The Wall Street Journal has a fascinating story on the front page about the operations of banks located inside Wal-Martalthough not really for the reason that the reporters believe it is interesting. The authors think it is interesting because they see it as exposing nefarious practices by these banks, which generate large overdraft protection fees and (it is implied) exploit consumers. In fact, the real story is exactly the opposite–it is compelling evidence of how consumers use alternative credit products, and I suspect, compelling evidence of the unintended consequences of government regulation of consumer credit products. [See update below.]

In an article that former Comptroller of the Currency Robert Clarke and I published in the fall, we discussed in detail how consumers use bank overdraft protection and payday lending and the competition between the two products. In particular, we note that available evidence strongly suggests that in choosing between the two products, consumers generally choose rationally. First, those who use these products (like all alternative lending products) are people who don’t have credit cards or would exceed their credit limits on their credit cards and trigger high fees. So they use these products because they are the least-bad alternative they have at the moment to meet pressing financial obligations.

But even more striking, when choosing between payday lending and overdraft protection, consumers generally choose rationally. As we note in the article, the similar uses but dissimilar pricing of payday loans and overdraft protection provide an excellent natural experiment to examine how consumers use the two products. Payday loans have a fee scale tied to the size of the loan–typically $15 per $100 borrowed. Overdraft protection, by contrast, is a fixed-rate product (mostly)–an overdraft charges a flat fee, typically $27-$35 or so, regardless of the size of the overdraft. As a result, for a single loan, payday lending will generally be less expensive for loans of up to $180 (depending on a particular bank’s fees) and an overdraft will be less expensive beyond that level.

This differential pricing gives rise to a natural experiment: if consumers are rational, they will tend to use payday loans to cover smaller bills and overdraft protection to cover larger bills. Sure enough, an ingenious study by Brian Melzer and Donald Morgan find exactly that result.

And as a matter of anecdote, that is exactly what the WSJ reports today (emphasis added):

CLEVELAND—On a rainy morning in April, Anna Proctor entered a Wal-Mart Supercenter near some of this city’s poorest areas to get $300 for urgent car repairs—money she didn’t have.

Inside, she joined a line at a Woodforest National Bank branch and intentionally overdrew her account. When her paycheck was deposited 12 days later, she said, the bank would take the borrowed sum plus a $30 fee.

“It’s cheaper than a payday loan,” said Ms. Proctor, a 35-year-old customer-service worker. If her overdraft and fee were calculated as a loan, the annual percentage-rate interest, or APR, would be over 300%. She said she overdraws “all the time.”


Some customers at banks inside Wal-Marts said they previously used payday lenders but switched to overdrawing bank accounts because it is less expensive. That was the case with Frank Owens, 38, who opened his Woodforest account in the Cleveland Wal-Mart because of “financial difficulties” at another bank.

Mr. Owens said he overdrew $300 in January and Woodforest repaid itself $330 from his next disability-check deposit three weeks later, the equivalent of a 174% APR. A typical Cleveland payday lender would charge $45 on a two-week $300 loan, local consumer advocates say—a 391% APR.

Moreover, consider this passage further down in the story:

Woodforest’s Mr. Marling said he is aware that some customers are using his bank’s overdraft program as a payday-lending substitute. Although “that’s not what the program was designed to do,” he said, Woodforest employees often advise customers who are short of cash to take out their entire overdraft allowance at once, incurring only one overdraft fee. “To me, that’s a very efficient way of somebody having a need and paying a fee to get it.”

Now let me emphasize that I am not ignoring the possibility (as also reported in the story) that some people can end up overusing overdraft protection, just as they can payday loans–or traditional mortgages, for that matter, as we saw over the past decade. Moreover, there are possible consumer protection issues here. Clarke and I discuss these considerations in our article as well. But that doesn’t detract from a key takeaway from the story–that heavy users of these overdraft services seem to be aware of what their options are and what they are doing and are aware of the relative prices of payday loans and overdraft protection and use their options accordingly. (Shhhh, don’t tell the editors of the Harvard and Chicago Law Reviews or believers in nontestable behavioral economic theories).

But there is another element of the story that is also interesting and potentially speaks to unintended consequences of government regulation. The story is accompanied by a chart of banks that derive an unusually high percentage of their revenues from overdraft fees. From looking at the chart, at least four of the 10 banks listed appear to be banks specializing in lending to military members or located on or near military bases. Why is that significant? Because in 2007 Congress passed the military lending act, which imposed a 36 percent APR price ceiling on loans to military members (essentially outlawing payday loans for soldiers)–but left overdraft protection unregulated. As I have discussed previously, in light of the fact that consumers can and do substitute between payday loans and overdraft protection, one would expect that the loss of access to payday loans would lead soldiers to increase their usage of overdraft protection. Given the unusually high representation of military banks among the 10 largest generators of overdraft fees, that appears to be exactly what has happened. I haven’t had a chance to dig into the FDIC data, but I would predict that if one examines the overdraft fees generated by the military banks one will find a dramatic upward surge when the Military Lending Act was passed.

Let me emphasize that I am not denying that not all consumers fully understand their usage of these products and dig themselves a hole–just as many others do the same with credit cards, home mortgages, or home equity loans. And they are certainly expensive and should not be used lightly–I am no enthusiast for these products and I wish we lived in a world where every consumer was financially responsible, had stable income and expenses, and had easy access to inexpensive, high-quality credit. But that’s not the world we live in. And we see that when we take away preferred choices from consumers, even choices that don’t look ideal to government bureaucrats, consumers are not necessarily going to be made better off as a result. Instead, in many cases they will simply be forced to use even worse and more-expensive choices. So regulators should be very cautious about taking away products from people who use them, and by all indications use them rationally in light of their particular circumstances.

As I’ve said before: “Regulators cannot wish away the need of low-income consumers for credit…. Congress can pass all the laws it wants, but it can’t repeal the law of supply and demand and the law of unintended consequences.”

Update: It has been pointed out to me that I unfairly interpreted the story as suggesting that the practices of these banks was “nefarious” and the like. That was my characterization after reading the story but that the story was more nuanced and neutral than that. I have removed that characterization and will let the article (an excellent piece of reportage, btw) speak for itself. My apologies to the reporters on the story.