According to the FDIC, payday lenders typically charge $15-20 per $100 advanced for a two week period, which represents an APR of almost 400%.

The Uniform Retail Credit Classification and Account Management Policy, which is a policy issued by the Federal Financial Institutions Examination Council, establishes guidelines for extensions, deferrals, renewals, or rewrites of closed-end accounts, which include payday loans. Under this Policy, institutions should:
(1) limit the number and frequency of extensions, deferrals, renewals, and rewrites; (2) prohibit additional advances to finance unpaid interest and fees and simultaneous loans to the same customer, and (3) ensure that comprehensive and effective risk management, reporting, and internal controls are established and maintained. This policy is probably why personalcashadvance.com has disclaimers that the loans are not for repeat use.

I also wonder to what extent Christine is right that payday loans involve little or no risk to the lender. Payday lenders don't get repaid directly from the borrower's paycheck. They either take a check to be cashed at a particular future date or receive the right to debit the borrower's checking account on a particular future date (which is what personalcashadvance.com does). If the borrower's bank account has insufficient funds on the collection date, then the check bounces or the lender is unable to make the debit. Not everyone has direct deposit and even if they did, they may be able to withdraw all or most of the funds before the check is cashed or the electronic debit occurs. This problem of insufficient funds is not unusual. If the credit rating of the borrower is truly wretched, then the lender may have significant difficulty collecting on the loan.

It is also worth noting that if a borrower has insufficient funds in his bank account when the check or debit is due, the borrower will have to pay a nonsufficient fund fee on top of his payday loan fees, plus a fee to the lender for having insufficient funds. All of these fees may significantly increase the total cost of the loan for the borrower.

Ron (and Christine)
you might also want to check out this:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=694141
Predatory Lending and the Military: The Law and Geograpy of 'Payday' Loans in Military Towns
Here's the abstract:

Abstract:
A heated national debate has developed over whether one type of high-cost predatory lender, commonly known as "payday lenders," are targeting financially vulnerable military families and whether the law protects them from such predation. Writing within the relatively new interdisciplinary "law and geography" movement, this article provides geographic evidence that payday lenders do aggressively target American military personnel, irrespective of most forms of legal regulation. This paper first provides a comprehensive introduction to payday lending business practices and to the financial vulnerability of military personnel. In conducting our empirical research, we examined 20 states, 1,516 counties, 13,253 ZIP codes, nearly 15,000 payday lenders, and 109 military bases. We consistently found high concentrations of payday lending businesses in counties, zip codes, and neighborhoods in close proximity to military bases. Our observations were controlled by comparing payday lender densities in military areas to statewide averages and also by comparing payday lender locations to bank locations. Of the twenty states involved, the only legal strategy which prevented payday lender targeting of military personnel was New York's aggressive enforcement of civil and criminal usury law. Going beyond the debate over predatory lending to military personnel, our research provides a realist check on pure legal reasoning and unfounded faith in current consumer protection rules.

Christine: If you really hate these programs, how about organizing a group of like-minded individuals to pool some money together and offer a similar service. You could charge a much lower rate and still make money because the industry, as currently constructed, is obviously charging rates which in no way reflect the risk involved or what the market should bear.....I'm confident this could work because academics and politicians (see WalMart & the Maryland legislature)invariably know more about a consumer service than the greedy @%$#@s actually providing the consumer service.

The study that Dan Markel mentioned above is truly embarrassing. It has no regressions. Not a single one. Consequently, it does not control for anything. A respectable study must control for the following characteristics of the nearby non-military populations: income, unemployment/underemployment levels, % home ownership, median age, % of households headed by a female, family size, racial composition, % of immigrants, education levels... you name it. This paper essentially assumes that all those characteristics are distributed equally across each state, and that military bases are placed in random locations. To calculate the “predicted” number of payday lenders, for example, it simply calculates statewide number of payday outlets per person and multiplies that by number of people in the zip code. It’s like saying that Beverly Hills and South Central LA should have the same number of payday lenders per person. Even for those silly “findings”, they don’t report t-statistics (is the difference between “predicted” and “actual” number of lenders significant?).

Pure garbage. Took me about 3 minutes to figure it out. Unfortunately, most papers that make similarly sweeping claims are of the same quality.

The paper Dan M linked is being published in Ohio State Law Journal. I think Kate's point about the weakness (or, in her view, complete irrelevance) of law review ranking/reputation as a signal of article quality is probably most persuasive with respect to empirical pieces. Law review editors have limitations as evaluators of traditional legal scholarship, but nowhere near as severe as their limitations evaluating empirical work; there's simply no reason to think that getting good 1L grades (the criterion for most law review editorships) indicates an ability to separate impressive from flawed empirical work.

It does seem bad that a law review picked it with such a glaring flaw, but a professor did just recommend it to another professor here on this site...

Another few thoughts, as this topic winds down:

First, it is clear that payday lending rates are much higher than credit card rates, probably about 20 times as high in APR terms (400% versus 20%). It also is relevant that these are not short-term rates in truth; the data we have make it clear that most of the interest comes from long-term semi-permanent borrowers. The hardest question for me, as to which I have no real data yet, is the extent to which those fees in fact understate the true rates. As commenters noted, there is considerable potential here for later fees for bounced checks in the like -- "shrouded" charges, to use Laibson's term -- that might drive the "true" rates higher. My guess, completely uninformed, is that those fees are prevalent in some payday lending niches but not others, but I don't yet have a good handle on that.

Kate is correct that my goal is to study payday lending, not to bury (or embrace) it. My coauthor Jim Hawkins and I picked this industry because it probably is the most reputable of the fringe lending industries, in that it involves set fees rather than complex APR calculations and doesn't pose problems with collateral valuation by consumers. Thus, in a way it presents the most interesting regulatory questions. If we can't tolerate this industry, then we have to take an unusually strong stance against lending to the marginally banked. I think we confront head-on the issue whether banks should be involved or not, the possibility for an increased role of the welfare state (if we choose not to permit this lending), and some interesting federalism questions. But, as Kate and Dan allude to, there are a number of dubious empirical facts that presently are being used to inform legislative debates. So we will need to untangle some of those issues, before making any informed proposals about short-term lending to credit-impaired borrowers. I thank everyone for the suggestions.

Law student editor: that was a funny comment, but not entirely fair. It’s one thing for a colleague outside your field to recall seeing something vaguely relevant and point it out to you without investigating the quality first – and it’s completely different for a journal editor to read a paper (presumably carefully) and recommend it for publication.

In a calmer mode, I should commend the authors of that paper for impressive data-gathering efforts. Good idea, a lot of work, bad implementation.

I thank Kate for the defense. I can't vouch in any way for the study's soundness, nor did I endorse it; I just thought it might be "vaguely relevant."

It seams we are stricken with the decision to limit the rights of the innocent to protect the rights of those who we perceive as not having the ability to make their own decisions.

I'm of the belief that this country, which was supposedly based entirely of the idea of freedome, should maintain that logic. In other words, the government should stay out of it.

Brandon Drury