09 September 2009

Overdraft Rant

I'm really rubbed the wrong way by blog posts like this one from Felix Salmon and this one from Kevin Drum based on today's NYT story on overdraft fees on checking accounts. I just can't get behind anyone who knows — knows for sure, with complete metaphysical certitude, as John McLaughlin would say — what any product or service ought to cost. "I think the fees ought to be $35." "Well I think they ought to be $20." "Well I think they ought to be free!" Why are these people pulling numbers out of the √¶ther and claiming they're the One True Price?

Look, I agree that it is better for banks' fees to be openly disclosed so they can be compared to the competition. And I agree that opting out of programs like overdraft protection ought to be easier rather than not. I will even consider proposals to limit fees to one per day, or having some provisions for customers with deposits pending to their account. But where do people get off thinking they know exactly how much something specific like an overdraft fee ought to cost? How is it any more justifiable than me saying "You know what should cost less? Felix Salmon columns! Publications should only be allowed to pay up to $109.42 for any written material from Felix Salmon."

If you don't like your overdraft fees there's a simple solution. Don't overdraft your damn account. You know how many times I've over drafted my account? None. Never happened. And believe me, it's not because I have so much money that there's always a big cushion. I wish it were otherwise. You know who's fault it would be if I did spend more money than I had? Here's a hint: not my bank's. The government is not our mommy or our daddy or our financial adviser. Man up and take some responsibility rather than demanding that we create new regulations to slap ever more foam padding over all the pointy bits of the universe.

My paternal grandmother never had much money, and she never once overdrafted her account. That's because she never used a debit card. My father picked up the habit from her; he's never used a debit card either. Neither of them wanted to give themselves any extra rope to hang themselves, either in terms of overdrafts, or in exceeding their self-imposed budgets, so it was cash or check all the time, and every check went in the ledger as soon as they got home, if not sooner. (Side note: using cash is faster than using cards at point of sale, on average. I would not have guessed that.)

(I feel like such a conservative right now, especially based on Tyler Cowen's characterization: "10. Responsibility is a more important value than either liberty or equality [to conservatives]." I'm having flash backs to my youth, before I discovered libertarianism and grew out of conservatism. I'm also feeling quite cranky, and come morning I'll probably regret being all "Bah humgbug, they deserve it!" (Even though they kind of do deserve it.))

Here's some info from Salmon:
And while back in July I said that 20% of bank customers pay 80% of the overdraft fees, in fact, according to the most recent FDIC report, it’s worse than that: the 13.9% of customers who get charged 5 or more fees per year pay a whopping 93.4% of the banks’ total fee income. And the 4.9% of customers who get hit 20 or more times per year are paying an average of $1,610 apiece in these fees. That’s money they really can’t afford.
I agree that's money they can't afford. Which is a great reason they shouldn't overdraft their accounts every other week. That sounds heartless of me, but seriously, you have some people that chronically spend money that isn't theirs. How is the solution for this to absolve them of responsibility? Also, the assertion that the $1610 is money they can't afford is true insofar as no one really wants to spend that money on fees, but the implication is that these 4.9% of customers are particularly less wealthy than the median. That may be, and probably is, true, but it's still assuming facts not in evidence. What I'd like to know is what proportion of our chronic overdrafters use the register in their checkbooks for debit card purchases or check their bank balance online regularly and often. I'll wager fewer of them do than the general population, from which we could only conclude that people who put little effort into avoiding overdrafts get stuck with the bill for the most overdrafts. The horror! The injustice!

Does anyone else get the impression that Salmon thinks that it's somehow scandalous that overdraft fees are not uniformly distributed throughout the population? Presumably responsibility will never be uniformly distributed, so why should the costs of irresponsibility be so? If the 93.4%/13.9% distribution is "worse" than the 80/20, would a 70/30 distribution be better? Would society really be better off in any meaningful way if the Gini coefficient for overdraft fees was 0.0?

This whole business about an epidemic of overdraft fees is also annoying because ... well because IT ISN'T HAPPENING. The plural of anecdote is not data, and all that. The total charges from overdraft fees have gone up 90% since the beginning of the decade. But the total amount charged on debit cards has gone up 333% over the same time period, and the total number of debit card transactions has gone up 378%. (That's data the NYTimes doesn't bother to present in their article, but can be derived from the exceedingly misleading charts they published with it. We're talking irresponsibly misleading. Lies, damn lies and statistics style misleading.) People are getting fewer penalties per transaction. Shouldn't that be evidence that overdraft problems are lessening? I may be being too curmudgeonly about people's overdraft fees, but can we at least agree that this appears to be a trend piece about a trend that doesn't exist?

(Hat tip to Coyote Blog for pointing out the error in that chart. He has more commentary on the matter here.)

PS This also bugged me in Salmon's post:
There’s also no debunking of this kind of thing:
Michael Moebs, an economist who advises banks and credit unions, said Ms. Maloney’s legislation would effectively kill overdraft services, causing an estimated 1,000 banks and 2,000 credit unions to fold within two years. That is because 45 percent of the nation’s banks and credit unions collect more from overdraft services than they make in profits, he said.
This simply doesn’t follow: just because a bank currently has more overdraft revenues than it makes in profits does not mean it’ll fold in the event those revenues go away. Especially not at credit unions, which don’t exist to make large profits: at most of those 2,000 credit unions, overdraft revenues are probably a very small proportion of total revenues and won’t make much of a difference if legislated away.
So he's annoyed at what he sees to be an unverifiable claim, and he responds with his own claim which is also based on no evidence. He's right that it is possible for banks to find something else to replace the lost revenue, but he's still making an argument that amounts to "I disagree, therefore I am right." And besides, if there are such easy revenue streams out there to substitute for overdraft fees, why aren't banks already tapping into them? Overdraft revenues probably won't make much of a difference anyway. Yeah, that's a solidly verifiable, debunkable, empirical claim for you. That's only a step above Well, I'm pretty sure it doesn't matter, so there! Nana-nana-nanaa pffffffffttt!

