Interchange fees: The latest salvo

By Felix Salmon
June 7, 2010
Tyler Cowen and Matt Yglesias have not had time to read Todd Zywicki's 63-page paper on interchange fees; it certainly doesn't need to be nearly as long as it is. But the fact is that you don't need to read all that far past the abstract to realize how silly and contentious it is.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Tyler Cowen and Matt Yglesias have not had time to read Todd Zywicki’s 63-page paper on interchange fees; it certainly doesn’t need to be nearly as long as it is. But the fact is that you don’t need to read all that far past the abstract to realize how silly and contentious it is.

For instance, Zywicki spends a lot of time making the argument that “the increased revenues merchants receive from shifting credit losses on sales to card issuers by itself exceeds interchange costs” (his italics):

Visa and MasterCard card issuers alone wrote off almost $50 billion in uncollected credit card debt in the U.S. in 2008, and $65 billion in 2009—more than 5% of the total volume of credit card purchases by their cardholders last year. This uncollected debt represents revenue or profits that merchants would not have received without issuing banks assuming the credit risk for those transactions and suffering the loss. In essence, these losses reflect revenues the merchants received from card issuers for sales that were made but not actually paid for by cardholders…

To the extent that a part of the interchange fee represents an allocation of the cost of increased credit risk to merchants (who undeniably benefit from the increased revenues it represents and who otherwise would have to bear that risk themselves), merchant claims that any portion of these fees above the “direct administrative costs” of operating a credit system are unjustified and meritless.

You can see why it might be hard to wade through 63 pages of this stuff, especially when Zywicki’s arguments are as disingenuous as this. Interchange fees are not, never have been, and never should be an attempt to charge merchants for the credit risk of their customers. The $65 billion of credit-card write-offs in 2009 is a large number — but it is significantly smaller than the total amount of interest charged by credit-card issuers. The extension of credit, for credit-card companies, is a profit center, not a loss center. If you’re just looking at the credit side of things, the card issuers should be paying merchants for bringing them new transactions from which they make so much money, rather than the other way around.

You also need to do a lot of footnote-chasing, with this paper. Consider:

Supporting the importance of risk-shifting for credit cards and their value to merchants is the fact that while total (credit card plus debit card) interchange fees collected have, on average, increased slightly, interchange rates on debit cards—which entail almost no credit risk—have been declining.

This surprised me, so I followed the footnote, which took me to this blog entry, written by a colleague of Zywicki. The blog entry, in turn, pointed me to Chart 3 in this paper. Here’s the chart in question:

interchange.tiff

The first thing you note is that the CAGR on the right hand side shows debit-card fees rising by 0.5% a year on average, not falling. And the second thing you notice is that the chart ends in 2004, long before the spike in debit-card fees that everybody is complaining about.

VISA1web.jpgBack in January, Andrew Martin published a much more up-to-date chart, which I blogged here: a glance at it makes clear that if you’re only using data on interchange fees which ends in 2004, you can’t be taken seriously in this debate.

At heart, Zywicki’s argument is that merchants benefit from payment cards, therefore they have nothing to complain about. But this is silly. Yes, merchants benefit from these cards. But that doesn’t mean that the card issuers can or should be able to continue to ratchet up interchange fees. The reason for the Durbin amendment isn’t so the level of interchange fees, so much as it it’s the rate at which they’re rising: these things were profitable for the card issuers when they were much lower than they are now, and all the recent increases have been pure gouging, made possible by the Visa/Mastercard duopoly.

Zywicki spends an inordinate amount of time arguing that payment cards are a good thing, that handling cash is expensive, and various other assertions which no one has any objection to. What he doesn’t do is provide any good reason for rising interchange fees: debit cards, in particular, should be significantly cheaper than cash, for merchants, and increasingly they’re not. Still, he gives it the old college try:

A small increase in interchange fees would make participation in the network more expensive for merchants, but it would also enable banks to enhance their card offerings. Thus, this would tend to increase the number of consumers using that network’s cards and thus make the network more attractive for merchants.

