Effects of Credit Scores on Consumer Payment Choice

Effects of Credit Scores on Consumer Payment Choice

Fumiko Hayashi and Joanna Stavins February 2012 RWP 12-03

Effects of Credit Scores on Consumer Payment Choice

Fumiko Hayashi and Joanna Stavins

February, 2012

Abstract: This paper investigates the effects of credit scores on consumer payment behavior,

especially on debit and credit card use. Anecdotally, a negative relationship between debit card use and credit score has been reported; however, it is not clear whether that relationship is related to other factors, such as education or income, or whether it is a mere correlation. We use a new consumer survey dataset to examine whether this negative relationship holds after controlling for various consumer characteristics, including demographic and financial characteristics, consumers' perceptions toward payment methods, and card reward status. The results based on a single-year survey as well as on panel data suggest that there is a significant negative relationship between debit card use and credit score even after controlling for various characteristics. We supplement the analysis with evidence from Equifax data. The results indicate that an increase in consumers' cost of debit cards--in response to regulatory changes, for example--would have an adverse effect on low-credit-score consumers (typically those with lower incomes and less education).

We then investigate what credit score implies. If credit score significantly influences consumer access to credit cards, credit limits, or the cost of credit cards, then the negative relationship likely results from supply-side constraints. If a lower credit score is associated with differences in underlying preferences, then the negative relationship is likely due to demandside effects. Preliminary evidence strongly suggests that supply-side factors play an important role in the cost of credit and in access to credit.

Keywords: credit scores, debit cards, payment behavior JEL Classifications: D12, D14, G21

We thank Carlos Arango, Ken Brevoort, Daniel Cooper, Bob Hunt, Scott Schuh, Oz Shy, and participants at the International Industrial Organization Conference, the Federal Reserve Bank of Boston seminar, and the Federal Reserve Bank of Kansas City seminar for valuable comments, and Elizabeth Antonious and Vikram Jambulapati for excellent research assistance. The views expressed herein are solely those of the authors and do not necessarily reflect the views of the Federal Reserve Bank of Boston, the Federal Reserve Bank of Kansas City, or the Federal Reserve System. Hayashi: Payments System Research Function, Economic Research Department, Federal Reserve Bank of Kansas City. Email:fumiko.hayashi@kc.. Stavins: Consumer Payments Research Center, Federal Reserve Bank of Boston. Email:joanna.stavins@bos..

1

1. Introduction

Over the last decade, debit card use has grown rapidly, and debit cards are now the most commonly used noncash method of payment in the United States. According to the 2010 Federal Reserve Payments Study (FRPS), 37.9 billion debit card transactions were made in 2009 in the United States, representing 35 percent of total noncash retail payments. Between 2000 and 2009, debit card use has grown at an average annual rate of 18 percent. In contrast, credit card use has grown at a much slower pace (3.7 percent at a compound annual rate) and has accounted for 20 percent of total noncash retail payments in 2009 (Federal Reserve System 2010).

The rapid growth of debit cards has stimulated several studies on consumer payment choice. Previous studies highlighted several important demand-side factors that influence consumer payment choice, such as consumer characteristics, transaction characteristics, payment method attributes, and price or reward of payment methods. Most of these studies did not include factors that would limit available payment methods to consumers, because very few datasets contain information necessary to examine the effects of such factors. There are a few exceptions. Rysman (2007) and Ching and Hayashi (2010) took account of merchant acceptance of payment methods when analyzing consumer payment choice. Zinman (2009) found that the closer the credit limit and the balance on credit cards, the more likely consumers are to use a debit card rather than a credit card.

The main goal of this paper is to investigate the effects of credit scores on consumer payment choice, especially on debit card use. Anecdotally, a negative relationship between debit use and credit score has been observed (Lightspeed 2009); however, it is not clear whether it is a mere correlation. We first examine whether this negative relationship holds even after controlling for various factors, such as consumer characteristics, payment method attributes, and price or rewards associated with payment methods. We use a unique consumer survey dataset: the 2008 and 2009 Survey of Consumer Payment Choice, or SCPC, a consumer survey conducted by the Federal Reserve Bank of Boston. We further investigate what credit score implies. If the credit score significantly influences consumer access to credit, credit limits, or the

2

cost of credit, then the negative relationship likely results from supply-side effects: consumers with a lower credit score cannot access credit by using their credit card, or accessing credit via credit card is too costly for them, and thus they use their debit card instead. Our dataset provides other variables that are indicative of consumers' current credit conditions, such as credit card balance and financial difficulties, which help to disentangle supply- and demandside effects.