PPS From the NYTimes: "Rather, regulators and lawmakers say they hope to curb abuses and make the fees more fair." InformationGain('fair') == 0.0.

PPPS Again from the NYTimes: "For instance, some banks have said they might slap a monthly fee of between $10 to $20 on every free checking account [if overdraft fees are limited]. At the moment, people who pay overdraft fees help subsidize the free accounts of those who do not." Super. By all means, let me prop up people with irresponsible financial habits. We're not doing enough of that in America yet. I am totally psyched about the prospect of spending ~$180 a year to free my fellow citizens of the burden of knowing how much money is in their checking accounts.

I also love the implication that people who spend money they don't have are somehow being victimized by being forced to subsidize the responsible account holders. Every business has different profit margins on different consumers. My groceries are "subsidized" by the people who impulse buy the candy bars in the check out aisles. My video rentals are "subsidized" by the people who keep their movies late. There's nothing wrong with that. I'm failing to see where the moral panic is in this.

This whole article is framed like some epic affront to social justice, rather than people just not liking the terms of their contracts. Behind what Veil of Ignorance does one stand to end up hoping to be born into a world where everybody shares the burden of some people spending money they don't have?

In order to play up the injustice angle the NYTimes also uses an anecdote of a mentally challenged guy whose caretaker sister couldn't un-enroll him from automatic overdraft protection on his B of A account. She claims she tried for years to get him out of it, but never (a) bothered to move his account to another bank without overdraft protection, or (b) got him one of those stored-value debit cards marketed for children that would cap his spending.

Also, let's make this explicit: the Times is saying that there's this one guy (who is perhaps representative of multiple guys?) who is legally allowed to have his own checking account, but is mentally incapable of understanding the consequences of using it, and thus should not be fully liable for its misuse. Because of this, everyone — all adults — need special protection from overdraft fees. We can't expect this one handicapped guy to be responsible, and so we shouldn't expect you, dear reader, to be responsible either. That's the argument they're making by using this guy as an exemplar.

I've gotten way off topic from ranting about people who know ("know") how much things ought to cost generally, and ended up grumbling about overdraft fee reforms specifically, so I'll bring things back around with this Megan McArdle anecdote from last week about someone who knows ("knows") how much insurance should cost:
I remember having a conversation with a coworker within three minutes a) complained that there was no reason that health insurance should cost so much and b) insurance was really important, because a few years ago his wife had had a baby prematurely with massive complications for her, and if they hadn't had insurance it would have cost several hundred thousand dollars.
This is a little more subtle, but the same problem. To assert something costs too much (or too little) is to implicitly make the claim that you would recognize the "correct" price for it. Even the people who set the prices don't know the "correct" prices, because there is no such thing as a correct, morally justifiable, philosophically grounded One True Price. There is only what people will agree to at any given time, and that is dynamic. Everyone claiming prices are wrong are Monday Morning Quarterbacks at best, and whiners at worst.


  1. I no longer overdraft my account. Years ago it'd happened on occasion (largely from something, usually a check, that I hadn't accounted for and wasn't cashed in a reasonable time). I fixed that by actually balancing my books on an ongoing basis.

    My wife is better than she used to be, but does occasionally falter. Her experience, however, is indicative of the one major issue I have with the fees.

    Banks debit the highest-dollar expense first, the next-highest second, continuing down to the lowest-dollar amount charged. So if you go buy a coffee in the morning, for $3, a magazine for $2, lunch for $8, a pack of gum for $0.25, and that night spend $25 on dinner, and you use your debit card, your daily expenses are $38.25. But they are not debited from your account in the order 3, 2, 8, 0.25, 25. They are debited 25, 8, 3, 2, 0.25.

    This means that if you have an unexpected unaccounted-for expense that leaves your account with only $30, you go overdraft on the $8, $3, $2, and $0.25. At my bank an overdraft charge is $35, so for going $8.25 overdrafted you're charged $140. If they charged in chronological order (or even, to be nice, in ascending order of dollar value rather than descending), the same overdrafts would only cause a single $35 fee.

    The banks do this claiming that they don't want to cause you a problem by denying a larger (and more important) payment while accepting the smaller and more inconsequential payments. But here's the kicker... I've NEVER seen them deny a payment. Even for my wife, who had this happen probably 4-5 times over the course of two years (establishing a pattern), they NEVER denied a charge.

    Based on this, it is clear that their preference of charging the highest dollar value first is done with the goal of maximizing fees, not (as they claim) protecting the account-holder. As an anarchist, I don't suggest government regulation is the answer... But it's shady and makes me think the banks are a bunch of swine.

  2. That's a fair criticism, and I admire the restraint in not calling for the government to Do Something About It. I'm not really convinced by the banks' argument for why they sort transactions that way myself. (I do have a suspicion though that any sorting method would lead to dissatisfaction of some sort.)

    This is part of the reason I said I'd be willing to consider limiting penalties to one per day rather than one per transaction. I'm not sure I support such a rule, but it's reasonable enough to discuss.

    I'd be willing to discuss plenty of things about the way these systems are set up. I just don't appreciate people trumping this up into a big social justice matter, or people claiming they know the correct price of some specific fee. I'm more interested in denouncing those two particular positions than I am in defending overdraft policies in the abstract and as a whole.