Liberty Street Economics

« The Tri-Party Repo Market Like You Have Never Seen It Before | Main | Beyond the Macroeconomy »

October 19, 2015

Reframing the Debate about Payday Lending


Except for the ten to twelve million people who use them every year, just about everybody hates payday loans. Their detractors include many law professors, consumer advocates, members of the clergy, journalists, policymakers, and even the President! But is all the enmity justified? We show that many elements of the payday lending critique—their “unconscionable” and “spiraling” fees and their “targeting” of minorities—don’t hold up under scrutiny and the weight of evidence. After dispensing with those wrong reasons to object to payday lenders, we focus on a possible right reason: the tendency for some borrowers to roll over loans repeatedly. The key question here is whether the borrowers prone to rollovers are systematically overoptimistic about how quickly they will repay their loan. After reviewing the limited and mixed evidence on that point, we conclude that more research on the causes and consequences of rollovers should come before any wholesale reforms of payday credit.

Payday Loan Prices: High but Justified?
The first complaint against payday lenders is their high prices: the typical brick-and-mortar payday lender charges $15 per $100 borrowed per two weeks, implying an annual interest rate of 391 percent! That’s expensive, to be sure, but is it unfair? For economists, the answer depends on whether payday credit markets are competitive: with healthy price competition, fees will be driven down to the point where they just cover costs, including loan losses and overhead.

Judging by their sheer numbers, payday lending is very competitive. Critics often fret that payday lenders outnumber Starbucks as if they—payday lenders, not Starbucks—were a plague upon the land. But shouldn’t competition among all those payday lenders drive down prices? They seem to. This study estimated that each additional payday firm per 1,000 residents in a given Zip code was associated with a $4 decline in fees (compared with a mean finance charge of about $55). In the later years of the study, the authors found that prices tended to gravitate upward toward price caps, but that seems like a problem with price caps, not competition. And of course, payday lenders also have to compete against other small dollar lenders, including overdraft credit providers (credit unions and banks) and pawnshops.

Competition seems to limit payday lenders’ profits as well as their prices. This study and this study found that risk-adjusted returns at publicly traded payday loan companies were comparable to other financial firms. An FDIC study using payday store-level data concluded “that fixed operating costs and loan loss rates do justify a large part of the high APRs charged.”

Is a 36 Percent Interest Cap in Order?
Even though payday loan fees seem competitive, many reformers have advocated price caps. The Center for Responsible Lending (CRL), a nonprofit created by a credit union and a staunch foe of payday lending, has recommended capping annual rates at 36 percent “to spring the (debt) trap.” The CRL is technically correct, but only because a 36 percent cap eliminates payday loans altogether. If payday lenders earn normal profits when they charge $15 per $100 per two weeks, as the evidence suggests, they must surely lose money at $1.38 per $100 (equivalent to a 36 percent APR.) In fact, Pew Charitable Trusts (p. 20) notes that storefront payday lenders “are not found” in states with a 36 percent cap, and researchers treat a 36 percent cap as an outright ban. In view of this, “36 percenters” may want to reconsider their position, unless of course their goal is to eliminate payday loans altogether.

“Spiraling” Fees?
A central element of the debt trap critique against payday loans is their “spiraling” fees: “When borrowers don’t have the cash come payday, the loan gets flipped into a new loan, piling on more fees into a spiral of debt for the borrower.” It’s certainly true that payday loan fees add up if the borrower extends the loan (like any debt), but do they spiral? Suppose Jane borrows $300 for two weeks from a payday lender for a fee of $45. If she decides to roll over the loan come payday, she is supposed to pay the $45 fee, and then will owe $345 (the principal plus the fee on the second loan) at the end of the month. If she pays the loan then, she will have paid $90 in fees for a sequence of two $300 payday loans. Payday lenders do not charge refinancing/rollover fees, as with mortgages, and the interest doesn’t compound (unless of course she takes out a new loan to pay interest on the first loan). Perhaps it is just semantics, but “spiraling” suggests exponential growth, whereas fees for the typical $300 loan add up linearly over time: total fees = $45 + number of rollovers x $45.

