Part 1: The Bureau’s Unsound Rationale for Why an $8 Late Fee Would Deter Late Payments
Developing sound regulatory policies ordinarily requires careful consideration of relative costs and benefits, supported by data. In the case of credit card penalty fees, Congress mandated such a standard for rulemaking when it passed the Credit Card Accountability Responsibility and Disclosure Act of 2009. Unfortunately, the Consumer Financial Protection Bureau’s Notice of Proposed Rulemaking on Credit Card Penalty Fees released on Feb. 1 falls well short of this standard.
Among other changes, the proposed rule aims to compress the existing “safe harbor” for credit card late fees, within which a fee is presumed to be reasonable and proportional to the customer’s violation or omission with respect to the card agreement. The proposal would reduce the safe harbor maximum to $8 from its current levels of $30 for a first violation and $41 for a subsequent violation within the next six billing cycles. The original amounts were established by rulemaking in 2010 based on empirical data and a balancing of the relative cost and benefits.
The CFPB, while authorized to establish a safe harbor for late fees, is also mandated to consider several factors in determining whether a penalty fee, including a late fee, is reasonable and proportional. One of these factors is the cost of the violation. Another is the potential incentive or deterrence role of the fee.
This note highlights major conceptual shortcomings in the Bureau’s consideration of the deterrence role of late fees and, relatedly, the Bureau’s failure to consider the risk-based pricing aspect of late fees. Follow-up notes (parts 2 and 3, respectively) will examine the inadequate empirical assessment of the deterrence role that is provided in the proposal, and the proposal’s flawed, cost-based calculation deriving the proposed $8 maximum.
In short, this note demonstrates that that the Bureau falls short of its statutory obligation to meaningfully consider the factors relevant to an assessment of late fees, in that the Bureau:
- Applies an excessively narrow definition of deterrence in rationalizing the $8 limit, centered on the immediate decision by a “rational” and “attentive” consumer on the day a payment is due whether to submit the payment.
- Ignores the risk pricing aspect of late fees — the fact that missing a payment date indicates increased likelihood of default, and banks price this increased risk via a late fee.
- Fails to give serious consideration to various potential incentive effects of late fees for consumers who are prone to inattention, mistakes or well-documented behavioral biases such as overoptimism (about ability to meet debt obligations).
- Asserts, without providing evidence or elaborating, that while the proposal may weaken deterrence, it “creates additional incentives for issuers to emphasize reminders, automatic payment, and other mechanisms that maintain similar or better payment behavior.”
1. The Bureau errs by looking at consumer behavior only at the payment due date, by assuming all consumers pay close attention to their upcoming payments, and by ignoring common behavioral biases, when assessing deterrence effects.
The Bureau acknowledges that it lacks “direct evidence on what consumers would do in response to a fee reduction similar to those contained in the proposal.” Nonetheless, the Bureau asserts that an $8 penalty for both first and subsequent late payments would provide sufficient incentive for on-time payment. The Bureau states that, in addition to “available empirical evidence” (to be examined in part 2), this assertion rests on “a comparison of the proposed $8 late payment safe harbor amount to minimum payment amounts on accounts in the Y-14 data.”[1]
This “comparison” as presented in the proposal consists of a simple calculation representing the choice made by consumers who are “rational” and “pay attention to financial penalties”, at the time a card payment is due, whether to delay the minimum required payment past the due date. The calculation equates this decision with the consumer choosing whether to “borrow” the minimum payment due, such that the “cost” of this borrowing is the late fee. Thus, annualized percentage rates (APRs) are calculated for hypothetical “borrowed” amounts and brief “borrowing” periods subject to the $8 late fee.
The CFPB then concludes that these APRs are sufficiently elevated to deter late payment. For example:
As the median minimum due was $39 for all cardholders between October 2021 and September 2022 in the Y-14 data, and around half of late payers made a payment in less than 10 days past the due date, the effective APR could be higher than 730 percent for some consumers. Thus, the Bureau has preliminarily determined that the proposed $8 late fee safe harbor amount is still a powerful deterrent to those consumers who pay attention to financial penalties.
This conceptual exercise is inadequate, however, for evaluating the overall, deterrence role of a late fee. It improperly assumes that late fees play a deterrence role solely on the payment due date. It also assumes that consumers are “attentive to their payment obligations”, and allows no role for behavioral biases whereby consumers make spending and borrowing decisions contrary to their long-run best interest or are prone to mistakes.[2]
(The proposal later mentions a few types of consumers who tend to make mistakes in their borrowing decisions, such as “consumers for whom high late fees serve as a valuable commitment device without which they would have a harder time responsibly managing their credit card debt.”[3] However, the proposal provides inadequate discussion of potential incentive effects of late fees for these consumer types. More on this below.)
