Regulating Interchange Fees will Promote Term Repricing that will be Harmful to Consumers and Competition

Todd Zywicki —  8 December 2009

Todd J. Zywicki is Foundation Professor of Law at George Mason University School of Law.

Although the mechanisms vary, legislation pending before Congress on interchange has a basic central purpose—to reduce interchange fees, either indirectly or directly.  If adopted, these efforts will likely succeed in their intended goal of reducing interchange fees.  But they will also likely have substantial unintended consequences that will prove harmful to consumers and competition and will roll-back the innovation in the credit card market over the past two decades.

Credit cards produce three basic revenue streams for issuers: finance charges (interest on revolving credit), merchant fees, and other fees on cardholders.  The ratio among these three streams has remained largely constant for almost two decades.  About 70 percent of issuer revenues come from interest paid by cardholders on revolving balances.  About 20 percent of revenues are generated by interchange fees.  And about 10 percent come from various other fees assessed on cardholders: two decades ago the bulk of this revenue was generated by annual fees, today it is predominantly from behavior-based fees such as over-the-limit fees, late fees, and other similar charges tied to a borrower’s actual behavior.  The precise ratios among these three streams varies slightly over time: in recent years, for example, greater use of home equity loans as a source of consumer credit led to a reduction in revolving balances and interest payments and an increase in risk-based fees.  At the same time, transactional use of credit cards has risen rapidly, fueled primarily by consumer use of rewards cards, leading to a growth in interchange fee revenue.

So what would happen if retailers get their way and interchange fees were cut by artificial governmental intervention?  The mathematics of the situation is inescapable: card issuers would have to increase the revenue generated from consumers from either interest payments or higher penalty fees or reduce the quality of credit cards, such as by reducing customer support or ancillary card benefits. In fact, this is exactly what happened when Australian regulators imposed price caps on interchange fees in 2003: annual fees increased an average of 22% on standard credit cards and annual fees for rewards cards increased by 47%-77%, costing consumers hundreds of millions of dollars in higher annual fees.  Card issuers also reduced the generosity of their reward programs by 23 percent.

The credit card system is essentially a closed system: a forced reduction in one stream of revenues generates efforts to substitute other revenue streams.  In a competitive market, the losses from one revenue stream have to be made up for somewhere else.  Americans have been recently reminded of this lesson, as Congress’s imposition of new limits on certain terms of credit card pricing through the Credit CARD Act over the summer has led to increased interest rates and higher annual fees to offset those restrictions.  Because it is more difficult to price risk accurately, issuers have reduced their risk exposure by reducing credit lines and closing accounts.  Congress may wail because its legislation failed to repeal the laws of supply and demand, but just as a minimum wage law increases unemployment or rent control creates housing shortages, regulation of some credit terms leads to predictable substitutions for unregulated terms.  Direct or indirect price caps on interchange fees would have similar negative consequences.

But while this type of substitution is bad enough for consumers there is an even more important systemic problem. The most important pro-consumer innovation in payment systems of the past two decades has been the general disappearance of annual fees on credit cards (except for rewards cards where the annual fee defrays the cost of program administration).  The elimination of annual fees has made it possible for consumers to carry and use multiple cards simultaneously.  According to Experian, consumers today have over five credit cards (including retail accounts) on average and over half the population has two credit cards or more. The consequences for consumer choice and competition have been profound—card issuers compete for consumers’ business literally every time they open their wallet to make a purchase.  Consumers can and do easily shift balances among different cards depending on which provides the best deal at any given time (according to a survey by ComScore, two-thirds of consumers say that they would consider switching their primary credit card if a better feature were offered).  Consumers can also stack credit lines when necessary.

An annual fee is essentially a tax on holding cards.  Policies that produced a return of annual fees would strangle this process of competition by making it more expensive for consumers to hold multiple cards, increasing switch costs and dampening competition.

Access to multiple cards (and their credit lines) is particularly important for the three-quarters of independent small businesses that rely on personal credit cards in their business and count on infrequently-used reserve lines of credit to exploit rapidly-developing business opportunities.  These reserve lines are especially important today as credit lines have been slashed. Forcing small businesses to pay an annual fee just to maintain access to these reserve credit lines would deter many of them from doing so, stifling entrepreneurship and an economic recovery.