Our results suggest that there is a negative relationship between debit card use and credit score even after controlling for various consumer characteristics, payment method attributes, and rewards on payment cards. A new rule starting on October 1, 2011, on debit card interchange fees1 has reduced the interchange fees charged on debit card transactions on the cards that are issued by large financial institutions. Some large financial institutions have reacted to this rule by announcing higher debit card fees to recover their lost interchange fee revenues.2 Because consumers with low credit scores, such as the FICO score developed by the Fair Isaac Corporation, are the ones who use debit cards more intensively, they are likely to be especially adversely affected if their banks introduce debit card fees.3 Based on our data, younger, less educated, and lower-income consumers would be more likely to be affected than other demographic groups, especially if their access to alternative means of payment is limited.

Because the SCPC provides only self-reported credit scores, we extend our analysis by using the Equifax dataset, which includes an external measure of credit score, closely correlated with a FICO score. While we cannot merge the SCPC and the Equifax data precisely, we extract the Equifax credit score for each finely decomposed socio-economic group and compare that external measure with the SCPC data.

The paper is organized as follows. Section 2 provides background and the main hypothesis of this study, and reviews related literature. Section 3 describes the data used for this study (SCPC) and compares the SCPC with other datasets in terms of credit scores. Section

1 A debit card interchange fee is paid to the card issuer by the merchant who accepts a debit card for the payment. For details, see 2 For example, Bank of America announced a $5 monthly fee to start in 2012, and Wells Fargo and Chase tested a $3 monthly fee on any customers using their debit card. The banks later retracted this policy. 3 Low-FICO-score consumers may be able to avoid debit card fees by switching to financial institutions that do not charge debit card fees. Merchants may also start offering discounts to their customers who pay with a debit card, because the merchants' costs of accepting debit cards are reduced.

3

4 focuses on the factors affecting FICO scores. Model and estimation results of adoption and use of payment cards are included in Section 5. Section 6 discusses the causes and implications of the FICO score effect, distinguishing between supply-side and demand-side effects. Section 7 concludes.

2. Background and hypothesis

2.1 Credit scores

Credit scores summarize consumers' credit history, and are used by various lenders and financial institutions to evaluate consumers' creditworthiness. A higher credit score indicates that a person is expected to have a lower probability of defaulting on his or her loan obligations. Even though most lenders--especially mortgage lenders--tend to use the same credit scores, a lot of confusion exists concerning how these scores are calculated. This is so because the formula that credit companies use to calculate credit scores is complex and proprietary, and also because a given consumer's credit score can change over time even when the consumer does not alter his or her behavior. Credit scores always correspond to assessed riskiness in a monotonic way, that is, a person with a higher credit score is always expected to have a lower probability of default than a person with a lower score, but any given value of credit score can be associated with a higher or lower probability of default over time.

The first model of credit scoring was developed by the Fair Isaac Corporation in 1956, although the models have evolved over time. By some estimates, more than a hundred different models of credit scoring have been developed. Credit scores are now calculated by the Fair Isaac Corporation, as well as by other companies, and are available through the major credit bureaus in the United States: Equifax, Experian, and TransUnion. FICO credit score ranges from 300 to 850, with 60 percent of scores falling between 650 and 799. Every person with a credit record has three credit scores for the FICO scoring model, as each of the three major credit bureaus (Experian, Equifax and TransUnion) has its own database. Data about an individual consumer can vary from bureau to bureau, even though each is designed to measure the risk of default and incorporates various factors in a person's financial history. The newer

4

models tend to have greater score dispersion and comparisons across the models should not be made.

Although the exact formulas for calculating credit scores are secret, FICO has the following components:

? Payment history (35 percent of the score): late payments on bills, such as a mortgage, credit card or automobile loan, can cause a FICO score to drop. Paying bills on time improves a FICO score.

? Credit utilization (30 percent of the score): The ratio of current revolving debt (such as credit card balances) to the total available revolving credit or credit limit. FICO score can be improved by paying off debt and lowering the credit utilization ratio, or sometimes-- but not always--by applying for and receiving the credit limit increase. Closing existing accounts will typically raise the utilization ratio and therefore lower the FICO score.

? Length of credit history (15 percent of the score): Longer credit histories are typically associated with higher FICO scores.

