks to look for other sources

of revenue, and consumer lending became a focus of intensive interest. 55 As one observer noted :

"Banks live from lending. Once they realized that companies had less and less need for them,

they turned to households with increasing ease as credit was liberalized." 56 France’s large banks

allocated large blocks of funds to a new push to extend their consumer lending business. Credit

Agricole alone allocated an additional 20bn francs to consumer credit in 1986 and 1987. 57

Through this push, they were able to increase dramatically their share of the consumer lending

market. From financial liberalization in 1984 to 1987, their share of the total consumer lending

grew from 18% to 25% 58

       They accomplished this in three ways. First, they offered lower rates than the consumer

finance companies. Whereas the latter were charging in the range from 15% to 18.25% at this

time, banks began offering rates from 12% to 16%. 59 They justified these lower rates based on

            Jean-Pierre Robin, "Credit a la consommation, les français s'americanisent," La Vie

Française, June 21, 1987.
            Françoise Fressoz and Philippe Boulet-Gercourt, «Les Français flambent à crédit, »

Libération, August 13, 1987.
            Florence Paricard, "Consommation: le taux de crédit doit baisser," La vie française,

March 30, 1986.
            Françoise Fressoz and Philippe Boulet-Gercourt, «Les Français flambent a crédit, »

Libération, August 13, 1987.
            Françoise Fressoz and Philippe Boulet-Gercourt, «Les Français flambent a crédit, »

Libération, August 13, 1987; Journale des Finances, May 30, 1987.
their lower cost of capital, but also on their detailed knowledge of their clients, many of whom

were long-term depositors. They also began aggressively expanding their credit agreements with

retailers. For the first time for a French bank, BNP in 1986 allied with the retailer Euromarché to

provide a Carte bleue credit card co-branded to the Euromarché. 60 Third, France’s big banks

started providing their best customers with access to revolving credit accounts. Traditionally,

banks lent in the form of personal loans with fixed repayment schedules. In the fall of 1987, BNP

launched Crédisponible, with revolving credit at 15.8%. Credit Lyonnais followed in September

1987 with Crédilion. Societé Génerale offered Crédiconfiance. 61 Credit Agricole formed a

consumer finance affiliate called Unibanque to offer revolving credit.

        As banks entered the consumer lending market, credit growth boomed. Non-mortgage

household debt in France rose from110 billion francs in 1984 to 370 billion francs in 1989, the

peak of the credit boom. Over the same period, the share of indebted households rose from 39%

to 53%. 62 Average household debt as a share of disposable income increased from 3.5% to 7%. 63

By 1990, the credit boom had dramatically slowed, from annual growth of 30% in 1987 to 20%

             Florance Pricard, « La Carte bleue fait credit, » La Vie Française, October 12-18,

             Florance Pricard, « La Carte bleue fait credit, » La Vie Française, October 12-18,

             « Les Français et le crédit," Figaro-Magazine, April 8, 1989, p 94.
             Roger Leron, Rapport surt l’application de la loi No 89-1010 du 31 Décembre 1989

rélative a la prévention et au réglement des difficultés liées au sûrendettement des particuliers et

des familles, 1990, p 7.
in 1988, 16% in 1989, to 2% in 1990. 64 Some attributed the reduced growth to the effect of the

first Gulf War; others thought that it was simply an artifact of France having “caught up” after

years of credit restraint. 65 The most important cause, however, was the failure of France’s large

banks in consumer lending.

       The banks failed for two main reasons. First, banks were not specialists at consumer

lending. They had not made the investments in skills or equipment that was required for success

in consumer lending. As Jean-Christophe Goarin, head of the consumer finance company S2P

(Groupe Paiement Pass), explained: "The handicap for banks is that they have too many

products. We have only a few, but we present them well and they are well adapted to the needs

of our clients." 66 Second, while consumer finance companies had embraced the idea of tracking

past repayment as a predictor of future repayment, banks were still relying on their relationships

with their depositors. They assumed that, because they had had known clients for years, that they

would be have low repayment risks. What they found was that they faced higher default rates

than did consumer finance companies. The high rate of defaults was aggravated by their effort to

expand their customer base rapidly, which led some banks to offer loans indiscriminately. Pierre

