1997 Survey of Credit Underwriting Practices

The Office of the Comptroller of the Currency (OCC) conducted its third annual Survey of Credit Underwriting Practices during the second quarter of 1997 to identify trends in credit risk within the national banking system. The questionnaire-based survey addressed changes in lending standards over the previous year for the most common types of commercial and retail credit offered by national banks. The OCC examiners-in-charge of the 80 largest national banks were asked to respond to the survey based on their firsthand knowledge of the banks they supervise.

Primary Findings

•A discernible shift in underwriting standards has occurred since the 1996 survey, with most of the surveyed banks now described as having "moderate" or "liberal" underwriting standards.

Across the 12 categories of credit covered by the survey, examiners described 38 percent of the surveyed banks as having "conservative" underwriting standards, down from 52 percent in the 1996 survey. "Moderate" standards were reported at 54 percent of the surveyed banks, compared with 45 percent a year earlier, and "liberal" standards were reported at 8 percent, up from 3 percent in 1996.

•The trend toward eased standards for commercial loans noted in last year’s survey has continued. Examiners reported competitive pressure was again the primary reason for easing of standards.

A majority of banks (59 percent) were found to have eased lending standards for one or more types of commercial loans. The trend was most pronounced in middle market lending, syndicated and national credits, and commercial real estate lending.

Examiners reported loan pricing that was less favorable to banks in all categories of commercial loans. Structural concessions to borrowers, including financial covenant, guarantor and collateral requirements, and lengthened maturities were also cited as evidence of a continuing trend toward eased commercial loan standards. As with last year’s survey, competitive pressure was again cited as the predominant reason for this continued easing of terms.

•More banks are tightening their lending standards for retail loans.

With the exception of home equity and residential real estate lending, more banks had tightened lending standards in their retail portfolios than had eased them. The tightening of retail lending standards was most pronounced in credit card portfolios, continuing and broadening what was predominantly a trend among just the largest banks in the 1996 survey.

• The level of inherent credit risk continues to increase for most com-ponents of both commercial and retail portfolios.

Compared with a year ago, examiners reported that the level of inherent credit risk in the portfolios of most of the surveyed banks had increased for at least one, and generally more, of the loan product components.

In commercial loan portfolios, higher inherent risk was most frequently cited in the syndicated/national credit, commercial real estate and middle market components. Examiners cited competition and the desire for loan growth as the principal causes of increased credit risk.

The retail components most frequently reported as having higher inherent risk were credit card, indirect consumer and home equity loans. This was attributed to the combined effects of high growth, acquisitions, increasing levels of consumer delinquencies and bankruptcies, and higher loan-to-value ratios.

Retail Lending Portfolios
Credit Card Lending

For the 66 surveyed banks that were engaged in credit card lending, including several monoline companies, the 1997 survey results parallel those from 1996, i.e., the overall level of inherent risk continues to increase despite tightened standards at most of the banks. Among the 31 Tier I and II and the 35 Tier III banks with credit card portfolios, 39 banks (59 percent) had tightened standards in the last year, compared with 30 percent that had tightened in the 1996 survey.

Examiners cited increased score card minimums (25 banks) as the most common technique used to tighten standards. In addition, some examiners revealed that their banks had tightened specific scoring components, including debt-to-income, length of residency, and length of employment requirements.

According to the examiners, changes in economic outlook (29 banks), market strategy (20 banks), and the competitive environment (13 banks) were the primary reasons the banks had tightened standards.

Examiners characterized current standards for credit cards as moderate at 52 percent of the surveyed banks, conservative at 41 percent, and liberal at the remaining 7 percent.

Although many banks had tightened standards for credit cards, examiners cited other reasons for the increased level of credit risk, primarily trends in consumer delinquencies and bankruptcies.

Home Equity Lending (HEL)

Among the 74 banks in the survey that had been engaged in home equity lending, examiners saw a continuing trend toward eased standards. The 1997 survey found easing at 28 HEL lenders (38 percent), compared with 16 percent in the 1996 survey. Three banks (4 percent) had tightened standards, similar to 1996 results.

Expanding Credit Card Debt: The Role of Creditors and the Impact on Consumers

Report of the Consumer Federation of America

In February 1997, the Consumer Federation of America issued a report, "The Consumer Impacts of Expanding Credit Card Debt," prepared by Stephen Brobeck, executive director of CFA. The study assessed the impact of rising credit card debt on households and discussed the implications of this indebtedness for educators and policymakers. Its findings included:

•55-60 million households revolve credit card debts averaging more than $6,000 which require payment of about $1,000 in interest and fees annually;

•This escalating debt has been the most important factor recently in riding personal bankruptcies and other financial insolvencies;

•The single most important reason for this expanding debt has been aggressive credit marketing and credit extension by issuers who have been increasingly targeting low and moderate income households; and

•The most effective way to reduce consumer insolvencies would be for credit card issuers to lend to households only affordable amounts of credit, which usually would not exceed 20 percent of incomes.

