On Poverty, Interest Rates, and Payday Loans

Felix Salmon responds rather pungently to my post on debt.  I certainly didn't mean to imply that Felix's position is unreasonable--it's not, and a lot of people hold it.  I just think it's tricky.

I'll cover some of our disagreements in a minute, but I think this is really interesting:

McArdle is far too generous to the lenders here. For one thing, I made it clear in my post that credit cards are very good for transactional credit: if you need to pay the car-repair shop today, using a credit card is a great way of doing so. But you should also have a good enough relationship with your bank that by the time the credit-card bill comes due, you can pay it with the proceeds from a personal loan or line of credit.

Secondly, I don't think for a minute that we should deny the poor credit; in fact I'm on the board of a non-profit institution which exists to provide credit to the poor, and I'm all in favor of that. It's credit cards I don't like, with their high fees and interest rates (and there are even exceptions to that rule, such as the ones provided by many credit unions). And I really dislike payday loans, which are pretty much universally predatory, especially when compared to similar products from community development credit unions.

Megan's conceptual mistake here is clear when she says that "credit extended to the poor carries high interest rates to cover the default risk". But in fact the interest rates on credit cards are really not a function of default risk at all. Mike Konczal had a great post on this back in May, where he showed pretty conclusively that credit-card interest rates were all about maximizing profit for the issuer, rather than compensating for default rates. And payday loans are even worse.

What earthly grounds does Megan have for saying that the number of people made worse off by payday loans is smaller than the number of people made better off by them? I suspect she considers the alternative to be no-credit-at-all-nohow-noway. But that's not what anybody is proposing. I, for one, think that credit should be available to the poor, very much so. But not in the quantities and at the rates that it's been available until now. There is such a thing as too much credit, and we crossed that line long, long ago.

It's an odd fact that poor people shun bank accounts at an astonishingly high rate. Rather than pay $10.00 a month for a checking account, they'll pay more than that to a check cashing place.  Of course, it's not like banks are going after those clients, because they're not very profitable--small accounts still have almost all the transaction costs and overhead of large ones.  But why don't the customers go after the banks?

The plausible reasons I've heard:

  • Check cashing places give you the money immediately
  • Poor people are disproportionately subject to judgments and garnishments that make it preferable to operate in cash
  • People working off the books don't want a trail for the IRS to follow
  • For people with low incomes, the costs associated with a mistake--bounced check fees, for example--can be devastating.  But if you don't have the fees, people will overdraw their accounts.
  • Check cashers keep longer attractive hours and have better service

As Felix could no doubt attest at great length, this problem has proven hella stubborn.

The problem of payday lenders and credit cards, however, is not a problem of the unbanked.  If you don't have a relationship with a bank, you almost certainly do not have a credit card, and you definitely aren't using a payday lender.

So why are people using credit cards and payday lenders?

Credit cards have low transaction costs, which is why, as Felix argues, people use them for sudden emergencies.  Many of them would be better off if they did go to their credit union for a personal loan to pay off the balance.  On the other hand, if you're planning to pay off the balance in a couple of months, that's overkill--and the loan inquiry will ding your credit.

Payday loans are a different question.  There's a lot of literature on them, but most of it agrees on a few points.  For our purposes, the salient characteristics of payday borrowers are a) they have little-to-no money in the bank b) they have moderate incomes and  c) they are fairly severely credit constrained.  Virtually all payday borrowers use some other sort of credit (Stegman and Faris, 2003). At least 60% of them have access to a credit card (Lawrence and Elliehausen, 2008).  73% of them have been turned down for a loan in the past five years, or received less credit than they asked for.   If they're turning to payday loans, it's because they have maxed out those other forms of credit, and they have some pressing cash flow need.

Payday borrowers do not necessarily turn to payday lending out of ignorance; a majority of them seem to be aware that this is a very, very expensive form of financing.  They just have no better options.

The biggest problem with payday loans is not the one-time fee, though that is steep; it's that people can get trapped in a cycle of rolling them over.  Paying $15 to borrow a few hundred bucks in an emergency is bad, but it's probably manageable for most people.  Unfortunately, since payday borrowers are credit constrained, have little savings, and are low-to-moderate income, they often have difficulty coming up with the principal when the loan is due to pay off.  The finance charges add up, making it difficult to repay the loan.

According to Lawrence and Ellihausen, about 40% of payday borrowers fall into that problem category:  they have rolled over a loan five or more times in the past year. A hard core of about 20% had rolled over 9 or more advances.

