Adam Levitin defended the Durbin Amendment imposing interchange price controls on debit cards on The Huffington Post Friday. From reading the post, it is not clear if he is just defending price controls in the abstract or whether he is actually defending the particularities of the Durbin Amendment. My reading is that this piece is only the former. Obviously someone writing in that context has limited space and limited reader attention, so I wouldn’t necessarily infer anything about whether he defends the details. Given that some of the details are really quite indefensible (and in my view highly indicative of the real rent-seeking motivation behind the whole law), however, I would be interested to know at some point whether Adam’s defense extends to the details of how the price control provisions are calculated and some of the additional provisions such as the ability of retailers to override network agreements that prohibit merchants from setting maximum charge levels.
Some of Adam’s arguments I have dealt with elsewhere, such as the argument that the market operates in an anticompetitive manner. As Josh Wright and I observed yesterday, the problem with this argument is that it only looks at half of the market. It ignores the heated competition on the consumer side of the market. More importantly, neither Adam nor any other defender of price controls has provided any evidence that there are sustainable economic profits produced by interchange fees once the full costs of operating the system are taken into account, including the pricing structure ont eh cardholder side of the market (note that this is not the same as positive accounting returns). Absent rents, there is no reason to infer anti-competitive market outcomes. The persistence of these arguments in this context brings to mind Ronald Coase’s famous observation: “One important result of this preoccupation with the monopoly problem is that if an economist finds something–a business practice of one sort or other–that he does not understand, he looks for a monopoly explanation. And as in this field we are very ignorant, the number of ununderstandable practices tends to be very large, and the reliance on a monopoly explanation, frequent.”
So let’s examine some of the other points raised by Adam (btw, I don’t speak for Visa, just for myself):
“Visa argues that swipe fee regulation will hurt consumers by leading to an increase in bank fees to consumers.” Adam acknowledges this argument but then he doesn’t address it, but instead goes on to address a different point (“To support this, Visa cites the case of Australian swipe fee regulation, which it claims did not result in any benefit for consumers.”) That’s actually apples and oranges–the first point is that bank fees went up for consumers, the second is a different point, which is that cost savings to merchants were not passed on to consumers to the same extent as bank fees rose for consumers. That is the correct overall question from a consumer welfare perspective–will the increase in bank fees on consumers be offset by a larger drop in retail prices as a result of lower merchant costs–but they are distinct.
As Adam notes, the best (really only) evidence we have on this is the Australian experience. All analysis of actual changes in prices to cardholders of which I am aware indicates that there was an increase in effective prices for Australian consumers as a result of interchange price controls.
Adam does not state exactly where he is drawing his support from, but I assume it is this report of the Reserve Bank of Australia on the effects of interchange price controls in Australia.
Here’s what the RBA report said about the effect of interchange fee caps:
The changes in price signals that have occurred reflect both changes in interchange fees and the introduction of surcharging on credit card transactions. Lower interchange fees in the MasterCard and Visa credit card systems have resulted in a reduction in the value of reward points and higher annual fees, increasing the effective price of credit card transactions facing
many consumers. For example, the effective price of a $100 transaction where the credit card balance is paid off by the due date has increased from around -$1.30 prior to the reforms
(reflecting the value of the interest-free credit and reward points) to around -$1.10 currently.
The RBA’s conclusions are consistent with all of the other direct evidence on consumer card prices that I have seen, including the study by Robert Stillman that found that annual fees increased by an average of 22% on standard credit cards and annual fees for rewards cards increased by 47%-77%. Available evidence also suggests that grace periods were shortened, other fees (such as over-limit and late fees) increased, and some have also argued that there were reductions in card quality (such as security) although that is contested by the RBA.
In fact, in Australia, the entire purpose of the interchange price control was to increase the costs of card usage by consumers (an important point that we’ll come back to later).
So, as of now, there seems to be little doubt that capping interchange fees on credit cards would increase other prices for consumers, especially for transactors who do not revolve their balances or pay much in the way of penalty fees. I can’t think of any reason why this same general analysis would not apply to debit cards. Some may argue that is a good thing as a normative manner for cardholders to pay more–this is the position of the RBA. As a purely positive manner, that seems to be the likely effect.