Those ungrateful merchants — they just can’t see that rising interchange fees are good for them!

A lot of the rest of the paper is devoted to the old arguments about how if you stop banks from making money in secret, through interchange fees, then they’ll just start making money more transparently, through higher annual fees and interest rates on credit cards. Which may or may not be true, but it’s a good thing either way, I think. Maybe it will help to swing the pendulum away from credit cards and back towards old-fashioned personal loans, as a way for individuals to borrow money. And yes, it’ll mean lower profits for the banks. I’m perfectly happy with that.

Update: Zywicki responds in the comments. I still don’t follow his argument: I don’t see why merchants, if they extended credit themselves at non-zero rates, would have have lost money rather than made money by doing so. But check out his comment for yourself.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
9 comments so far

If retail, in general, has high levels of competition wouldn’t transactional costs such as interchange be another cost-of-good sold input? If so, while reducing interchange rates would reduce costs of good sold, wouldn’t competitive pressures result in lower prices and the merchants not making any additional money as they rely on price competitive strategies? I’ve struggled with what merchants will get from reduced interchange. The only benefit I can see is if they can move more of their sales to lower cost channels than a competitor they can gain advantage, but I doubt the mom-and-pop stores that are used so much as the face of this fight will be the ones to figure out how to do that.

I guess on the margin a few more sales could be generated due to lower prices, but that seems trivial given the cost savings here.

Welcome any other perspectives.

Posted by TRKAdvisors | Report as abusive

TRKAdvisors- “If so, while reducing interchange rates would reduce costs of good sold, wouldn’t competitive pressures result in lower prices and the merchants not making any additional money as they rely on price competitive strategies?” The market would have to be oftly competitive for none of that money to stay in the merchant’s pocket. And as you say, the cost savings is pretty small on a per-item basis- thus its unlikely stir up that much price competition. And if it does, I think most of us would rather that money be in a consumer’s pocket rather then a bank’s pocket.

Posted by AdamJ23 | Report as abusive

Felix Salmon really should write the Critique of Pure Gouging, and draw a distinction between “pure gouging” and other ordinary profit-maximizing activities. Which is not illegal, not so far.

Posted by billyjoerob | Report as abusive

billyjoerob- I don’t think it’s particularly easy to distinguish between gouging and profit-maximizing most of the time. But when you start seeing marginal costs drop and yet prices still increase it starts to smell like a rat. Not to mention credit cards are charging merchants to assist the credit card company in extending credit- a profitable activity in and of itself. In a competative marketplace that really shouldn’t happen- card companies should be competing to get merchants to use their cards. And low and behold there’s a duopoly in play. Neither Visa nor Mastercard is going to compete on interchange rates- they know they’ll just force the other card company to drop their rates too and not gain any significant market share- neither company will win in a pricewar like that. So instead of a pricewar, they raise interchange rates, and wait for the other card company to follow- which they do, because its more profitable to raise prices then try and eek out a slightly larger market share with price competition. Plus, merchants cannot realistically refuse to accept the major credit cards and they can’t put the costs directly on the credit card using consumer- they have to hide and spread the costs between all customers. Merchants can’t forward the costs solely onto their credit card customers because they’re risk losing those customers- consumers get pretty nasty when they start seeing new fees pop up on their bill- even if the fee is completely warranted. Card companies can keep increasing interchange fees simply because consumers are ignorant of the hidden cost and merchants hands are tied.

Posted by AdamJ23 | Report as abusive

billyjoerob, I’m not talking about legality here, I’m talking about payments. It’s perfectly reasonable for the national payments system to be regulated, no?

Posted by FelixSalmon | Report as abusive

Felix:
Thank you for taking the time to read my paper. I look forward to hearing your contributions tomorrow.

Having said that, I think you’ve misunderstood my paper a bit. The point is a bit more subtle, I think, then you’ve grasped. It is certainly more subtle than the straw man arguments that you claim that I make.