Do Payday Lenders Target Minorities?
It’s well documented that payday lenders tend to locate in lower income, minority communities, but are lenders locating in these areas because of their racial composition or because of their financial characteristics? The evidence suggests the latter. Using Zip code-level data, this study found that racial composition of a Zip code area had little influence on payday lender locations, given financial and demographic conditions. Similarly, using individual-level data, this blog post showed that blacks and Hispanics were no more likely to use payday loans than whites who were experiencing the same financial problems (such as having missed a loan payment or having been rejected for credit elsewhere). The fact is that only people who are having financial problems and can’t borrow from mainstream lenders demand payday credit, so payday lenders locate where such people live or work.

Do Economists Agree about the Perils of Payday Lending?
On the contrary, the roughly half-dozen studies published in academic, peer-reviewed journals are thoroughly mixed on “the big question” of whether payday loans help or hurt their users. On the harm side, researchers have found that access to payday loans leads to more difficulty paying bills, more involuntary bank account closures (due to overdrafts), and reduced preparedness by “airmen.” On the help side, researchers found that access is associated with reduced foreclosures after natural disasters, fewer bounced checks, and less difficulty paying bills. This study and this study find that access to payday credit does not affect users’ credit scores one way or the other. That’s a notable nonresult because if payday loans caused further financial problems, as critics allege, those problems would presumably show up as a falling credit score as borrowers began missing other debt payments—yet it doesn’t.

It’s All about the Rollovers
So if payday loan fees are competitive and don’t spiral, and if lenders don’t target minorities, and if the academic research on the pros and cons of payday credit is so mixed, what’s left in the critique against payday lenders? Rollovers. Payday lenders often pitch their two-week loans as the solution to short-term financial problems, and, true to form, about half of initial loans (those not taken out within fourteen days of a prior loan) are repaid within a month. Potentially more troubling is the twenty percent of new payday loans that are rolled over six times (three months) so the borrower winds up paying more in fees than the original principal.

Critics see these chronic rollovers as proving the need for reform, and in the end it may. A crucial first question, however, is whether the 20 percent of borrowers who roll over repeatedly are being fooled, either by lenders or by themselves, about how quickly they will repay their loan. Behavioral economists have amassed considerable evidence that, contrary to tenets of classical economists, not all people always act in their own best interest; they can make systematic mistakes (“cognitive errors”) that lower their own welfare. If chronic rollovers reflect behavioral problems, capping rollovers would benefit borrowers prone to such problems.

Unfortunately, researchers have only begun to investigate the cause of rollovers, and the evidence thus far is mixed. This study found that counseling prospective borrowers about how the cost of rollovers add up reduced their demand by 11 percent over the subsequent four months. Their finding suggests “cognitive bias” among some customers and implies that capping rollovers might benefit such borrowers (although the authors themselves did not advocate limiting rollovers). By contrast, this more recent study found that the majority of borrowers (61 percent) accurately predicted within two weeks when they would be debt-free. Importantly, the study reported that borrowers who erred were not systematically overoptimistic; underestimates of borrowing terms roughly balanced overestimates. After reviewing the available evidence, one expert in behavioral economics concluded that the link between overoptimism and overborrowing (that is, rollovers) “. . . is tenuous at best, and arguably non-existent.”

Reform or More Research?
Given the mixed evidence on the “big question” and the smaller, but crucial question of whether rollovers reflect overoptimism, more research should precede wholesale reforms. A handful of states already limit rollovers, so they constitute a useful laboratory: how have borrowers fared there compared with their counterparts in “unreformed” states? A delicate welfare calculus should also precede reform: while rollover caps might benefit the minority of borrowers prone to behavioral problems, what will it cost the majority of “classical” borrowers who fully expected to rollover their loans but can’t because of a cap? Without answering that question, we can’t be sure that reform will do more good than harm.

The views expressed in this post are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.