Perverse consequences would likely follow on the issuers’ side of the market as well.  Issuers will find it more burdensome to retain customers who pay their bill every month (the lowest-risk group of customers), creating incentives to pursue customers who revolve and the riskiest class of customers who pay behavior-based fees.  This substitution will lead to a less-diversified revenue stream (more dependent on non-interchange revenue), making credit card operations riskier and more dependent on the swings in the business cycle.  Credit unions and community banks rely especially heavily on interchange revenue as they tend to cater to lower-risk customers that are less prone to revolve balances and pay penalty fees.  Reducing revenue from interchange fees would force these issuers to either abandon the credit card market or to pursue a riskier customer base.  It is hard to see why Congress would want to adopt policies that punish the most conservative financial institutions, encourage the very risk-seeking behavior that helped to spawn the financial crisis, and encourage a less-diversified and more risky customer base.  Yet squeezing interchange fees would do exactly that.

There is no free lunch.  In a competitive market, a reduction of the stream of revenues from interchange fees will have to be made up somewhere else.  Not only is this term repricing likely to be inefficient by replacing voluntary contract terms with governmentally-created prices but they will likely dampen competition and innovation, thereby harming consumers in the long run.

7 responses to Regulating Interchange Fees will Promote Term Repricing that will be Harmful to Consumers and Competition


    Like Todd, I’ll do some discussing of the cross-subsidy argument tomorrow. But I think one thing worth noting here is that complaints about cross-subsidies are fundamentally at odds with the conventional posture of the interchange fee debate as a competition policy issue. In other words, the existence of cross-subsidies is a function of the very nature of competition in two-sided markets rather than market power. For example, payment systems with small market shares must engage in the same balancing of both sides of the market which results in cross-subsidies. Further, as Todd alludes to, consumers are “bombarded” by cross-subsidies every time they enter the supermarket, coffee shop, bookstore, etc.

    It is important to recognize that the competitive process generates these sorts of pricing “inefficiencies,” which are inefficient primarily in the sense of comparison to a blackboard model of perfect competition, are often the consequence of the normal competitive process. It is in this light that the posturing of interchange fee legislation as antitrust intervention appears troublesome.


    I think that whether this inflates prices is a matter of relevant cross-elasticities of demand and supply in the market. Empirical evidence so far finds no evidence that the reduction in interchange is passed on to consumers in the form of lower prices. So, this may mean that the merchants increase quality (perhaps better customer service, longer hours, or free gift-wrapping), higher profits, or simply dissipate the rents in some other way (perhaps excessive advertising to try to attract customers).

    And it is open to question whether the alternative payment schemes are actually less-expensive (once the full social and externality costs are accounted for) and that they are also subsidized (the government prints currency and clears checks at par).

    So it is equally plausible to assert that consumers overuse cash and checks because they don’t pay their full price.

    As I’ll discuss tomorrow, the idea that credit cards makes non-card customers subsidize cash customers is a red herring–those sorts of cross-subsidies have been going on for about a century.


    Thanks for your comment. This is a tentative response, because I’d want to re-read Josh’s paper more closely and read more thoroughly on the academic discussion of it, but here’s my initial response.

    I certainly understand Josh’s findings, but they remain intuitively puzzling to me. Mandell reports that when annual fees were first imposed on credit cards in the US in 1980, some 8% of the outstanding credit cards in the market were canceled, or about 9 million cards. If any of my card issuers try to impose an annual fee on me, then I’d cancel that card. I can’t imagine I’m unique in this. Which should lead to attenuated competition.

    So while I understand Josh’s findings, it is hard for me to wrap my head around the logic behind the findings. As Geoff notes in his Comment to Josh’s post, these sure seem like some awfully inelastic consumers. And it suggests that consumers are much more inelastic about credit cards than they seemed to be in 1980.

    And it isn’t clear to me why that would be. The only possibility I can think of would be that at that time many retailers offered layaway, store credit, and their store credit cards, so maybe consumers who were making credit purchases didn’t need bank cards as much. And usury regulations made bank cards much less attractive than these other types of credit. But it hardly seems like we want to say that the goal of interchange regulation would be to resuscitate layaway plans.

    Ditto for surcharging (which I gather he’ll discuss tomorrow). I use my credit card for everything, but I’d probably switch to a debit card if my credit card was surcharged. Obviously that’s not data, but I can’t imagine that I’m unique in that.

    So if there is an intuitive logic to back this up, I’d be interested in understanding it (or perhaps Josh is going to explain it tomorrow).

    But there is another issue–if there is no welfare effect because there is no change in consumer behavior or the overall economic effects, what is the point of the regulation? Just to change the distribution between the merchants and consumers? I thought the point was to try to change consumer behavior to substitute away from credit cards to other payment schemes.