Of course, within the narrow context of the hypothetical scenario to which the above calculation applies, a small late fee should suffice to motivate compliance with the payment due date. Provided the late penalty is non-trivial and the consumer has the cash on hand and no dire, alternative need for it, the non-behaviorally-biased consumer in that narrow context will pay. What’s the rationale for hanging on to the payment amount for a week or two only to pay it back with a penalty (again, absent dire circumstances)? But the claim that the incentive role of the late fee is limited to such a scenario is an overly narrow perspective.[4] Late fees serve as a motivating deterrence factor over the long term as consumers are making financial choices.
The incentive role of a late fee primarily is to motivate greater, long-run attention by consumers to their spending and finances and how these may affect their ability to meet their debt obligations. However, the CFPB appears to believe that is relevant only on the payment due date – such that just on this day, assuming the consumer is “rational” and “attentive”, the late penalty motivates the consumer to submit his or her payment.
In other words, the deterrence or incentive effects of a late fee are not appropriately measured at the time a payment comes due. Rather, they should be considered and observed over the prior month or, better yet, over the course of many prior months when the consumer is making choices about his or her spending and saving. In general, by the time the payment is due, a consumer is not faced with a choice whether to pay; either the consumer is able to pay, or not.
In sum, the Bureau’s argument is overly simplistic and tangential and does not appropriately consider when the deterrence effect of a late fee would be observed – during the month or months leading up to the payment due date. In doing so, the Bureau avoids substantive discussion of the potential adverse consequences for deterrence of reducing the safe harbor maximum.
2. The Bureau errs by failing to account for the risk pricing aspect of a late fee.
One essential role of a late fee is to price delinquency risk, which increases with a late payment. A rational consumer would factor this price into the overall APR of the card to inform his or her borrowing decisions – that is, in deciding to open a credit card account and deciding whether to accumulate a revolving balance. For purposes of the statute, the risk of nonpayment constitutes a “cost associated with the violation”.
Consumers, when selecting and then using the card, would weigh the possibility that they might have to delay a payment beyond the due date and thereby incur the late fee, knowing that this risk increases as the balance and minimum required payment increases. The higher the late fee, the more strongly the consumer would be encouraged to mitigate the risk of late payment by using their card more cautiously.[5] In other words, the higher APR may motivate the consumer to reduce the balance and the associated likelihood of paying late. Moreover, with fee income potentially increased and credit risk thus reduced, the bank may be able to offer a lower baseline APR.
The CFPB’s proposal ignores this justification for a late fee entirely.
3. The Bureau errs by downplaying how late fees affect consumers who are prone to inattention, mistakes or well-documented behavioral biases.
The case of consumers who make mistakes. The proposal contains a few references to consumers who make mistakes. These are:[6]
- Consumers who ignore or downplay late fees because they do not expect to be late, although many may end up paying late.
- Consumers who pay more attention to late fees than to other consequences of paying late, who thus “might be harmed in the short run if a reduction in late fees makes it more likely that they mistakenly miss payments.”
- “Consumers for whom high late fees serve as a valuable commitment device without which they would have a harder time responsibly managing their credit card debt.”
The Bureau makes little effort to assess the potential importance of incentive effects of late fees for such consumers. Nor does the Bureau provide any empirical evidence on the fraction of the population with credit cards that are not prone to behavioral biases and the fractions associated with each of the above three categories.
Once bitten, twice shy. The proposal describes the first category of consumers above as those who “may mistakenly expect high fees to be unimportant to them, as they are overly optimistic about not missing a payment.” The Bureau asserts that “such consumers would benefit from the proposed changes to late fee amounts, which lower the cost of this mistake,” thus dismissing any potential incentive role of late fees for such consumers.
The Bureau appears to be saying that someone who doesn’t see any likelihood of missing a payment, as mistaken as that notion might be, would have no reason to behave any differently in the presence of a late fee, large or small. But on further reflection, it seems overly simplistic to conclude that late fees have no role in relation to this type of consumer.
While a consumer might discount late fees prior to incurring a first one, the consumer would be highly likely to consider late fees from that point forward. To the extent that the late fee changed consumer behavior for the life of the lending relationship, it would substantially reduce risk to the lender.