Robert DeYoung is the Capitol Federal Distinguished Professor in Financial Institutions and Markets at the University of Kansas School of Business. He published a study (mentioned in the post’s pricing section above) on payday lending regulation and competition in Colorado in 2013. He testified on payday lending legislation to the Missouri House of Representatives in 2011 and wrote an op-ed article on federal payday loan legislation for the Wall Street Journal in 2009.

Ronald J. Mann is the Albert E. Cinelli Enterprise Professor of Law at Columbia University. Over the course of his career, he has served as a consulting expert and lawyer on behalf of consumers, governments, and financial institutions regarding matters relevant to the payday lending industry and consumer finance industries more generally. He has never testified at a state or federal governmental hearing about an issue related to payday lending. He received no payment from the data provider, any payday lender, or any other external source for work on his paper mentioned in the post’s rollovers section above.

Morgan_donaldDonald P. Morgan is an assistant vice president in the Federal Reserve Bank of New York’s Research and Statistics Group. He has published two coauthored papers and a blog post about payday lending, which are mentioned above in the pricing, targeting, and “Do Economists Agree…” sections. He presented some of his findings on payday lending to the Virginia State Senate Committee on Commerce and Labor at its request in 2008.

Michael R. Strain is the Deputy Director of Economic Policy Studies and a resident scholar at the American Enterprise Institute. He was formerly an assistant economist in the Federal Reserve Bank of New York’s Research and Statistics Group. He and Morgan coauthored a paper on payday lending (mentioned above in the “Do Economists Agree…” section) while he was pursuing his Ph.D. at Cornell University.


Posted by Blog Author at 07:02:00 AM in Household Finance, Inflation

Feed You can follow this conversation by subscribing to the comment feed for this post.

Thanks for comment. The substitution question you raise is important; if sharp restrictions or bans force payday borrowers to substitute toward alternative, possibly costlier types of credit, the restrictions might be counterproductive. Payday borrowers tend to be maxed out on their credit cards and there doesn’t seem to be much substitution along that margin after payday loan bans. You are right about overdraft credit being a close substitute. My paper (with Strain and Seblani) finds evidence that households bounce more checks after bans and that fee income at banks increases (suggesting increased overdraft demand). That is notable because overdraft credit can be more expensive than a payday loan. In surveys, payday loan borrowers often mention using payday loans to avoid overdrafts.
This working paper by Bhutta et al. finds no substitution toward credit cards but they do find more pawn loan usage and evidence of more overdraft credit usage after bans. From their intro:
We find that payday loan bans do not reduce the number of individuals who take out alternative financial services (AFS) loans. Although payday loans decrease following the bans, that reduction is offset by an increase in the amount of borrowing from pawn shops. We also document that payday loan bans are associated with an increase in involuntary closures of consumers' checking accounts, a pattern that suggests consumers may substitute from payday loans to other forms of high interest credit such as bank overdrafts and bounced checks.

The Pew Trust paper finds that in states where payday loans are heavily restricted and there is as a result much less payday loan usage, that 'borrowers are not driven to seek payday loans online or from other sources'. Is there any more detail on this point? It would be interesting to know whether there is greater use of credit cards in such states, or greater incidence of overdrafts (the 'payday loan' of choice that I see among many of my working poor clients in a state where there are no payday lenders).

Giving people who cannot afford it these types of loans is predation. Trying to justify doing it with “but they want it and/or need it” is no different than a pusher justifying his sale of crack the same way. The same goes for the absurd claim that we need 5000% interest rates because of the high default rates. If your institution cannot make money with a 35% rate then you need to stop making those loans. You do not help people by pushing them down a slide towards inevitable default – that is predation, not help.

To actually help these people you need to come up with new products that are secured by requiring direct deposit of their paychecks into a locked account, that first makes payments on the short term loan, then pay essential bills and finally pay out what remains in cash. That would be actual help to a financially distressed and/or illiterate person.

If we're going to pass regulations on these loans, they have to be based on actual best practices, and not some pie-in-the-sky aspiration to provide cheap credit for risky borrowers. For every state or locality that has put regulations in place, people are still using these loans, and little is changed. You can't legislate a better product, lenders actually have to be able to make loans sustainably.