This report, issued 10 months later, updates the first one. It supplies new information about credit card debt levels, the consumer costs of these debts, the role of a few large institutions in the creation of these debts, the effect of proposed bankruptcy reforms on debt levels, and the most effective way to reduce these debt levels.

Credit Card Debt Levels

Since the first report was issued, the level of revolving consumer credit levels has increased steadily. By the end of September, according to the Fed, it stood at $526 billion, an increase of 14 percent in nine months. In addition, in June the Fed estimated that only 9 percent of credit card debt was paid off before interest charges were incurred. Thus, to estimate the level of credit card debt incurring interest charges, one must only reduce the $526 billion figure by 14 percent (the 9 percent plus the estimated 5 percent that is not credit card debt), yielding a figure of $452 billion.

The first report estimated that 55-60 percent of all households carried credit card balances incurring interest charges. There is no reason to believe that these percentages have changed appreciably over the past year. Since there are now about 100 million households in the United States, that would represent 55-60 million households. Dividing $452 billion by 55-60 million yields about $7,000-$8,000 of credit card debt per household with a revolving balance.

The Consumer Costs of Credit Card Debt

The first report estimated that, in 1996, consumers paid at least $55 billion in interest on credit card debts. More recently, RAM Research estimated that, in 1997, cardholders will pay $63 billion in interest charges.

These expenses do not include the interest and fees paid on credit card debt that has been refinanced as home equity debt. Using Federal Reserve data, we estimate that at least one-third of the roughly $250 billion in home equity debt represents refinanced credit card debt. That would suggest that at least $6 billion in interest on home equity loans is related to credit card borrowing.

There is considerable evidence that, as a class, lower middle-income households are most burdened by large credit card debts.

As reported in the first study, the typical household in severe financial difficulty belongs to the lower middle-income class and is burdened by high credit card debts. As indicated by research conducted by Michael Staten and published by the Credit Research Center (then at Purdue, now at Georgetown), chapter 7 bankrupts studied in 1996 had annual after-tax incomes averaging $19,800 and credit card debts averaging $17,544. As reported by the National Foundation for Consumer Credit, 1995 participants in consumer credit counseling service debt management programs averaged $25,680 in annual income and over $10,000 in credit card debts.

Role of Creditors in Credit Card Debt Creation

While several factors help explain recent growth in credit card debt, more aggressive marketing and credit extension by issuers appear to be the most important.

In 1997, according to data published in the October 1997 issue of Card Marketing, the marketing of credit cards continues to expand.

After declining from 2.7 billion in 1995 to 2.4 billion in 1996, credit card mail solicitations increased in the first half of 1997, with second quarter mailings of 881 million representing a record.

From 1995 to 1996, the hours billed to credit card issuers for telemarketing rose 30 percent (from 18.6 million hours to 24.1 million hours).

At the same time, card issuers were extending an increasing amount of credit. In March 1993, consumers had available $0.6 trillion in unused bank card limits. By March 1995, that amount had increased to $1.0 trillion. By the middle of 1997, these limits had grown to $1.6 trillion.

Was the extension of all this credit prudent? According to one Wall Street Securities analyst: "The banks and other credit card lending institutions...shot themselves in the foot by using some of the weakest and most pitiful loan underwriting techniques that I have ever witnessed." [Congressional Quarterly, October 18, 1997, p. 2536.] The best evidence of this imprudence was a rising bad debt charge-off rate—debt losses as a percentage of outstanding loans—that, according to the American Bankers Association, exceeded 5 percent in the second quarter of 1997 [Wall Street Journal, October 17, 1997, p. A3]. For purposes of comparison, in the 1970s and early 1980s, this charge-off rate averaged well under 3 percent.

What is not widely known, however, is that a relatively small number of banks extended much of this credit and suffered a significant percentage of the charge-offs. According to a Veribanc News Release of September 16, 1997, only 50 of the some 5,801 banks that offer credit cards dominate the industry’s credit extension and debt losses. These 50 institutions held 40 percent of all card debt outstanding. But their second quarter (1997) net charge-offs represented 59 percent of all credit card-related charge-offs by banks. Twenty-three of these institutions, which held 20 percent of all bank credit card debts, suffered an average charge-off rate for this quarter that was 140 percent higher than that of the rest of the industry—2.4 percent versus 1.0 percent.

Among the largest credit card banks of the 42 largest credit issuers, there was a substantial range of net charge-off rates for the period June 30, 1996 to June 30, 1997—2.1 to 9.0 percent. Four banks had unusually low rates. They were:

MBNA (Wilmington) 2.1%
Peoples (Bridgeport) 2.4%
Travelers (Newark) 2.7%
First USA (Wilmington) 2.9%

Five banks had unusually high rates. They were:

Mellon (Wilmington) 9.0%
Hurley State (Sioux Falls) 9.0%
Wells Fargo (Phoenix) 8.6%
First Union (Charlotte) 8.4%
Advanta (Wilmington) 8.2%