The still-pending $7.25 billion settlement of the long-running antitrust case against Visa, MasterCard and more than a dozen of their member banks would, among other things, allow merchants to charge their customers fees for using credit cards for payment at the checkout. Although much has been written about this particular provision of the proposed agreement, it is unlikely to have any significant effect on consumer payment behavior, not least because even if the settlement is approved, such surcharges are already banned in 10 of the country’s most populous states and many others are planning to do the same.
However, as I’ve noted on this blog many times before, merchants have, and always have had, access to other tools to help them drive — or “steer”, in industry parlance — customers away from credit card use and toward more merchant-friendly payment types like cash and, since the Durbin Amendment took effect, debit cards. My favorite example is the discount that has for a very long time (ever since I can remember) been offered by many gas stations here in Massachusetts to cash-paying drivers. And yet, such incentives have been used very sparingly. Why? Well, anecdotal evidence has always suggested to me that the benefits from implementing such measures would be outweighed by the disadvantages, primarily in the shape of consumer backlash. After all, if you want to pay by credit card, for example because you get a 3-percent cash back for doing so, you are unlikely to frequent a gas station, which charges you extra for it. But it turns out that there is more to it than that. A new paper from Tamas Briglevics and Oz Shy from the Boston Fed strongly suggests that offering a cash and, especially, a debit card discount may not even make financial sense for the merchants themselves. Let’s see why.
Calculating the Cost of “Steering”
The Boston Fed researchers offer the following rough calculations:
Consider a merchant selling to consumers who spend on average $30 for each transaction. Suppose that the merchant fee on credit card transactions is 2 percent (hence, 60?ó per average transaction). Suppose that the merchant fee on debit cards is a flat 25?ó per transaction. Note that the debit card fee does not vary with the value of the transaction. Footnote 10 explains why 25?ó is a realistic estimate of a debit card merchant fee.
Conjecture 1. Because 60?ó > 25?ó, profit is always enhanced when merchants steer the average customer to pay with debit cards instead of paying with credit cards.
Whereas Conjecture 1 makes sense, it seems unrealistic that buyers who pay with credit cards would agree to pay with a debit card without receiving any monetary incentive. Therefore, the merchant must provide customers who pay with credit cards an incentive to pay with their debit cards. A frequently observed method of steering (among the few merchants who steer) is to offer a 1-percent discount to customers who pay with a debit card.5
Suppose that the 1-percent debit card discount is a sufficient incentive to induce all customers who pay with credit cards to switch to paying with debit cards. Then, the merchant saves 60?ó, which is the credit card merchant fee on a $30 transaction. On the other hand, each consumer who switches from paying with credit to paying with debit increases the merchant cost by 25?ó per transaction. In addition, the merchant loses 1 percent from the debit card price, which equals 0.01 x $30 = 30?ó. Hence, each switching buyer adds 25?ó + 30?ó = 55?ó to the merchant’s cost of accepting debit cards.
Conjecture 2. Because 60?ó > 55?ó, profit is always enhanced when a merchant gives the average buyer a 1-percent discount on paying with a debit card.
Our investigation in the remainder of the paper is motivated by the following result:
Result 1. Conjecture 2 is incorrect.
Intuitively, Conjecture 2 is incorrect because it neglects to take into consideration that the 1-percent debit card price discount also applies to buyers who do not pay with credit cards even in the absence of any effort exerted by merchants to steer customers to pay with debit cards. The revenue loss from these buyers may outweigh the gains from buyers who switch from paying with credit to paying with debit.
The researchers then proceed to prove their result and do so, although they are very careful in their wording, but I think that the above reasoning should be sufficient to convince doubters.
I should note that the 2-percent credit card transaction fee used in the researchers’ calculations lies toward the very high end of the spectrum of probabilities. Unless you use Square or something similar for payment processing, your per-transaction credit card fees would be lower than that. In any case, the credit card processing fees paid by bigger merchants are much lower than 2 percent, which means that offering a debit card discount would make even less sense for them.
Now, Briglevics and Shy are careful to point out that there are differences between the effects of surcharging and discounting. Whereas the use of card surcharges allows merchants to vary the size of the fees to account for the differences in the processing fees they pay on different types of cards, a discount is applied universally and does not vary, which gives merchants less pricing flexibility. But the more important point is that neither approach would bring much benefit to merchants — surcharges would alienate credit card using customers, while discounts may turn out to be too costly to make sense.
Image credit: Flickr / 401(K) 2013.