We want to thank everyone for their comments. We anticipated this post would generate a lot of discussion and it has indeed.

Googling “payday lenders target minorities” generates many hits, suggesting it is far from a straw man. In fact, it is a common element of the narrative against payday lending and that is why we wanted to show evidence to the contrary. Please see our reply to Ryan on whether they target low income households.

1. When we referred to the “majority” of borrowers in the last paragraph, we were referring to the 61 percent of borrowers that correctly predicted their borrowing term, not the 50 percent who repaid after one loan.

2. On rollovers, the share of borrowers that correctly predict their likely schedule of repayment is critical. One of the central elements of the critique against payday lending is that they harm unsuspecting borrowers who systematically underestimate how much they will wind up paying. As we said right at the outset, this potential behavioral problem is a completely legitimate cause for concern and a possible reason for policy intervention. Unfortunately, the evidence on whether rollovers reflect behavioral problems is limited and mixed so more research is called for. Once we have a better estimate on that number we can do the sort of welfare calculus you have in mind (and that we mentioned in our post).

The facts are correct in the “Spiraling Fees?” section. If the borrower pays the fee each period out of pocket and just rolls over the principal (as we indicated we were assuming), the fees add up linearly with rollovers -they don’t “spiral” or compound. You analyzed the case where the borrower used an additional loan to pay interest on first loan. Of course, in that case, the fees compound as they would on any type of debt.

Thank you for responding to my comments. After reading through the Pew study you cited, it seems as though the evidence being used to support arguments in the blog post is being cherry-picked. In your reply, the average income of a payday borrower is cited as evidence of them being of 'low to moderate' income, certainly not poor. However, the data in the Pew study show that the income bracket most likely to use payday lending services is between $15,000 and $25,000 per year, and this is household income. Also, your example of Jane borrowing is much more detailed in the Pew study. On average, borrowers borrow $375 eight times a year, and pay $520 in interest. This is to cover everyday living expenses. Also, according to Pew, odds are that Jane rents her home, earns less than $40,000, is divorced or separated, and is African American. Finally, the zip codes of payday establishments most likely only sheds light on where poor Americans need to travel each day to work. It does not make a strong argument for where they live.

Having been in the "Payday Lending" space for over ten years and as a non-profit Credit Union, we have a different perspective than many based on actual experience with our members. We have also innovated in this space with new mobile applications that use data analytics and SaaS (software as a service). For WSECU, it started with a teller who noticed our members buying money orders to payoff loans at Payday Lenders. We call our loans "small dollar" loans, because the "Payday Lending" brand is so tarnished. Our mobile app innovations are expected to disrupt our traditional underwriting practices and delivery mechanisms. We have shared our experiences with the CFPB and we believe our involvement has been productive for the development of their rule-making process. Suffice it to say there are many inaccuracies in the depiction of how this market works, and many of the experts and policy advocates are not well informed regarding consumer needs and available choices to meet those needs.
Kevin Foster-Keddie

Thanks for your comments, Ryan. It’s important to define what you mean by “poor.” Payday credit users must have a job (or fixed income) and a checking account to qualify so they are not unbanked, or destitute, or the poorest of the poor. A 2012 Pew Foundation study (link below) found the typical borrower had “low to moderate income.” (p. 8). In the 2007 Survey of Consumer Finance, mean income of people who reported using payday loans was about $33,000 and their mean assets were about $83,000. Like Pew, I would call that “low to moderate income.”

Pew also noted that “other factors can be more predictive of payday borrowing than income.” Consistent with that, I found in a blog post “Do Payday Lenders Target Minorities” that once we accounted for past debt problems, income did not help predict payday credit usage. Rather than “targeting” the poor, it is more correct to say (as we did) that payday lenders cater to people having financial problems who can’t qualify for cheaper, mainstream credit who may happen to be lower income, or be minorities.