“The Reserve Bank of Australia concluded that as the result of swipe fee reform, rates dropped, and merchants passed on significant cost savings to consumers: around $1.1 billion a year, in a market a fifteenth the size of the US. In competitive retail markets, merchants’ cost savings on get passed on to consumers.” As noted, Adam’s response to this first point is actually non-responsive: in response to the argument that the cost of cards will increase for consumers, he instead argues that the cost of retail shopping will decrease. While, as noted, there seems to be little serious doubt that the cost of cards for consumers will increase substantially if interchange fees are cut, the question of how much of this cost savings will be passed through to consumers is a different question.
Surely some of the cost savings will be passed on in some markets–how much and in which markets is the real question. And more importantly, the real question from an overall consumer perspective is whether the amount of the overall price savings to consumers from lower interchange fees is likely to exceed the overall price increase to consumers from higher card prices. That’s an empirical question. But there is serious reason to doubt that consumers will benefit from that trade.
Let me stress at the outset of this discussion that I assume that the only reason that merchants or card issuers pass on benefits to their customers is because they are forced to do so by market competition. Both merchants and issuers will want to keep as much for themselves as they can.
Adam refers to the “conclusion” of the RBA that “rates dropped” and that “merchants passed on significant cost savings to consumers: about $1.1 billion per year.” I can’t find anything in the RBA report or anywhere that says that “rates dropped,” so I’ll just ignore that claim unless and until I see the basis for that. What about the conclusion that merchants passed on savings of about “$1.1 billion per year”?
Actually, while Adam refers to it as a “conclusion,” it would be more accurate to characterize it as “conclusory.” Here’s the entire analysis of the RBA on that point (pages 22-23):
5.2.6 Merchant pass-through of savings
One issue that has attracted considerable attention since the
reforms were introduced is whether the cost savings that merchants have received from lower merchant service fees have been passed on to consumers in the form of lower prices for goods and services than would have otherwise been the case. The schemes argue that there has been no, or little, pass-through, while the merchants argue that the cost savings have been passed through. The Bank’s estimate is that over the past year, these cost savings have amounted to around $1.1 billion.No concrete evidence has been presented to the Board regarding the pass-through of these savings, although this is not surprising as the effect is difficult to isolate. The Bank had previously estimated that the cost savings would be likely to lead to the CPI being around 0.1 to 0.2 percentage points lower than would otherwise be the case over the longer term (all else constant). It is very difficult to detect this against a background where other costs are changing by much larger amounts and the CPI is increasing by around 2½ per cent per year on average.
Despite the difficulties of measurement, the Board’s judgement remains that the bulk of these savings have been, or will eventually be, passed through into savings to consumers. This judgement is consistent with standard economic analysis which suggests that, ultimately, changes in business costs are reflected in the prices that businesses charge. A similar conclusion
was reached by the House of Representatives Standing Committee on Economics, Finance and Public Administration when it considered the Bank’s payments system reforms in 2006.
Note what the RBA’s facts actually support–that merchants saved $1.1 billion in costs. But the RBA has no evidence of any pass-through, how much pass-through occurred, or–most importantly–whether the pass-through to retail customers was larger than the pass-through to card customers prior to interchange regulation. And, of course, none of this even accounts for the external benefits of card usage versus paper, such as reduced crime, tax evasion, and the social costs of printing cash. The RBA’s “conclusion” is not so much a “conclusion” as a bald assertion without a single shred of actual evidence to support it. It could be true, but the RBA certainly hasn’t demonstrated it.
As to the RBA’s assertion that it is “confident” that the “bulk” of these savings will be passed through, the question is much, much more difficult than the RBA acknowledges. The degree of pass-through in the retail market versus the card market depends mightily on the elasticities of supply and demand in those markets. On this, there is good reason to doubt the unsupported assertion that the “bulk” of savings will be passed through. Certainly, there is no question that the amount of pass-through will vary significantly among different markets.