? Types of credit used (10 percent of the score): Having a variety of different types of credit (installment, revolving, consumer finance, mortgage) can lead to a higher FICO score.

? Recent search for credit (10 percent of the score): Opening new accounts is associated with greater credit risk, and new accounts lower credit scores. Credit inquiries, which occur when a person is seeking new credit, can hurt the consumer's FICO score. Although all credit inquiries are recorded, credit inquiries that are made either by an individual himself (to check his credit), or by his employer (for employee verification), or by companies initiating prescreened offers of credit or insurance have no impact on the credit score.

A more complete set of variables included in the FICO score is listed in the appendix. Even though the models used to calculate credit scores differ across the companies that calculate them, the steps involved are the same for all of them. They start by using the data on each consumer in their models to predict a likelihood that a person will default on his or her

5

credit obligations within the next two years. Next, they group the consumers with others who exhibited similar credit history events. Finally, each person's probability of default calculated in the first step is mapped to a credit score, based on where the person is grouped in the second step. The process yields score-probability relationships, which then allow prospective lenders to make their decisions whether or not to lend to each consumer. As mentioned above, the information is updated all the time, and so the score changes over time.

2.2 Literature

The literature on consumer payment choice has been growing since the late 1990s, but most of the previous studies focused on the effects of consumer characteristics, payment method attributes, and prices or rewards of payment methods. Consumer characteristics, such as demographic and financial characteristics, and tendencies to adopt new technologies are correlated with the adoption and use of payment methods (for example, Kennickell and Kwast 1997, Stavins 2001, and Hayashi and Klee 2003). Mantel (2000), Jonker (2005), Klee (2006), and Schuh and Stavins (2010) found that payment method attributes as perceived by consumers are strongly correlated with consumer payment choice. Prices or rewards offered on payment methods are also highly correlated with the use of payment cards (for example, Borzekowski et al. 2008, Ching and Hayashi 2010, Simon et al. 2010).

A few studies included factors that would limit available payment methods to consumers in their analyses. Rysman (2007) and Ching and Hayashi (2010) took account of merchant acceptance of payment methods. Rysman found that a consumer's favorite card network is positively correlated with the number of local merchants who accept that network's cards. In the Ching and Hayashi model, a consumer's choice set consists of payment methods that the consumer believes are accepted at a given type of store. Zinman (2009), on the other hand, analyzed how a consumer's credit limit and balance on a credit card would influence his choice of debit or credit cards. He found that the closer the credit limit is to the balance on credit cards, the more likely consumers are to use a debit card rather than a credit card. A few other studies also found a negative relationship between credit card balance and credit card use (for

6

example, Ching and Hayashi 2010 and Simon et al. 2010) or a positive relationship between credit card balance and debit card use (Lee et al. 2007, and Sprenger and Stavins 2010).

Our study is closely related to Zinman (2009). Unlike Zinman's, our dataset does not contain each consumer's credit limit. However, our dataset does include credit scores as well as other variables that are indicative of consumers' current credit situations. These variables are used to disentangle supply- and demand-side effects on consumer payment choice. When estimating the effect of credit scores on consumer payment choice, we also control for consumer characteristics, perceived payment method attributes, and prices or rewards of payment methods. Avery et al. (2010) examines a relationship between credit scores and demographics, but does not deal with payment behavior.

2.3 Do credit scores imply supply-side or demand-side effects?

A negative relationship between debit card use and credit score could imply supply-side effects, demand-side effects, or a combination of both. As explained above, credit scores are used by various lenders, including credit card issuers, to evaluate consumers' creditworthiness. Since a lower credit score indicates that a person is expected to have a higher probability of defaulting on his or her credit card loan obligations, credit card issuers may provide relatively lower credit limits to those consumers with lower credit scores than to those with higher credit scores. Or credit card issuers may not issue a credit card to a consumer whose credit score is below a certain threshold, or they may make credit cards more costly to low-score consumers, by offering them credit card plans with higher fees or higher interest rates. These are potential supply-side constraints that might force consumers to use debit cards more heavily, even if they preferred to use credit cards.

Thus, supply-side effects of consumers' credit scores on consumer payment choice include the limitations on consumers' ability to conduct credit card transactions--credit card issuers may have failed to approve their credit card applications or may have set their credit limits too low to allow consumers to make their desired number of credit card transactions, or may have set the cost higher than that offered to people with high scores. The supply-side

7

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download