First, I am not arguing that credit risk is overall a money-losing proposition for card issuers for exactly the reason you state–purchases is what generates a customer base for issuers. What I am arguing is that merchants want to argue that the only costs that matter are the marginal costs of processing transactions. That ignores the costs of credit risk. Once that variable is considered, with respect to credit cards, it is not obvious at all that merchants are being overcharged on interchange fees. For reasons discussed in the paper, merchants likely would have issued much of this credit in-house (or foregone sales) and would have suffered similar, and probably larger, credit losses. The argument is not one about charging merchants for the credit risk of their customers (I’m not sure why you read it that way, actually) but to recognize the benefits that merchants get from credit cards. From there it is just a Coasian bargain if credit card issuers can bear risk at lower cost than merchants.

I think maybe you’ve also missed the larger conceptual point here because you’ve misunderstood the purpose of the credit-loss point which is not to justify cost-based pricing: the details of the pricing (rising interchange fee rates) flows out of the two-sided market analysis (newspaper subscription and advertising rates bear little relationship to the relative costs of servicing those two groups), the credit-loss point is related to the fairness point.

For reasons described in detail in the paper, I believe you are simply wrong about the supposed benign value of annual fees. The return of annual fees would be a major disaster for consumers and would substantially reduce competition in the credit card industry. They are also regressive in that they bear no relation to the amount a person uses their card or the amount they charge. I don’t see any reasonable argument to the contrary.

As to whether these costs will be passed on to consumers (“which may or may not be true”) the only research I’ve seen is the Australia experience. And I don’t know of anyone, especially supporters of interchange regulation, who doubts that costs to consumers would rise. It appears that you have some reason to doubt that result. If so, I’d be interested on your basis for equivocating on that point.

As for the final block quote, that is a hypothetical example used to illustrate a theoretical point–I think you’ve simply misread the purpose of that analysis.

As for your charge that this is a duopoly–are you saying that there is an antitrust violation here? If so, why do you think the DOJ doesn’t sue? Are they simply unaware of it?

And if it is a duopoly, what exactly is the supply restriction that you see going on here? Are you saying that there are sustainable economic rents somewhere in the system–which is my understanding of the effects of monopoly? Are you saying the rents aren’t dissipated somewhere in the system, perhaps on the consumer side of the market? Are consumers victimized by the interchange duopoly too? If not, what exactly is your theory as to why the purported duopolists only inflict their harm on merchants rather than consumers? How do you square the arguments of those who say that the problem arises from what amounts to hypercompetition in the consumer side of the market in the form of benefits, etc.?

Or are you just throwing around the term “duopoly” as an epitaph? Because I don’t really see the sustainable rents in the system, the output restrictions, or the logic of how the monopoly rents are supposedly being extracted here.

As for the debit card point, I relied on the research by Tim Muris (former FTC Chairman) and Tom Brown and was not aware of your subsequent blog post with the new data.

The basic problem overall is who bears the cost of payment systems. Merchants like the legacy systems because they are able to externalize the underlying costs of those systems on consumers and taxpayers. Consumers bear the costs of having to go to the ATM, the risk of loss and theft of cash, and much of the costs and risks of using checks. The government prints money and requires that checks be cleared at par, thereby allowing merchants to externalize all of those costs.

The complaint of the merchants, it looks like to me, is that unlike those subsidized systems they have to pay for the benefits of payment cards and they’d like to be able to externalize those business costs onto the rest of us. I understand why they like that. I just don’t understand why the rest of us would.

The real question is whether they should be able to force consumers to subsidize their business costs through higher annual fees, higher costs, and lower quality payment cards. I think not. If you want interchange fee regulation, then that’s the argument you need to be willing to make.

Posted by Zywicki | Report as abusive

Oops, obviously I meant “epithet” not “epitaph.”

Posted by Zywicki | Report as abusive

Interchange fees should be standardized, and independently audited. That definitely isn’t the case at present, which is how a lot of the confusion on this topic arises. Merchants are paying all kinds of different amounts to various service providers, some of whom are banks and none of whom basically do very much to deserve any of the fee amounts they suck out of the system.

Posted by HBC | Report as abusive

Your home is valueble for me. Thanks!…

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/