Marleix, head of the trade union-affiliated consumer group AFOC, explained that banks "were

            Le Figaro, May 30, 1991.
            La Tribune de l’Expansion, April 8, 1991.
            Catherine Rigollet, "Le Bazar des produits financiers," L'Express, March 6, 1987.
increasingly unconcerned about finding reliable borrowers; they simply wanted to increase their

loan volume in order to be big. " 67

       By the early 1990s, most banks had already moved to close smaller and delinquent

accounts, and it was clear that they would not be leading players in consumer lending, at least

not directly. Those that continued to lend did so mainly to their affluent customers. 68 Banks like

Credit Agricole that had created their own consumer finance affiliates closed those. 69 What they

did instead was begin to acquire existing consumer finance companies. BNP Paribas bought

Cetelem. Credit Agricole bought Sofinco. GE Capital bought Sovac. Credit Lyonnais bought

‘Citifinancement’ from Citicorp France. 70 The failure of bank lending was good for the dedicated

consumer finance companies. In 1992 alone, Cetelem stock rose 60% in value. 71

       The most enduring impact of liberalization was the shift of consumer credit from

installment credit and personal loans toward revolving accounts tied to payment cards. Until

liberalization, the active management of credit volumes required by ‘encadrement’ meant that

open or revolving accounts were difficult to operate. Not only did consumers control the amount

            Françoise Fressoz and Philippe Boulet-Gercourt, «Les Français flambent a crédit, »

Libération, August 13, 1987.
            Philippe Reclus, « Les grands réseaux s’alignent sur le taux à 9%, » la Tribune

desfosses, September 7, 1993.
            Le Figaro, May 30, 1991.
            Renaud de la Baume, “Les banques peinent sur le marché du crédit à la

consummation,” La Tribune de L’Expansion, 26 march 1992.
            Roland Laskine, « Au mieux de sa forme, » La Vie Francaise, January 9-15, 1993.
of credit they used, they also tended to rely more heavily on their revolving accounts at exactly

the times when the Banque de France was reining in lending. All of this changed with the end of

encadrement. In 1985, Cetelem launched their new Aurore card. 72 For Cetelem, the Aurore card

was a way to counter the aggressive move by banks into personal lending. They offered it as a

co-branded card that allowed them to reinforce their ties with retailers. They also already

understood the technology, which they had showcased twenty years earlier with their ‘credit en

poche.” By end of 1987, the Aurore card already had 600,000 members. 73 Other consumer

finance companies quickly launched their own revolving products. They included Accord by

Auchan, Plus by Cofinoga, and Pass by S2P. Credit Commercial de France (CCF) created an

innovative revolving account with rates tied directly to the interbank lending rate. 74 Amid all of

the new entrants, the Aurore card would dominate the field. By 1995, ten years after its launch,

Cetelem’s Aurore card was used by 5 million French (and 1 million other Europeans). 75 By

2000, there were 13 million Cartes Aurore in use, with 250,000 points of sale. 76 Interest rates on

            The name « aurore » meant dawn, but also sounded like « or-or », or « gold-gold ».

Catherine Rigollet, "Le Bazar des produits financiers," L'Express, March 6, 1987.
            Jean-Philippe Vidal, « L’explosion du crédit à la consommation, » Tribune de

l’économie, October 27, 1987.
            Florence Paricard, "Consommation: le taux de crédit doit baisser," La vie française,

March 30, 1986.
            Figaro-Economie, June 19, 1995.
            Philippe Herail, “Cetelem: une histoire des valeurs, un avenir,” Cetelem Historical

Archive, August 2002.
these loans ranged from 12.4% to 16.1%. 77 Some banks also began launching Cartes Bleue tied

to revolving credit accounts, including Alterna by Societé Generale, and Provisio by BNP.

Interestingly, bank revolving accounts that were initially marketed as luxury products did not

remain that way for long.

       By the late 1990s, France’s banks were facing a surge in personal loan non-payments.

