The report was witten by Julapa Jagtiani, and from
We investigate whether the Credit Card Accountability, Responsibility, and Disclosure (CARD) Act of 2009 influenced the debt structure of consumers. By debt structure, we mean the proportion of total available credit from credit cards for each consumer.The act enhances disclosures of contractual and related information and restricts card issuers' ability to raise interest rates or charge late or over-limit fees, primarily affecting non-prime borrowers. Using the credit history via the
* * *
A profound concern over decades about the credit card market has been weak consumer protection (
It is a priori unclear whether the proportion of available credit line from consumer cards increases or decreases after the implementation of the act. On the one hand, research shows that restrictions on issuers' risk pricing reduce their willingness to lend (Han, Keys, and Li, 2018; Elliehausen and Hannon, 2018), which can translate into a lower proportion of available credit on consumer cards ("the supply effect"). On the other hand, the act induces better disclosure and consumer protection, which makes credit card products more appealing to consumers. As a result, they may increase the demand for credit through credit cards relative to other forms of borrowing. Consequently, consumers should borrow more through credit cards, relative to other financial products ("the demand effect"). Thus, the net effect of the act on the proportion of available credit from credit cards is ultimately an empirical question.
We use a novel data set that traces the quarterly credit history for nearly all the
Following the literature (Han et al., 2018; Elliehausen and Hannon, 2018), we use a difference-in-differences research design that compares changes in the proportion of available credit from consumer cards over time for non-prime borrowers, who were primarily affected by the CARD Act, to changes in the proportion of available credit from consumer cards for prime borrowers, who were little affected. As the bill was implemented in three stages, a recent study by Agarwal et al. (2015) examines the consequences in three periods relative to their pre-period (March 2008-April 2009): the passage of the bill and Phase 1 (May 2009-February 2010), Phase 2 (March 2010-August 2010), and Phase 3 (September 2010-December 2011). This approach, however, is criticized on the ground that it did not capture the anticipation effect when the bill was first introduced in
In regression analysis, we control for an array of variables that could potentially influence available credit from credit cards. They include the card utilization ratio, the number of credit inquiries, the presence of a card with more than 60 days past due, the presence of a non-credit card loan with more than 60 days past due, and the state of the local economy as measured by the state-level initial unemployment insurance claims. We also include consumer, year-quarter, and state-fixed effects in the regressions, and two-way cluster the standard errors at the consumer and subperiod levels.
Our baseline analysis suggests that the proportion of available credit from credit cards started to decline in the anticipation period for non-prime borrowers relative to prime borrowers.
The decline persisted through all of the three phases and the post-Phase 3 period. The reduction is economically meaningful and ranges from 4% to 11% of the mean level of total credit limits on consumer cards divided by total available credit. The results suggest that the supply effect outweighs the demand effect, yielding a net reduction in the proportion of available credit from credit cards. We consider two potential alternative explanations. First, the freezing of securitization markets in the financial crisis may reduce the issuance of credit cards because it is more difficult to securitize them. However, according to the
We then extend our analysis in a number of directions. First, a key assumption underlying our identification strategy is that non-prime and prime borrowers would have exhibited parallel trends in the proportion of available credit from credit cards in the absence of the CARD Act. We assess the validity of this parallel-trends assumption by examining six quarters preceding the anticipation period. We find no decline in those quarters. Second, to rule out the possibility that changes in secured borrowings drive our results, we exclude secured debt (i.e., mortgages, home equity lines of credit, and auto loans) from the denominator. Our results are robust to using this alternative dependent variable (total credit limits on consumer cards to total available unsecured credit). Third, we replace the non-prime indicator with Equifax Risk Score ranks based on six groups and continue to find robust results. Fourth, our results are robust to dropping the transition quarter of each subperiod. Fifth, we add state-year-quarter fixed effects to account for state-level legislative and economic changes (e.g., changes in real estate prices across states). These fixed effects permit a comparison of consumers within the same state-year-quarter. Our inferences are unaltered. Finally, in the baseline analysis, we use credit scores at the beginning of our sample period to identify non-prime consumers. Our results are resilient to using updated credit scores as of each quarter-end.