As for the size of the loan, the Pew Study cited above found the average loan was $375. For many years, $300 was the mode loan size, but it may have increased a bit. A recent study by Bhutta et al. (link below) said loans range up to a maximum of $1,000.

On your third point, we did not say payday borrowers were overoptimistic. We said that if future research could prove that they are overoptimistic, then that could be a reason to intervene.

Roy and Jonathan: The Pew Trust paper the authors referenced (which is actual scholarly research, as opposed to the essay being discussed) state plainly what happens when payday loans are heavily restricted or banned outright: "In states that enact strong legal protections, the result is a large net decrease in payday loan usage;
borrowers are not driven to seek payday loans online or from other sources." and "In states that restrict storefront payday lending, 95 of 100 would-be borrowers elect not to use payday loans at all—just five borrow online or elsewhere."

So there's the answer. These loans are *not* unavoidable, and when they are restricted, distressed people find other ways to solve their problem. *Some* of those other ways are less destructive than payday loans. Sounds like a net improvement to me.

Robert DeYoung, Ronald J. Mann, Donald P. Morgan, and Michael R. Strain:

Let’s not add media deception to the rip-off, please. This post has its facts wrong in the section “Spiraling Fees”. The fees will increase as the principal increases with each rollover. The payday lender will not charge the same amount for a rollover of $345 as it did for the initial loan of $300, and subsequent rollovers that would include each higher fee would incur increasingly larger fees. Your example of a $300 payday loan costing $45 would equate to a fee of $1.50 per every $10 or 15 cents for each dollar. Therefore, the fee for a the first rollover of $345 would be $51.75. The second rollover of $396.75 ($345 + $51.75) would cost $59.51, the third $68.44, the fourth $78.70, and the fifth, when the initial principal plus all subsequent fees would be $603.40, would cost $90.51. Thus, the repeat payday borrower, with an initial loan amount of $300, would owe nearly $700 ($693.91) after 3 months.

Frank M. Miranda

You work for poverty wages. The rent is due NOW. You and your children face eviction. Or you have to buy medicine for an ill child NOW. Or your car breaks down, it's your only way to work, and you'll get fired if you cannot get it fixed NOW. Or you just cannot bear to see your kids go to bed hungry one more day. Such situations are not uncommon. When you face these consequences, you will do anything — repeat, anything — to deal with the resulting Hobson's choice. Even if it means taking out a payday loan, right down the street. Even when you know the cost.

Voila, Wall Street predators get to book another exorbitant debt trap.

As Elizabeth Warren said at the Pew conference on postal banking (July 2014), predatory payday lending constitutes a "massive market failure." By contrast, postal banking ('non-bank financial services'), the public option, can provide affordable small loans to millions of hardworking families. Thus it can drive out predatory lenders while providing the lifeline that desperation borrowers need.

A virtuous circle of positive impacts will result. As just one example, low-cost small loans will reduce defaults by making repayment affordable and, by eliminating exorbitant fees, restore billions of dollars to the lower-income families who earned it — in turn reducing their need for desperation borrowing.

For more, please refer to“public-option”-banking. Thank you.

No human being should ever have to pay interest rates over 12% (Usury should be a crime).

This study plays with numbers and words to make its point. For example, it sets up the straw man argument that payday loans target minorities. Well, no, they don't necessarily target by race, but they do target the working poor and those about to become the working poor, regardless of their race. But that's a harder argument to refute, so straw man it is.

It also uses small numbers to make the egregious fees and interest rates seem less onerous but provides no figures on the average amount borrowed from payday lenders. In fact, it provides no empirical evidence at all in this regard. Why?

In short, take this "study" with a very healthy dose of salt.

This analysis is AWFUL, especially the part on rollovers.

1. The authors trumpet the fact that about half of initial loans are repaid straight away (and refer to this "majority" later) - but this is like claiming drunk driving is okay because a majority of drunk drives do not lead to accidents.

