Board of Governors of the Federal Reserve System

June 21, 2024

Estimating Retail Credit in the U.S.

Jessica N. Flagg, Simona M. Hannon, Cisil Sarisoy, and Mark J. Wicks^*

In this note, we estimate the size of an understudied segment of the consumer credit market—retail credit. Retail credit consists of the amounts owed to retail stores by their customers for purchases made on credit extended in partnership with a financial institution.¹ Although retail credit is included in the consumer credit totals in Federal Reserve Statistical Release G.19, "Consumer Credit," a principal economic indicator providing the official estimate for consumer credit holdings in the U.S., it is not separately categorized on the release.² Using several credit bureau data sets, we provide an estimate showing that the retail credit outstanding totalled $130 billion, or over 2.5 percent of total outstanding consumer credit as of the end of 2023.³ Moreover, nearly 85 million individuals (or about one-third of the entire U.S. adult population with credit records) had a retail credit account, with a median outstanding balance per account of $194 and a median monthly payment of $29.⁴ Furthermore, more than 60 percent of retail credit balances were held by borrowers with nonprime credit scores.⁵ Retail credit is more than 90 percent revolving in nature.

In the remainder of this note, we present a method for estimating retail credit, and then we provide some estimates and additional details for the sector.

Estimating Retail Credit Outstanding

To provide a complete overview of the retail credit outstanding universe in the U.S. over nearly two decades, 2005:Q1 to 2023:Q4, we combine data sets across two versions of the Federal Reserve Bank of New York's quarterly Consumer Credit Panel (CCP), a database on consumers' credit use and payment performance drawn from anonymized Equifax credit bureau records.⁶ The first version of the CCP is a newly released account- (or loan-) level version, while the second is at the individual borrower level. Both versions are needed to construct time series for retail credit, while the account-level version allows for a deeper look into the details of the sector. The resulting sample has coverage of up to 10 credit card accounts and up to 4 auto and personal loans per individual, and it contains industry code and account type indicators that allow for the categorization of the accounts.

As our definition of retail credit consists of the amounts owed to retail stores by their customers for purchases made on credit extended in partnership with a financial institution, the sector is likely primarily revolving in nature, consisting of all private-label credit cards and of nonrevolving credit holdings of sales finance companies. Accordingly, to construct our estimate, we examine three CCP account-level data sets—for credit cards, consumer finance accounts, and auto loans—in addition to the individual-level data set.⁷

The account-level data contain a rich set of variables for each loan, including industry and account type indicators, which are salient to our analysis. Specifically for credit cards, the industry indicators allow the categorization of outstanding balances or accounts by retail expenditure sector, or where the credit cards were used—for example, department stores, clothing stores, home furnishing stores, etc. For auto and personal loans, these indicators can be used to further differentiate the finance company holdings by finance company type—personal loan, sales finance, and a miscellaneous category—thus allowing the isolation of sales finance company holdings from other finance company holdings, which is key for our estimation. In addition, account type indicators allow the separation of accounts by credit type—revolving and nonrevolving.

Our estimation proceeds in four steps. First, we isolate private-label credit card holdings from the credit card account-level data using a method proposed by Rahman (2022). The method consists of classifying each credit card account by the industry code indicator associated with it and estimating the total outstanding private-label retail credit as the sum of those accounts with industry codes indicative of spending reflecting purchases at specified retailer types, such as those for automotive parts, at clothing stores, at contractors, at department stores, of home furnishings, of jewelry and cameras, at oil companies, or at sporting goods stores.⁸ The remainder of accounts—those with industry codes indicative of specific institution types or credit card companies—are considered to be part of the general-purpose credit card universe and, thus, are not included in our estimation.⁹ In column I of Table 1 we show that as of the end of 2023, private-label credit card holdings reflected in the credit card tradeline portion of the credit bureau data amount to $64.9 billion and are held in 107.3 million accounts by 68.5 million borrowers with a median age of 54 and a median Equifax Risk Score of 747. The median balance per account is $127, and the median account monthly payment is $29. More than 60 percent of balances are held by nonprime borrowers. The delinquency rate on these accounts, or the fraction of balances that are at least 30 days past due (excluding severely derogatory loans), is 6.2 percent.