Indeed, there is solid empirical evidence of a learning effect associated with incurring a late fee. For example, a National Bureau of Economic Research study (Agarwal et. al 2008), using a proprietary dataset from a large U.S. bank, found that incidence of late fee payments declines sharply as a new credit card account ages.[7] The analysis attributes this decline to consumers’ learning to pay on time after experiencing a late payment penalty.[8] Thus, even inexperienced consumers are likely to be incentivized by late fees, albeit only after experiencing an initial late payment event.
Also, if consumers were overly optimistic about their ability to make required payments on time, then it seems likely that they would be prone to other mistakes in managing their debt, such as a tendency to overborrow. To the extent that consumers do not consider the full cost of their borrowing, they are apt to become overextended.[9] Thus, following their first late payment event, such consumers may not only become more attentive to paying on time, but also are likely to make better financial decisions in a variety of ways, such as reducing the number of credit cards they hold, for example.
By considering only the “reduced cost of mistake” aspect of a reduced late fee while ignoring these other relevant factors, the Bureau fails to meaningfully consider the deterrence role of late fees.
Late fee saliency. The second type of behavioral bias-prone consumer mentioned in the proposal is consumers who “pay more attention to late fees than to other consequences of paying late.” In other words, for these consumers the late fee is more important than other factors that another borrower would consider. The latter factors could include “interest charges, penalty rates, credit reporting, and the loss of a grace period.”[10]
Again, the Bureau offers an assessment, which boils down to the following assertions:
- These consumers “might be harmed in the short run if a reduction in late fees makes it more likely that they mistakenly miss payments”.
- “The Bureau has not quantified this effect.”
- The Bureau “notes that reducing late fees may increase issuer incentives to find other approaches to make the consequences of late payment salient to consumers, including reminders or warnings.”
The first statement subtly downplays the role of late fees by referring to “mistakenly missed payments,” suggesting that the role of late fees is limited to countering inattention or forgetfulness. But, as noted previously, late fees can also have an important role in countering consumers’ behavioral biases, such as by discouraging over-spending that increases the likelihood of missing a payment due date. Moreover, to the extent that they are more salient than other consequences, late payment penalties will be more effective in this way. Thus, the Bureau sidesteps a full discussion of incentive effects.
Furthermore, the assertion that a lower safe harbor maximum “may increase issuer incentives to find other approaches to make the consequences of late payment salient to consumers, including reminders or warnings” comes without evidence or explanation.
It is standard practice for many banks to send reminders and warnings to loan and credit card customers who have missed a payment. Typically, banks also send out reminders of upcoming due dates to credit card customers who have been regularly paying on schedule. Thus, it is not clear why the Bureau believes that banks will increase their use of reminders if the safe harbor maximum is lowered – many already use these tools.[11]
Moreover, the Bureau provides no evidence to suggest that warnings and reminders are as effective as late fees in affecting consumer behavior. Relatively few studies have explored the effectiveness of payment nudges on likelihood of late payment on non-delinquent credit cards or other consumer loans, and these are limited in context and scope.[12] The studies suggest that reminders or warnings can reduce the frequency of late payments or have other positive effects in specific contexts with certain consumers.[13] However, at least one study found that reminders may have adverse side effects, in the form of increased overdraft activity among a segment of customers.[14] All of these studies hold in place existing fee structures; little is known about the effectiveness of substituting reminders and warnings for reduced late fees.
Significantly, another study compared the effectiveness of reminders with and without mention of a penalty fee for missed payment and found that the reminders that referred to the penalty were far more effective — further corroborating the important deterrence effect of appropriately sufficient late fees.[15]
Sticking to a plan. Another type of consumer prone to behavioral biases mentioned in the proposal is one for whom a sufficiently high late fee provides an incentive to stick to a monthly spending and debt management plan. The Bureau notes that without the incentive provided by the late payment penalty, these consumers “would have a harder time responsibly managing their credit card debt.”
The notion that there are “present-biased” consumers who may fail to stick to self-set debt paydown plans has a strong foundation in the behavioral consumer finance literature.[16] Thus, the Bureau is acknowledging an important potential incentive effect of card late fees.
But again, the Bureau sidesteps meaningful discussion of the issue, instead asserting:
“To the extent that late fees benefit some consumers in this way, any harm to such consumers may be mitigated to the extent that the proposal creates additional incentives for issuers to emphasize reminders, automatic payment, and other mechanisms that maintain similar or better payment behavior, as discussed below.”