One solution, they discovered, was to convert the distressed personal loan accounts into

revolving accounts. With longer repayment periods and a flexible repayment schedule, write-offs

could be avoided. Indeed research at the time found that only 1.1% of revolving accounts ended

up in legal recovery, compared to 3.2% for traditional small personal loans. Driven first by

luxury and then by necessity, revolving lending grew from 8.5% of consumer borrowing in 1996

to 17% in 1990. 78 By 2000, 26% of all non-mortgage household borrowing would take the form

            Le Figaro, March 9, 1987.
            Roger Leron, Rapport surt l’application de la loi No 89-1010 du 31 Décembre 1989

rélative a la prévention et au réglement des difficultés liées au sûrendettement des particuliers et

des familles, 1990, Annex II..
of revolving credit. 79 For consumer finance companies, revolving loans accounted for 37% of all

loans. 80

        The emergence of revolving credit elicited concerns from the French public. Many feared

that it offered “easy credit” that would drive households excessively into debt. Bankers were

defensive, arguing that borrowers with access to revolving credit were able to be more financial

responsible because of the greater control that revolving credit gave them. Lenders tended to use

developmental and evolutionary metaphors when talking about their revolving credit clients.

LaJaques Lenormand, the director of the personal lending department at Crédit Agricole, was

typical in claiming "clients with access to revolving credit are far more adult, more evolved,

better informed and, in this sense, are better managers." 81 French consumers were evidently more

circumspect. A survey of French consumers in 1992 found that only 30% knew what revolving

credit was. Among those who were familiar with it, two-thirds called it “very useful.” Yet half—

             François Henrot, « Les instruments du financement à crédit de la consommation en

France, » in André de Lattre (ed.), La consommation et son financement en France et dans le

monde (Paris: Berger-Levrault, 19xx), pp 56-57.
             François Henrot, « Les instruments du financement à crédit de la consommation en

France, » in André de Lattre (ed.), La consommation et son financement en France et dans le

monde (Paris: Berger-Levrault, 19xx), pp 56-57.
             Figaro-Economie, May 14, 1990.
and therefore presumably some significant share of those who called it “very useful”—also

called it “a trap from which one never escapes.” 82

       A second concern was with the impact that the new revolving credit arrangements would

have on consumer protection in sales contracts. One of the core pillars of French consumer

protection was the 1978 “Scrivener” law. This law had the effect of linking credit and sales

contracts. Under the law, installment credit contracts were binding only with the delivery of a

product. Conversely, sales contracts were binding only with the approval of a loan. The law also

created a 7-day cooling off period for credit sales that required sellers to take back their products

if the buyer decided to nullify the credit contract. In practice, sellers commonly waited seven

days before making delivery to avoid having to take back used products if a customer opted to

reverse a credit purchase. 83 As French consumers moved away from installment borrowing and

toward revolving credit plans, the benefits of these consumer protections diminished. Yves le

Duc, general secretary of the French consumer group CSCV, writes: "The Scrivener law

…established a link between the credit contract and the sale of a product. This link is

dissolving….The large retailers and banks that offer their own cards no longer need to know if a

            Paul Defourny and Josette Bienfait, Données d’image sur le credit, Paris: Cetelem-

BVA, 1992, p 7.
            Law 78-22, 10 January 1978, transcribed as articles 311 and 313 of the French

Consumer Code; La Croix, May 29, 1976.
good is delivered, or if it is in good condition…whatever the case, payment is required. This is

an important regression with respect to the law of 1978." 84

Re-Regulation in France

       From 1980 to 1986, the share of French households with outstanding consumer debt grew

from 33% to 40%. 85 During the same period, the household savings rate fell from 17.5% to

12%. 86 The growth in the use of credit led to a wide-spread public debate about the causes and

consequences of over-indebtedness. Some saw it as a ‘demoralization’ of credit that allowed

households to behave more like companies. 87 The dominant interpretation, however, was that

French families were using consumer credit as means to compensate for stagnating purchasing

power. 88 Yves Ullmo, secretary general of the National Council on Credit, described the effect:

"Since [1983], disposable income has increased little. To maintain a certain standard of living,

use of credit has been a solution.” 89 Policymakers on the left and right worried about the