Table 1: Retail Credit Characteristics as of 2023:Q4

Second, using information about the account type, we identify additional private-label credit card holdings not reflected in the credit card account-level data set that appear in the form of revolving credit accounts in the consumer finance account-level data set that covers holdings of sales finance companies, personal loan companies, and miscellaneous companies.¹⁰ In columns II, III, and IV of Table 1, we show that among finance companies, sales finance companies hold the largest outstanding balance—$32.2 billion that is held in 20.3 million accounts by 18.7 million consumers with a median age of 51 and a median Equifax Risk Score of 723. The median balance per account is $787, and the median account monthly payment is $39. A little more than half of the sales finance revolving credit balances are held by nonprime borrowers, and the delinquency rate on these accounts is low—1.6 percent. In contrast, larger shares of personal loan and miscellaneous company revolving credit balances are held by nonprime consumers—55.6 percent and 83 percent, respectively—and the delinquency rates on these accounts are also higher—3.4 percent and 7.4 percent, respectively. Moreover, we note differences in revolving credit holdings at finance companies across a number of dimensions. For example, the median personal loan company borrower age is the highest at 56, while the median miscellaneous company borrower age is the lowest at 46. Similarly, the median personal loan company borrower Equifax Risk Score is the highest at 737, while the median miscellaneous company borrower Equifax Risk Score is the lowest at 600. That said, although there are significant differences in the median balances across revolving retail credit categories—ranging from $127 for credit card tradelines, to $787 for a revolving sales finance company account—the common characteristic among these accounts is the relatively low median monthly payment—$29 or $30 for most accounts, and $39 for revolving sales finance accounts.

Third, to complete our estimate, we examine the nonrevolving credit holdings of sales finance companies reflected in the consumer finance and the auto loan-level data sets. These holdings consist of retail sales financing, installment sales contracts purchased at a discount and, in a small share, of loans categorized as auto loans in the data.¹¹ Although sales finance companies are less in the spotlight nowadays, it is important to account for their holdings, as these companies are "specialized financial institutions which extend credit through retail dealers to consumers" and "which customarily purchase at discount the credit instruments that are given to retail dealers by their (...) customers" (see Plummer and Young, 1940, pp. 33-53) and, thus, are included in our definition. In columns V and VI of Table 1, we show the installment credit balances and those sales finance company balances that are categorized as auto. Installment credit balances amount to $10.4 billion and are held in 1.5 million accounts by 1.5 million borrowers with a median age of 42 and a median Equifax Risk Score of 628. The median balance per account is $3,605, and the median account monthly payment is $179. More than 60 percent of balances are held by nonprime borrowers. The delinquency rate on these accounts is 2.6 percent. Auto loan balances represent a very small share of retail credit and amount to $2.1 billion held in 100,000 accounts by 200,000 borrowers with a median age of 40 and a median Equifax Risk Score of 592.

Summing up the account-level estimates using data from the three sources, in column VII of Table 1 we present our estimate of total outstanding retail credit. As of 2023:Q4, we find that the retail credit sector totals $130 billion and consists of nearly 150 million accounts held by nearly 85 million borrowers with a median age of 53 and a median Equifax Risk Score of 742. The median balance per account is $194, and the median account monthly payment is $29. More than 60 percent of retail credit balances are held by nonprime borrowers, defined as those with Equifax Risk Scores of 719 or less. The delinquency rate on retail credit accounts is 4.5 percent.

Additional Details

As our method of estimating total outstanding retail credit in the U.S. employs cross-sectoral data from multiple sources, once we assemble the relevant information from these sources, we can further separate retail credit holdings by credit type or sector, as shown in Figure 1. In the first bar we present total outstanding retail credit by type—thus showing that retail credit is primarily revolving in nature. In the second bar, we make the most of the information available to us across the multiple sources. We present retail credit outstanding by expenditure sector and finance company type. This level of detail allows us to show that significant shares of balances are spent at department stores (in gray) and are held by sales finance companies (in red).

Figure 1. Retail Credit Perspectives

Finally, in Figure 2, we present retail credit trends by credit risk group—prime, near prime, and subprime. Notably, retail credit has not recovered to its pre-pandemic levels for two out of three borrower risk groups.¹²

Figure 2. Retail Credit by Credit Risk Group

Conclusion

In this note, we provide an estimate of and some details for the retail credit sector in the U.S. by combining data sets across two versions of the CCP. Our data reveal that at the end of 2023, the retail credit sector amounted to $130 billion and that nearly 85 million individuals had a retail credit account. The median balance per retail credit account is $194, and the median monthly payment on such an account is about $29. Retail credit is primarily revolving in nature.

References

Lee, D. and W. van der Klaauw (2010). "An Introduction to the FRBNY Consumer Credit Panel". Staff Report no. 479, New York: Federal Reserve Bank of New York, November 10.

Plummer, W. C. and R. A. Young (1940). Sales Finance Companies and Their Credit Practices.

Rahman, M. (2022). "CCP Data Updates and Beyond," presentation at the 2022 Credit Bureau Data Users Conference.