Judging who is worse off is a pretty tricky task.  Would payday borrowers be better off if they had no other debt, and could go to their credit union for a tidy personal loan?  That's unquestionable.  By the time they're at the payday loan stage, however, that doesn't seem as if it's usually an option.  I'd say that the people who are rolling over 9 or more loans are definitely worse off, the people rolling over 5-9 loans are probably worse off, and the majority who are rolling their loans over no, or a few times are probably better off, given the circumstances they were in when the time came to get the loan.  People who roll over loans only a few times are not trapped in a debt cycle, and (I'd guess) are unlikely to have been using the loans for ordinary expenses.

There's some experimental and empirical evidence to support this.  Wilson, et al (2008) built an experimental model of credit-and-cash constrained households, and found that adding payday loans contributed significantly to household financial survival in the lab.  Which seems to also be true in real life, according to their paper:

Georgia banned payday loans in May 2004 while North Carolina banned them in December 2005. These two events provide the authors with an opportunity to empirically investigate several effects of the removal of payday loans on household behavior. Morgan and Strain find that relative to households in other states, households in Georgia bounced more checks, complained more frequently to the Federal Trade Commission about lenders and debt collectors, and were more likely to file for bankruptcy under Chapter 7 after the ban of payday loans . . . The results for North Carolina, which the authors regard as preliminary, given the shorter period in which payday loans have been banned, are similar to those for Georgia.   

But as Bart Wilson told me the last time I saw him, they also found a minority were made much worse off by the loans.  Those were the people who took out ten or more--and just as Lawrence and Elliehausen found in the real world, those extreme borrowers made up about 20% of the group.

There is, of course, the question of what happens to people between the time when they had no debt, and the time when they need the payday loan.  If we could constrain them during that period from maxing out their available credit, they'd never need a payday loan.  People who have maxed out their credit and are getting turned down for loans could probably have used an intervention that would force them to match income to outflow.

But I'm not sure how you do that.  Say we slap on a usury law that makes credit card lending to poor people unprofitable, so people use personal finance loans instead.  Well, the people who are getting payday loans now would, in this alternative universe, have already maxed out this line of credit.  How do we know that?  Because they seem to have done it in this universe. I don't know whether that's because they're irresponsible, or because they had a string of really crappy bad luck.  I'm not sure it matters.

The core problems we would actually need to solve to get rid of payday loans are first, that some people have marginal incomes and no capital, and second, that when credit is available, some of those people do not exercise the incredibly tight spending discipline which is required to achieve financial stability on such an income.  Because their incomes are marginal, and the lives of the working poor are fraught with all sorts of extra problems, like cheap cars that break down constantly and landlords who turn the heat off, the people who do not keep very tight control of their money are fairly likely to end up in a place where they have exhausted all other credit lines, and are forced to pawn something, hock their car title, or take out a payday loan.

And those loans are jaw-droppingly expensive.  Even non-profit payday lenders apparently charge about a 250% APR, because the loans have a 10-20% default rate, and the transaction costs on lending small amounts are very high. Of course, the profits are usually quite substantial, with APRs often double the non-profit rate . . . and even I have to wonder how a guy who made his fortune lending money at 600% o society's most financially unstable people, smiles at himself in the mirror every morning.

In principle, I agree that many poor people would be better off if they were able to borrow a lot less money at better rates (though even then, I always wonder if I'm not just imposing my monetary time preference on others).  Only when I look at any given rule aimed at accomplishing this, it always hurts a lot of people, even as it helps others--I think the last twelve months have proven fairly conclusively that the supply and price of credit are not entirely unrelated to default risk.  While it is absolutely true that credit card issuers maximize their returns through hefty stealth charges, and payday lenders charge absolutely rapacious interest rates, it is also apparently true that these awful loans often help avoid even worse fates.  And I don't see any way to cut off the credit to people who are ignorantly or irresponsibly getting into trouble, without also cutting it off to a bunch of people who need it.

So I think focusing on the lender side is usually a mistake, though I can't say I'd be sorry to see caps on what payday lenders can charge.  The lender side makes us indignant, because hey, they're getting rich by charging outrageous rates to those least able to pay them!  But if we want to actually improve the lives of the borrowers, we need to intervene before they get to the payday loan point, rather than try to stop them from getting one once they're there.  Felix is doing God's work on just that problem, as are many other people in many other ways.  I think we'll be better off when payday lenders go out of business due to lack of demand, not prohibited supply.