            Philippe Lebellec, « Crédit a la conosmmation : prudence, » La Croix, November 11-

12, 1987.
            Le Monde, June 19, 1987.
            Le Monde, January 5, 1989.
            Martine Gilson, « Credit : les français craquent, » Le Nouvel Observateur, January 26-

February 1, 1989.
            Martine Gilson, « Credit : les français craquent, » Le Nouvel Observateur, January 26-

February 1, 1989.
            Roger Leron, Rapport surt l’application de la loi No 89-1010 du 31 Décembre 1989

rélative a la prévention et au réglement des difficultés liées au sûrendettement des particuliers et
sustainability of this strategy. The result was a series of decisions that progressively restricted

access to credit for French consumers.

        The first target of reregulation was the so-called ‘free credit’ that emerged with the

relaxation of government credit ‘encadrement.’ These were sales loans that were offered without

any direct interest charge. Their goal was to increase sales. In the furniture sector, which quickly

embraced the practice, 20.5% of all credit offered in 1984 was without interest. As a

merchandising strategy, fee credit seems to have been effective. The two furniture retailers that

were most aggressive in offering furniture credit for free—Galeries Lafayette and Castorama—

were able to consolidate their position as France’s leading furniture retailers. Galeries Lafayette

at the time was offering 94% of its credit sales at zero interest. 90 For the government of François

Mitterrand, who in 1984 was struggling to reign in inflation, free credit raised the old fear that

unbridled consumer credit would stoke the flames of inflation and so hurt competitiveness.

Laurent Fabius, Prime Minister under Mitterrand, attacked the practice as “pushing customers to

purchase using credit they don’t really need.” 91

        The Banking Law of 1984 reined in free credit sales. Retailers could no longer advertise

free credit terms outside of the point of sale. More importantly, when free credit terms were

offered, retailers were required to display a lower price for goods when they were purchased

des familles, 1990, p 7; Jean-Marc Biais, « Le Piège à mauvais payeurs," L’Epress, February 3,

1989, p 62.
             Direction du Développement, “Credit Gratuit,” September 4, 1984.
             Laurent Chavane, “L’enterrement discret du ‘crédit gratuity’”, Le Figaro, 23 July

without free credit. This lower price was set by a formula, based on the average interest rate plus

50%, that for average periods of credit gave prices that were roughly 20% below the ‘free credit’

price. 92 Interestingly, most consumer groups supported the new law, as did small retailers, who

did not have the financial means to compete with ‘free credit.’ 93

       The second move to restrict credit access focused on the practice of credit rating. In 1988,

Jacques de la Rosière, governor of the Banque de France, announced their intention to create and

manage a mandatory listing of credit data on all consumer borrowers. This would re-establish a

system created in 1946 in which banks reported all of their loans to the central bank. While this

process had been retained for commercial lending, consumer lending had made exempt as part of

the credit liberalization campaign of the 1950s. Consumer groups strongly supported the creation

of a positive or ‘white’ credit database, in part because they thought it would force France’s

banks to behave more responsibly. Consumer representative Louis Mesuret warned: “[the banks]

do not respect their obligation of prudence and offer credit willy nilly (‘atort et a travers’).” The

idea was to allow all lenders to know the total debt held by credit applicants so that they could

better evaluate the likelihood of repayment. The Association of French Banks came out strongly

against the white list idea, which they saw as an attempt by the non-bank finance companies to

gain access to their own clients. Banks reasoned that the long-term relationships they had built

with depositors would give them an advantage in offering them loans. Consumer finance

companies, by contrast, had traditionally offered loans via retailers, and thus had very little direct

            François Renard, “Le credit gratuity va pratiquement disparaître,” Le Monde, 25 July

1984, p 19.
            Laurent Chavane, “L’enterrement discret du ‘crédit gratuity’”, Le Figaro, 23 July 1984
contact with their customers. By sharing data on their customers, banks stood to lose valuable

information and gain very little. 94 France’s large banks are reported to have used their political

influence to get France’s powerful data privacy body, CNIL, to recommend against the creation

of a positive credit listing of the sort proposed by de la Rosière.