For example, when the RBA permitted surcharging of cards, Qantas announced that is was going to surcharge card purchases. First, as I note in my white paper, the surcharge it imposed far exceeded the actual cost of cards. Second, it announced that it would not reduce its prices for cash users. Now Qantas is unusual in that it likely faces a relatively inelastic demand curve for use of cards for various reasons. But it illustrates the point–the amount of pass-through varies a lot, and there is no reason to believe that there was any but minimal pass-through to Qantas customers. Qantas price cards according to inelasticity of demand, not just its costs. Every other business does too.
There are other reasons to believe that pass-through might be less than expected. For example, today merchants have an absolute right under the Truth in Lending Act to offer cash discounts. Very few do. Many merchants also run their own in-house proprietary credit systems. By definition, those credit customers cost more to service than cash customers. Yet merchants charge the same price for cash and their own credit–and in fact often discount for their own in-house credit (such as when department store offers a discount for opening a store credit account). Pure economic theory would predict that if it were the case that all that mattered in a competitive market was cost, in those markets you would expect widespread cash discounting for both bank-type cards and propriety cards. But you don’t. Which suggests that pass-through is modest.
By contrast, competition in the credit card market is fierce. There is no evidence that issuers are earning rents off of interchange fees. There is evidence, however, that issuers compete aggressively for card holders (a key point here).
The RBA may be correct that merchants saved $1.1 billion in cost. There is no evidence at all however that the “bulk” of that has been or will be passed on or that the pass-through will more than offset the higher costs to cardholders. And again, that excludes the social benefits of greater card use.
“Visa also inaccurately claims that swipe fee reforms are only supported by large retailers. In fact, there is very broad-based support for swipe fee reform.” I don’t speak for Visa, so I don’t know what they supposedly said. But I doubt they would say that “only” large retailers support interchange price controls. Nor would it be accurate for them to say that–certainly retailers of all sizes support politically-determined interchange price controls.
It is accurate, however, to say that the primary beneficiaries of the Durbin Amendment are large retailers. In some part this is from the price control provision itself if we assume that the average purchase at larger retailers tends to be larger than at smaller retailers because of the way fees are charged. it would also be the case if a larger percentage of the transactions at large retailers are made with cards (which seems plausible, but I’ve seen no evidence).
The real smoking gun, however, is the provision in the Durbin Amendment that permits retailers to establish maximum charge levels for using cards. The only possible explanation I can see for that provision is to enable department stores and other big retailers who can afford to establish or maintain their own proprietary credit operation to essentially steer consumers to their own in-house credit for big-ticket purchases. Small retailers can’t afford this. And I can’t see any possible way that this provision eliminating competition in the provision of credit for big-ticket purchases could possibly help consumers.
I’ve not seen any comment from Adam as to whether he supports this provision of the Durbin Amendment.
On the other hand, it seems quite plausible that smaller merchants are the primary beneficiaries of the provision that permits merchants to set minimum purchase levels for card usage. Basically the importance of this provision is that it allows merchants take advantage of tourists, business travelers, and others who don’t have enough cash in their pocket or time to go to an ATM, by making them buy more stuff before they can use a card. Often this means that the consumer might have to choose between finding an ATM and paying an out-of-network ATM transaction fee (I note that many convenience stores helpfully provide ATM’s in store from which the store owner draws a cut!) or buying more stuff to get the bill up to the minimum purchase level. As I’ve said previously, I can see why this is a good deal for the merchant, but I don’t know why the rest of us would favor this.
So it is likely true that small merchants are rent-seeking at the expense of consumers just as larger merchants are. They both benefit, which is why both groups have been willing to spend millions of dollars rent-seeking, but it nonetheless looks like larger merchants probably benefit more.
So what’s really going on? I think the politics of this is pretty clear. Politicians have many retailers in their districts. They have relatively few banks and the major card issuers are located in a handful of districts (which is why politically the carve-out for smaller banks makes political sense even if makes not economic sense). Merchants have spent millions lobbying on this issue–and they haven’t spent that much just so they can pass on their gains penny-for-penny to consumers. Issuers, of course, have lobbied on the other side. But based on available evidence, banks are making a normal return on these operations, not any sustainable economic rents. Consumers, as always, are the least represented players in the political process, but we’ll be the ones who bear the cost of higher costs for debit cards, and eventually credit cards.