No evidence is provided in support of these assertions. As noted above, reminders are no substitute for late payment fees. Automated payment plans have unintended, costly side effects not acknowledged in the proposal. For instance, automated payment transactions are the most frequent type of overdrawn account transaction according to a CFPB report, and the Bureau’s educational material on automatic debt payments cautions consumers about this risk.[17]
4. Conclusion
The Bureau acknowledges that “late fees are a cost to consumers of paying late, and a lower late fee amount for the first or subsequent late payments might cause more consumers to pay late” and that the Bureau “does not have direct evidence on what consumers would do in response to a fee reduction similar to those contained in the proposal”. Despite these acknowledgments, the proposal proposes an $8 maximum safe harbor penalty for the first and subsequent late payments, arguing that this would provide enough incentive for cardholders to pay on time.
There are multiple conceptual flaws in the Bureau’s assertions regarding late fees. First, the Bureau relies heavily on a problematic and superficial argument that an $8 late fee suffices to deter late payment by a “rational” and “attentive” consumer. Second, the Bureau ignores the risk pricing role of late fees. Third, the Bureau fails to offer serious consideration of various potential incentive effects of late fees for consumers who are prone to inattention, mistakes or well-documented behavioral biases. Finally, the Bureau asserts, without adequate conceptual or empirical support, that “the proposal creates additional incentives for issuers to emphasize reminders, automatic payment, and other mechanisms that maintain similar or better payment behavior” compared to higher late fees.
References
Paul Adams, Benedict Guttman-Kenney, Lucy Hayes, Stefan Hunt, David Laibson, and Neil Stewart (2022). “Do Nudges Reduce Borrowing and Consumer Confusion in the Credit Card Market?”, Economica, volume 89, issue S1, pages S178-S199. Available at: Economica – Wiley Online Library
Sumit Agarwal, John C. Driscoll, Xavier Gabaix, and David Laibson (2008). “Learning in the Credit Card Market.” NBER Working Paper No. w13822. Available at: https://ssrn.com/abstract=1093660
Anat Bracha and Stephan Meier (2015). “Nudging Credit Scores in the Field: The Effect of Text Reminders on Creditworthiness in the United States,” Research Department Working Paper no. 15-2, Federal Reserve Bank of Boston, available at: bostonfed.org
Daniel Campbell, Andrew Grant, and Susan Thorp (2022). “Reducing Credit Card Delinquency Using Repayment Reminders”, Journal of Banking and Finance, volume 142 (September). Available at: ScienceDirect
John Gathergood, Hiroaki Sakaguchi, Neil Stewart, and Joerg Weber (2019). “How Do Consumers Avoid Penalty Fees? Evidence From Credit Cards.” Available at: https://ssrn.com/abstract=2960004
Theresa Kuchler and Michaela Pagel (2021). “Sticking to Your Plan: The Role of Present Bias for Credit Card Paydown,” Journal of Financial Economics, Volume 139, Issue 2, pages 359-388. Available at: https://doi.org/10.1016/j.jfineco.2020.08.002
Paolina C. Medina (2011). “Side Effects of Nudging: Evidence from a Randomized Intervention in the Credit Card Market”, The Review of Financial Studies, Volume 34, Issue 5, Pages 2580–2607. Available at: oup.com
Stephen P. Roll and Stephanie Moulton (2019). “The Impact of Automated Reminders on Credit Outcomes: Results from an Experimental Pilot Program”, Journal of Consumer Finance volume 53, issue 4, pages 1693-1724, available at: Journal of Consumer Affairs – Wiley Online Library
Stephen Roll, Yung Chun, Olga Kondratjeva, and Sam Bufe, (2022). Prevention, Resolution, or Counseling: Testing Three Approaches to Preventing Debt Payment Delinquencies in a Large-Scale Field Experiment. Available at SSRN: https://ssrn.com/abstract=4282388
H. Sakaguchi, N. Stewart, J. Gathergood, P. Adams, B. Guttman-Kenney, L. Hayes, and S. Hunt (2022). Default Effects of Credit Card Minimum Payments. Journal of Marketing Research, 59(4), 775–796. Available at: https://doi.org/10.1177/00222437211070589.
Daniel Schwartz (2021). “The Rise of a Nudge: Field Experiment and Machine Learning on Minimum and Full Credit Card Payments”, in NA – Advances in Consumer Research, Volume 49, eds. Tonya Williams Bradford, Anat Keinan, and Matthew Thomson, Duluth, MN : Association for Consumer Research, Pages: 362-363. Available at: acr_vol49_3000661.pdf (acrwebsite.org)
Jonathan Slowik (2012). “Credit Card Act II: Expanding Credit Card Reform by Targeting Behavioral Biases,” UCLA Law Review volume 59, pages 1292-1341. Available at: heinonline.org.