       With the failure of a white list, France’s association of finance companies (Association

des sociétés financiers, ASF) launched their own debtor black list in October 1988. This list,

which was voluntary to its members, included only borrowers who were at least 3 months late on

their payments. Members were encouraged to check applicants against this list before making

loans. All but the largest lenders participated, accounting for 35 of the 37 member organizations

and 70% of all finance company lending. 95 The Banque de France responded the following year

by creating their own black list. Unlike the ASF list, all consumer lenders were required to check

loan applicants against the official government list. Unlike the proposed white list, it included

only non-payment information. The database, called the Fichier national des incidents de

remboursement des credits aux particuliers (FICP), was accessible only to credit providers and

to individual debtors who wished to view their own record. 96 In practice, the FICP database

became a no-credit black list for borrowers. In 1991, it included 800,000 late payers; by 2003,

that number had grown to 2.3 million.97

            Libération, December 7, 1988.
            Le Monde, December 8 1988.
            Chatain, Pierre-Laurent and Frédéric Ferrière, Surendettement des particuliers, Paris:

Editions dalloz, 2000, p 188.
            Libération, March 20, 1993.
       Since its creation, debates periodically surfaced concerning the potential advantages of

collecting and distributing more extensive ‘positive’ credit data, including information on

outstanding loans, taxes, income, and assets. In one proposal, which came before the National

Assembly in 2005, the center-right UDF party argued that a positive registry would give

financial institutions a better sense of total lending and help them to better assess a consumer’s

ability to repay. France’s financial institutions were divided on the proposal. Some, including the

consumer lending institution Cofinoga, argued that positive data on potential borrowers would

help them to avoid adverse selection in selecting customers, thereby reducing both defaults and

credit rationing. But many other financial institutions, supported by France’s association of

financial companies (ASF), argued that a positive rating system would only assist foreign

financial firms—like UK-based lender Egg—to identify and exploit new clients in France. 98

France’s consumer groups generally agreed with this view, worrying that a positive list would

become a tool for more aggressive commercialization of credit, leading to higher levels of

consumer indebtedness. 99 CNIL again blocked the proposal as an illegitimate use of private data.

       The third major regulatory move was in the area of bankruptcy. In 1989, France adopted

new legislation (the Law on Prevention and Regulation of Individual and Household Over-

indebtedness) that introduced for the first time a civil personal bankruptcy procedure. The goal

            Catherine Maussion, “Credit: tout le monde dans le même fichier?” Libération, 5

October 2002.
            Bertrand Bissuel and Anne Michel, “Les établissements de credit accusés de favoriser

le surendettement, Le Monde, 28 April 2005; Sylvie Ramadier, “La bataille du credit à la

consummation,” Les Echos, 21-22 January 2005, p 8.
of the project was to prevent the over-indebted from becoming permanently marginalized in

society. Until 1989, France provided no useful recourse to over-indebted consumers. Creditors

with defaulting borrowers brought their cases to the courts, which summoned the defaulting

consumers, who often simply did not appear. Courts would then grant creditors the right to

repossess property and attach salaries, typically leading to eviction. 100 For creditors, this ad hoc

response created coordination problems, as creditors rushed to secure insufficient assets. With

liberalization of consumer credit in 1984-1987, a rising incidence of over-indebtedness led

France’s consumer and finance associations to negotiate a novel solution. The law of 31

December 1989, called “Neiertz law” after France’s minister of consumption, Véronique Neiertz,

created a new administrative instrument that would work beside the legal system to help resolve

cases of consumer over-indebtedness. The law would be reformed several times as the French

government has moved slowly toward a workable system that provided for the discharge of

personal debt.