    If I’m understanding the bottom line correctly, what Josh seems to show is sort of curious. Merchants argue that they “must” accept cards because of business necessity–this gives rise to the idea that merchants are the more inelastic side of the market. But Josh seems to be finding that consumers are highly inelastic in their demand for credit cards too. That could be, but this seems like it is starting to look like a pretty strange market. As suggested above, this may be the case with goods almost exclusively bought on credit (like appliances), but it doesn’t seem generalizable.

    And, of course, this doesn’t even consider all the welfare benefits of electronic payments to which you allude.

    So this is not to question Josh’s empirical findings, but to just say that they seem incongruous to me and that this seems to just be a distributional issue with pretty weak policy implications.

    On behavior-based fees, the only research that I’m familiar with (unless I’ve missed something) is Massoud, Saunders, and Scholnick,, which finds that behavior-based fees appear to be consistent with risk-based pricing. If that is true, then presumably that improves overall efficiency.

    As for small business, you are right that I am not making a broad welfare claim. I don’t think that it has been studied thoroughly. I’m making the casual observation that small businesses may have an even greater need for residual credit lines to deal with greater short-term fluctuations in credit needs, so that having a no-annual fee card that can serve as a back-up credit line when needed may be valuable, but would be more costly if the small business has to pay an annual fee to keep it open.

    Omri Ben-Shahar 8 December 2009 at 1:49 pm

    Even if you are correct in your conjecture that issuers will make up for lost revenue resulting from interchange fee caps, it is not clear that this substitution is bad. In fact, the substitution can be efficient. If interchange fees inflate product prices, which is probably the case, and if these products can be more efficiently paid for without incurring up to 5% payment fee, then we are dealing with a potentially substantial distortion in the decisions to buy products. It is possible that this fee helps pay for the credit element in the transaction, but for those who don’t need this element there is not reason to tie it to the product. So, I don’t think that the substitution argument necessarily leads to the conclusion you draw.

    Bob Chakravorti 8 December 2009 at 12:40 pm

    I agree that increases in interchange fee would result in increases in other types of fees and charges. Would this shift in fees have any welfare implications? According the Josh’s post, there were no real effects to the economy. Would you expect something different for the US market? Do we have any data regarding the benefits of switching from annual fees to “behavioral” fees and its impact on the real economy?

    It is difficult to predict what are the optimal levels for each fee. Clearly, there are benefits to merchants in accepting payment cards and they should share in the cost to enjoy these benefits. One measure that was mentioned elsewhere in this symposium is the cost of directly providing credit and in some cases subsidizing that credit to increase sales and gain loyalty. Other issues include possible better fraud protection and timely receipt of good funds. However, the optimal level of the interchange fee from a social perspective for a specific market has not been identified.

    An interesting shift in the US is greater use of debit cards and linkages to lending facilities. These may be worse for the consumer in the sense that there is no 21-day float versus paying a credit card monthly and may result in additional service fees but merchants may be better off because of lower fees.

    I am less convinced that interchange fees should be kept high so that annual fees are zero to aid small business borrowing. Interestingly, “business” credit cards have higher interchange fees that can be availed by small business owners. Furthermore, it is not clear who is subsidizing whom here and whether society is better off for it.

Trackbacks and Pingbacks:

  1. TRUTH ON THE MARKET » Zywicki on Interchange Fee Legislation - January 5, 2010

    […] colleague (and TOTM Credit Card Symposium participant — posts here and here) Todd Zywicki has an excellent op-ed in the Wall Street Journal today on Congressional […]

  2. TRUTH ON THE MARKET » The Merchants’ Insincere Concern About Cross-Consumer Subsidies - December 9, 2009

    […] Josh on Regulating Interchange Fees will Promote Term Repricing that will be Harmful to Consumers an…Josh on The Myth of Consumer Protection Through Disclosure.Geoffrey Manne on What happened in Australia?.Richard Epstein on Why Now? The Faulty Economics of Credit Card Reform.Omri Ben-Shahar on The Myth of Consumer Protection Through Disclosure.Joshua Gans on What happened in Australia?.Peter Schwartz on EPA's Legislative End-run Strategy.Allan Shampine on The Myth of Consumer Protection Through Disclosure.Bob Chakravorti on What happened in Australia?.Todd Zywicki on Regulating Interchange Fees will Promote Term Repricing that will be Harmful to Consumers and Competition. […]