This study adds to the debate on the costs and benefits of the CARD Act (Bar-Gill and Bubb, 2012; Jambulapati and Stavins, 2014; Agarwal et al., 2015; Debbaut,
Elliehausen and Hannon (2018) find a decrease in the number of bank credit card accounts held by non-prime borrowers relative to prime borrowers. The implications of their findings to consumer debt structure, however, are unclear for two reasons. First, the number of accounts cannot fully capture the dollar amount of available credit via credit cards. Second, consumers may also experience a contraction in total available credit (the denominator of our debt structure variable) so that the debt structure remains unchanged. Our results of a reduction in the proportion of available credit from credit cards complement their finding and enhance our understanding of how the CARD Act influenced the debt structure of households.
Santucci (2015) conducts a univariate comparison between two vintages of credit card accounts, those opened in 2005 and 2011. He finds that the latter exhibit lower credit limits than the former. His comparison, however, does not trace the evolution of the CARD Act, examines only new accounts, and does not account for the contraction in the total available credit or other secular trends. As such, he acknowledges that he cannot separate the effects of the CARD Act from those of many omitted events. In contrast, our tight research design allows us to better attribute our findings to the CARD Act.
The rest of the paper proceeds as follows. In Section 2, we discuss the CARD Act. We describe data and sample construction and lay out our methodology in Sections 3 and 4, respectively. Empirical results and a number of robustness tests are reported in Section 5. We conclude in Section 6.
* * *
We investigate whether the Credit Card Accountability and Disclosure (CARD) Act of 2009 influenced the debt structure of consumers, or the proportion of available credit from credit cards.
The CARD Act enhances disclosures of contractual and related information and restricts card issuers' ability to raise interest rates or charge late or over-limit fees, and primarily affects non-prime borrowers. Using the credit history via the
* * *
An Act to Amend the Truth in Lending Act to Establish Fair and Transparent Practices Relating to the Extension of Credit under an Open End Consumer Credit Plan, and for Other Purposes. Public Law 111-124.
Angrist, J., and
Bar-Gill, O., and
Debbaut, P., A.
Durkin, T. A.,
Elliehausen, G., and
Jambulapati, V., and
Lee, D., and W. van der Klaauw. 2010. "An Introduction to the
Levitin, A. 2011. "Rate-Jacking: Risk-Based & Opportunistic Pricing in Credit Cards." Utah Law Review 2: 339-367
Mian, A., and A. Sufi. 2009. "The Consequences of Mortgage Credit Expansion: Evidence from the
Nelson, S. 2020. "Private Information and Price Regulation in the US Credit Card Market.
Pinheiro, T., and
Sandler, R., and
Santucci, L. 2015. "A Tale of Two Vintages: Credit Limit Management Before and After the CARD Act and Great Recession."
Santucci, L. 2016. "What Happened to the Revolving Credit Card Balances of 2009?"
Tian, X., and
Zywicki, T. 2016. "No, the Credit Card Act Is Not a Free Lunch." The
* * *
2 The act requires monthly credit card statements to display prominently the interest savings from paying off balances in 36 months rather than making minimum payments. The act restricts interest rate increases on new transactions within the first year of opening the account and on existing balances except when the prior rate was an introductory rate or the minimum payment has not been received for 60 days. The act also prohibits a card issuer from imposing fees on consumers who make a transaction over an account's credit limit unless the cardholder explicitly opts in for the issuer to charge such a fee. Furthermore, an over-limit fee could be charged only once when the limit is exceeded, and over-limit fees are capped at the actual over-limit amount. A card issuer cannot impose a late fee of more than
3 See H.R. 5244 -- Credit Cardholders' Bill of Rights Act of 2008.
* * *