       The core of the new system was a set of Departmental Commissions for Over-indebted

Individuals’ managed by France’s central bank and to which any consumer could apply for debt

restructuring or relief. The commissions had broad authorities. They could suspend payments for

up to two years, restructure the payment period for loans, and modify interest rates. The new

repayment schedule was based on the Commission’s assessment of the “minimum vital income”

to meet the claimant’s most basic needs, and repayment plans could last up to but not longer than

ten years. Formally, Commissions could only propose repayment solutions. If both parties to the

negotiation did not voluntarily accept it, the case went before a judge, who, from 1995, had the

             “Un peu d’oxygène pour les familles surndettées,” Nous 103, March-April 1990, p 9.
right to enforce the recommendations of the Commission on both parties (or, in rare cases, to

design a new repayment scheme). A series of studies conducted in the 1990s showed that most

cases went to the full 10 years, that interest rates were reduced on average from 13% to 9%, and

that average monthly payments were reduced from 6,000 francs to 3,800 francs. 101 Within

certain limits—business-related debt was excluded, and consumers had to show ‘good faith’ in

presenting their case to the commission—nearly all cases were accepted. The number of cases

filed with the commissions grew dramatically since their founding. From 90,000 cases in their

first year, 1990, the number of cases had risen to 190,000 by 2004. 102

       Two features of the 1989 law were distinctive. First, the commissions included no formal

requirement for credit counseling, although in practice consumer and family associations often

helped indebted households to navigate the administrative procedures. Second, private credit

intermediaries were banned in France, although credit consolidation loans remained legal. 103 In

assessing cases, the commissions distinguish first between ‘active’ cases of over-indebtedness, in

which consumers have simply taken on too much debt, and ‘passive’ cases, in which external

causes have made it impossible for the borrower to continue paying. In 2001, active cases

             Balaguy, Hubert, Le crédit a la consummation en France (Paris: Presses

Universitaires de Paris, 1996), p 106-7.
             The commissions were administrative bodies. Each had four members, representing

the Banque de France, the treasury of the department, a consumer association, and a

representative of a lender association. In 2003, two non-voting members were added: a social

worker, and a lawyer. INC, Le Dispositif Juridique (internal documents).
             Familles de France magazine 664, September 2000.
accounted for 36% of all cases. By 2004, active cases accounted for 27% of all cases. Because

the total number of cases had grown substantially, the number of active cases has stayed nearly

the same – approximately 50,000 cases per year. In 2004, the most important sources of passive

cases were unemployment (30.8%), divorce or separation (14.7%), and sickness or accident

(10.8%). 104 As the number of cases seen by the commissions each year has risen, the share of

‘active’ cases has consistently fallen.

       Consumer creditors were initially highly skeptical of the Neiertz procedure. Initially, in

1990, only 45% of solutions proposed by the commissions were accepted by creditors. Paul

Defourny, head of Cetelem, estimated in 1992 that 50,000 of the 160,000 cases heard up until

that point represented instances of clear cheating by borrowers—borrowers who had gone from

lender to lender in order to borrow as much as they could without intention of repaying. He

estimated that the loi Neiertz had cost the lending industry 1bn francs in its first two years. As he

describes it, “the [Neiertz] law…moralized our profession by creating a sort of deontological

code.” 105 Yet as cases of over-indebtedness rose, and as the commissions showed that they were

able to produce workable solutions, industry acceptance of commission proposals rose, to 75%

by the end of the 1990s.

       The final major area of consumer credit re-regulation concerned usury. The 1989 Neiertz

law had redefined usury to accommodate different risk classes while still limiting lending outside

of a legally defined range. (See chapter 7 above.) The effect was to restrain the more costly kinds

             INC, Le surendettement en chiffres (internal publication)
             Renaud de la Baume, “Surendettement: un banquier dénonce les ‘tricheurs’,” La

Tribune a l’Expansion, 19 March 1992.
of loans while still allowing lenders flexibility in adjusting their rates over time. 106 From the

mid-1990s, the interest cap for the most expensive class of loan, the revolving credit accounts,

ranged between 20% and 25%. In practice, this meant that certain smaller loans to riskier classes

of borrowers were not financially viable without cross-subsidy. A cap at 25% meant that some

people would not have access to credit. As in the past, however, what mattered for usury caps

was the political will to enforce them. The first test for the Neiertz law came in 1996.