2. The authors then provide a meaningless statistic: 61% of borrowers accurately predict when they will pay off the debt, with about half of the remainder paying it off earlier than expected. This still leaves ~20% of borrowers in debt beyond their estimates - which may or may not be fine, depending on *how badly their estimates undershot the true repayment schedule*. That is the meaningful number here, and the authors don't bother to provide it or address this issue in the slightest.

I'm not interested in what percentage of borrowers are "classical". I'm interested in the following: let's partition loans into "classical" and "malignant" based on how many times they're rolled over. The question is, what fraction of the profits made by this multi-billion-dollar industry are from the "malignant" type? That might actually address the question of how beneficial the industry is on the whole. It's already evident from this post that the majority of profits are from malignant loans (20% of loans are rolled over six times or more).

Ryan: If they are unavoidable, then what happens when the payday lenders are shutdown?

"The key question here is whether the borrowers prone to rollovers are systematically overoptimistic about how quickly they will repay their loan."

Why is that the key question? Why is it even a relevant question?

Buyers of iPhones might be systematically overoptimistic too. So what?

Without a robust theory of "reform," explaining under what circumstances and for what reasons a "reform" is justified, how can any question be selected as "key"?

If the loans are unavoidable, and there is no choice, then what happens when they are banned?

Very interesting analysis. One might expect an unintended consequence of wholesale reform/elimination of payday lending would be to drive demand towards illegal forms of borrowing via organized crime.

Payday lending targets the poor. A Board study cited by the authors reveals that payday lending has the highest number of establishments per person in poor states. Indeed, payday lending is concentrated in areas, where “other income and wealth measures (including educational attainment)” are lower.

A fee of 15 percent every two weeks may not seem like much when the example given – Jane borrowing $300 with a $45 fee – uses small dollar amounts, but these rates of interest can become quickly insurmountable. If, for instance, Jane borrows $5,000 for two weeks, there is a $750 fee. If that amount is not paid back in 10 weeks, the amount owed in fees is $3,750. In just two and a half months, Jane owes 75 percent of the principal borrowed in fees alone. The principal, $5000, may be rolled over for another 10 weeks, but the fees owed, $3,750, would be debited from Jane’s next paycheck before the process starts again.

The authors argue that those borrowing from payday lenders are ‘overoptimistic’. In his study, Mr. Bhutta explains that “potential payday loan customers may have behavioral biases or limitations in analytical ability that make a ban on payday lending welfare enhancing”. Really? Borrowers understand the terms of the loan, but their economic conditions make these loans unavoidable. There is often no choice, no options to be weighed, or a cost-benefit analysis to be carried out.

The comments to this entry are closed.

About the Blog
Liberty Street Economics features insight and analysis from economists working at the intersection of research and policy. The editors are Michael Fleming, Andrew Haughwout, Thomas Klitgaard, and Donald Morgan.

The views expressed are those of the authors, and do not necessarily reflect the position of the New York Fed or the Federal Reserve System.

Economic Research Tracker

Liberty Street Economics is now available on the iPhone® and iPad® and can be customized by economic research topic or economist.

Useful Links
Comment Guidelines
We encourage your comments and queries on our posts and will publish them (below the post) subject to the following guidelines:
Please be brief: Comments are limited to 1500 characters.
Please be quick: Comments submitted after COB on Friday will not be published until Monday morning.
Please be aware: Comments submitted shortly before or during the FOMC blackout may not be published until after the blackout.
Please be on-topic and patient: Comments are moderated and will not appear until they have been reviewed to ensure that they are substantive and clearly related to the topic of the post. We reserve the right not to post any comment, and will not post comments that are abusive, harassing, obscene, or commercial in nature. No notice will be given regarding whether a submission will or will not be posted.‎
Disclosure Policy
The LSE editors ask authors submitting a post to the blog to confirm that they have no conflicts of interest as defined by the American Economic Association in its Disclosure Policy. If an author has sources of financial support or other interests that could be perceived as influencing the research presented in the post, we disclose that fact in a statement prepared by the author and appended to the author information at the end of the post. If the author has no such interests to disclose, no statement is provided. Note, however, that we do indicate in all cases if a data vendor or other party has a right to review a post.