Jialan Wang (2022) “To Pay or Autopay? Fintech Innovation and Credit Card Payments”. Available at: Wang 2022 – To Pay or Autopay.pdf (dropbox.com)
[1] See pages 53-54 of the NPR.
[2] See Slowik (2012) for a discussion of behavioral biases in the credit card context.
[3] See page 108 of the NPR.
[4] Furthermore, a missed payment is not a loan—it is, plain and simple, a missed payment. Any decision made on the due date is not one related to the deterrence effect of the late fee or a hypothetical APR, but rather, the perceived benefit to the consumer of holding onto the money for some additional days or weeks; that is, the “opportunity cost” of not retaining the minimum payment for a short period.
[5] For example, suppose a consumer has accumulated an average balance of $3,000 on a card with a baseline APR (assuming no late payments) of 15 percent and late payment penalty of $8, and expects with this sized balance to miss one payment due date over the next year. Then the consumer’s total expected interest and fee payment during the year, inclusive of the late fee, will be $458, which translates into an APR of approximately 15.3 percent. By comparison, if the late fee instead was $40, the effective APR would increase to 16.3 percent.
[6] See especially pages 107-108 of the NPR.
[7] Another, more recent study using data from the U.K. (Gathergood et. al 2019) similarly finds that incidence of late payment fees declines sharply over time for new accounts. The analysis suggests that this effect is tied to consumers adopting automatic payments as a preventive measure after experiencing a late payment event.
[8] The analysis also indicates some backsliding as memory of the late payment event recedes. On net, however, “knowledge accumulation dominates knowledge depreciation” so that over time, fee payments decline dramatically.
[9] As discussed in a previous BPI research note, evidence shows that behavioral biases, including over-confidence, can adversely affect consumers’ use of credit card debt.
[10] By listing “interest charges” and “loss of a grace period” together with “penalty rates” and “credit reporting”, the NPR is conflating consequences for an individual who general pays in full each month and misses a due date with consequences for an individual who has a revolving balance. The former, by having missed the due date, would now have a revolving balance with interest charges and loss of a grace period. Those who already are revolvers may see their interest rate increased or their account being reported as delinquent, but typically only if the late payment episode extends beyond 30 days.
[11] The NPR suggests that banks “profit from late payment fees” and therefore have insufficient incentive to discourage customers from paying late. But that reasoning fails to consider banks’ primary motivation for issuing warnings and reminders: to reduce costly incidence of delinquency and default.
[12] Most studies examining payment nudges focus on nudges targeted at increasing repayment by delinquent borrowers, which are used extensively and often; see, for example, Campbell, Grant, and Thorp (2022). Other studies focus on motivating consumers to increase their monthly payments above the required-minimum amount; see, for example, Adams et. al (2022).
[13] For instance, Bracha and Meier (2015) and Roll and Moulton (2019) focus on the effects of reminders for samples drawn from special credit counseling programs. Some individuals within each sample were randomly selected to receive monthly reminders linked to their financial goals and payment obligations, and outcomes for the treatment group (those who received reminders) were compared to those for the control group (those who did not). The reminders and outcomes concerned general debt obligations, not specifically credit card. Bracha and Meier (2015) found a positive effect of reminders on credit scores for subprime consumers, apparently associated with a reduction in debt, but not for others. Roll and Moulton (2019) found that “consumers offered reminders were 21% less likely than the control to experience a 60 day or longer payment delinquency and 12% less likely to experience a 30+ delinquency”. Roll et. al (2022) report on the outcomes of a randomized field experiment conducted in a large Midwestern credit union, testing various types of nudges intended to improve payment outcomes of credit union members on non-mortgage installment loans. The experiment was conducted mostly in 2021, and thus overlapped with the pandemic period. A series of monthly reminder emails containing a basic reminder of the due date, an offer of financial assistance resources if needed, and a link to set up automatic payments (for those that did not already have autopay set up), and additional information intended to encourage timely payment was found to be effective in reducing delinquency incidence.
[14] See Medina (2021).
[15] See Schwartz (2021).
[16] See, for example, Kuchler and Pagel (2021).
[17] Also, research suggests that automatic payment can have adverse side effects for some consumers in the form of slower paydown of existing balances and higher cumulative interest costs. See Sakaguchi et al. (2022) and Wang (2022).