        In that year, the UK-based Thorn group launched its first rent-to-sell store, called “Crazy

George’s,” in Bobigny, France. It followed with a second site in Havre. The store format was

based on the company’s highly successful Rent-a-Centre chain, launched in 1992 in the United

State, and the equally successful ‘Crazy George’ rent-to-own chain in the UK. The idea of the

store was to offer goods under a rental contract that would eventually lead to consumer

ownership. Their French stores targeted communities with high concentrations of poor and

elderly, and where credit was not readily available. The effective interest rate on the rent-to-own

purchases ranged up to 56%. Because the rental was not formally a sale, it did not fall under

France’s usury restrictions. 107

        Initially, Thorn had envisioned offering the rent-to-own service through its existing

French consumer electronics rental company, Visea. Beginning in 1994, Visea began posting

              Gaudin, Michel, Le Crédit au pariculier: Aspects économiques, techniques, juridiques

et fiscaux, Paris: SÉFI, 1996.
              Crazy George’s 40 outlets in the UK operated on far lower effective interest rates,

ranging between 20% and 25%. This was likely due to the more highly competitive consumer

lending market in the UK, which gave Crazy George customers other borrowing options.
billboards with slogans that read: "If we lend only to the rich, why did we invent credit?” and “If

we deny credit to those who are unemployed, what good is solidarity?" 108 The reference to

solidarity was a direct attack on the French idea of equal citizenship without equal access to

credit. Thorn finally opted to introduce the rent-to-own (location avec option d’achat, LOA)

format as an American import. To launch its first Crazy George’s store in Bobigny, they paraded

around town in convertibles festooned with balloons and American flags. The French media

quickly focused on the new format as an ‘Anglo-Saxon” import. Public interest heightened when

the first store was shut down by France’s commercial police (the DGCCRF) on the grounds that

they were not accurately reporting the final total price on their goods.

       For Crazy George’s customers, many of them retired, on a pension, and without access to

traditional consumer credit, Crazy George’s provided their only possibility to purchase large

household goods. Unlike consumer lenders, Crazy George’s accepted social payments as income

to qualify for a loan. In fact a third of all customers reported no household salary income. To

manage repayments, the stores sold only to customers within a 5 kilometer radius and insisted

that customers drop off their payments in person and on a weekly basis. All applicants were

required to give the names and telephone numbers of three friends and two relatives, so that

application information could be confirmed. 109 These measures kept default rates below 1%. 110

A survey of customers found that 66% were first-time buyers of basic household goods like

             La Croix, January 18, 1994.
             Odette Terrade, “Projet de loi de finances pour 1998 - Consommation et concurrence,”

87/9, French Senate, 1997.
             “Crazy George’s Déclare forfeit,” LSA 1594, August 27, 1998.
televisions and computers. 111 Critics argued that it was wrong for the poor to pay over twice as

much for products. And a public poll found that 70% of French opposed the rent-to-sell

format. 112 Behind the opposition was a coalition of the Christian right and labor left. France’s

Catholic newspaper, La Croix, took a hard line: "consumer to be condemned, and the

old Christian tradition of usury, meaning a total prohibition on the charging of interest, not only

abusive interest rates, should be retained.” 113 The union-affiliated consumer group CSCV

(Confédération syndicale du cadre de vie) brought a law suit against them for exceeding France’s

usury cap. 114 In 1998, Thorn was acquired by Nomura Investment, and the French stores were


Credit and the Third Way

          In the United States, consumer credit was also closely associated with concerns about

welfare. The difference was that American policymakers saw credit access as promoting

solidarity. The basis for this consensus was an evolving cross-class coalition in favor of credit

extension. The coalition proceeded in four stages. In the 1920s, newly efficient installment

lenders were embraced as a means to save household borrowers from the extremely high interest

rates and harsh collection techniques practiced by unlicensed loan sharks. In the 1950s and

1960s, credit boosted demand for household manufactured goods. This was seen to increase

industrial production and jobs, stimulate new business investments, and help workers to secure

                CSA, survey of 157 customers, October 1997.
                Le Monde, 25 November 1996.
                Hugues Puel, “Crazy George’s met l’éthique au défi,” La Croix, 11 December 1996.
                Crazy George’s à contrecourant, Humanité, February 20, 1998.
the benefits of the industrial society they were helping to build. During the 1970s, both women

and blacks made credit access a successful part of their campaigns for equal rights. This new

access came at a low economic cost, as rising inflation caused by the two oil shocks dramatically

reduced the real interest rate on fixed interest rates consumer loans. Across all of these first three

periods, industrialists tended to embrace credit as a way to de-radicalize the work force; liberals

saw it as a means to support the poor without raising new government resources.

       By the late 1980s, the seeming benign relationship between working class aspirations and

consumer credit turned more sinister. A combination of deregulation, new technologies, low

inflation, and low interest rates contributed to making consumer loans highly profitable for

lenders. Declining tax rates and revenues as a share of GDP and a new embrace of markets as a

means to promote social progress led left governments to embrace ‘third-way’ policies that

actively promoted credit market access as a social goal. These two trends increased American

households’ access to unprecedented liquidity, and they used it to maintain their purchasing

power. Under normal circumstances, this sort of debt-led strategy could not have lasted long. But

the late 1990s were not normal. An appreciation in house values and a relaxation of mortgage

lending terms led American households that were already in debt to tap their homes as a new

source of equity. What appeared at the time as a vindication of the “third way”—with rising

house prices seemingly buoying the wealth of even modest homeowners—was in reality only a

temporary patch to fill growing wedge between stagnating real wages and rising real


       It has commonly been asserted that Americans in the early 2000s treated their houses as

ATMs, but this is only partly right. What they were doing in part was using their mortgages as

debt consolidation loans. Fully a third of all value extracted from the housing stock via cash-out
refinancing or home equity loans was used to pay down existing consumer debt. At the time, this

transfer of debt from high-interest rate consumer loans to lower interest rate mortgage loans

seemed like a responsible financial decision. But it created two problems. First, households did

not stop their consumer borrowing, even as they transferred the debt to their mortgages. Second,

the depletion of household assets pushed up the leverage in the US housing market, making

family balance sheets especially sensitive to fluctuations in house prices. In a real sense, the US

housing crisis that would precipitate the 2008 financial crisis started as a consumer debt crisis

with its roots in the 1990s. Why was this allowed to happen?

       The idea that free access to financial markets could play a role in generating social

equality dominated the ‘third way’ politics of the Bill Clinton presidency. To some degree, the

focus on financial markets as a boon to the working poor simply made a virtue of necessity. As

tax rates fell, rising social demands met declining government revenues, and the political left

turned to the markets to finance their social goals. They pushed for state lotteries, with winnings

going for progressive causes like education. They advocated for legalizing betting in order to

sustain social spending in the face of declining revenues. They embraced the housing and

financial assets as a means to invest for future retirement—an approach that came to be known as

“asset-based welfare.” In the UK, where a new-left synthesis was also emerging under the

Labour government of Tony Blaire, asset-based welfare took the form of Individual

Development Accounts—investment accounts opened for every newborn child. In the US, the

emphasis of asset-based welfare was on tax-deferred private retirement savings and on home


       The fall of the US housing market sounded the death knell for the Third way left.

Through the early 2000s, the high mortgage leverage of poor families was a virtue. The high
degree of leverage made possible by new low- or no-down-payment mortgages meant that as

house prices went up, returns to those least able to make down-payments were disproportionately

high. 115 Household debt apologists noted that although debt was rising as a share of household

income, it was falling as a share of household wealth. The decline in house values of the late

2000s and the high incidence of foreclosures and bankruptcies took away what leverage had

given them. For America’s lower middle class, the financial crash of 2008 stripped away the

wealth of a generation of new home owners. The greatest welfare setback in a generation had its

roots not in the give-and-take of wage bargaining or labor market reform, but in financial

markets that allowed middle class Americans to gamble with their homes. For Third Way

Democrats who had embraced markets as a route to social inclusion, these reverses represented a

startling betrayal. They also pointed to a deeper problem. If extending financial market access

was regressive in its consequences, then the left faced a stark dilemma: one could extend market

access at the cost of social and economic cohesiveness, or one could increase social and

economic cohesiveness at the cost of market access.

             Anne Kim, Adam Solomon, Bernard L. Schwartz, Jim Kessler, and Stephen Rose,

“The New Rules Economiy: A Policy Framework for the 21st Century,” Third Way Report,

2007, p 8.

To top