The Benefits of Accepting Credit Cards

In a previous post I wrote about the costs of accepting credit cards. It is important to note that merchants derive lots of benefits from card acceptance, including and lower credit risk (at least compared to checks), easier bookkeeping and currency conversion, lower theft risk (less cash in the till), and improved checkout speed. The most important benefit, though, is increased sales. (See an example of this claim here.) But is that right? Do consumers really purchase more because they are purchasing with credit cards? Put more technically, does price elasticity depend on payment method?

Consider what MasterCard and Visa state on their websites. Visa lists the following among the benefits of accepting its cards:

• There is an increase in average ticket size for card purchases
• Customers tend to purchase higher margin goods with cards
• Card acceptance increases your revenue and reduces losses

Likewise, MasterCard’s website lists "Increase Sales" as a benefit of accepting cards: “Consumers spend more when they're not constrained by cash on hand. You may see increased purchases of higher-margin products as well as specialty items. And customers may visit your store more often.”

Let’s unpack these statements. Does card acceptance really increase revenue? I have seen no published study to this extent. Instead, there’s a logic error occurring in statements attributing revenue growth to card acceptance. No one disputes that ticket sizes for credit card purchases are higher than those for cash (but less than checks). But this correlation hardly shows a causation, much less a direction for that causation. Do people spend more because they purchase with credit cards? Or are credit cards simply the payment instrument of choice for larger purchases (because of legal protections, convenience, security, etc.)? There’s surprisingly little evidence on this question. A few, rather studies indicate that price elasticity might vary by payment method, but only for luxury goods, but none of these studies look at data from a number of merchants before and after credit card acceptance. Instead, they use artificially created experiments.

Logic suggests that in most cases, larger ticket amounts for credit card purchases appear to be a function of a preference to use cards for large purchases. No one (in the US) wants to carry around thousands of dollars in cash (Japan is another story). And it's a lot easier to dispute a purchase made on a credit card than one made with cash. Consumers are more likely to be concerned with disputes over large ticket items. And so on. It seems likely that the correlation of large purchases and credit cards is explained by a causation flowing from purchase to cards, not cards to purchase. Instead of card use causing large purchases, large purchases cause card use.

Such a limited statement makes comports with common sense. Common sense tells us that cash consumers are limited to the cash presently in their pocket, debit consumers are limited to the cash presently in their bank account, check consumers are limited to the cash in their bank account in a couple of days, and credit card consumers are limited to their credit limit. Which of these will be the highest is uncertain, but credit limits seem like a reasonable guess. So, arguably credit card consumers are not as constrained in their immediate purchasing power as other consumers (although they will find themselves more constrained down the road if they do not pay their bill in full and on time). But, how many consumers will buy more milk (or buy it at a higher price) because they are paying by credit card? A consumer might splurge on luxury goods because of credit cards (strange to think that these are exactly the type of purchases that are non-dischargeable in bankruptcy), but is unlikely to purchase more staples because of credit cards. And it is only the cash-flow constrained consumer who will really purchase more because of credit cards even in these circumstances.

Of course rewards points tilt the scale slightly towards greater purchases. If I get 1% cash back, I might rationally purchase $101.01 with a credit card rather than $100 worth of goods with cash. But given that merchants pay well over $1.01 for the rewards card purchase, the rewards are actually a net loss for the merchant, not a benefit. If card acceptance merely shifts larger ticket purchases to cards (and to the most expensive cards), rather than generating larger ticket purchases, it is hardly a benefit for merchants.

To be sure, there are cases where card acceptance undoubtedly increases revenue. It’s hard to sell anything over the Internet without accepting cards. Start accepting cards as an Internet merchant, and its hard not to see a revenue increase because prior revenue was likely zero. And if a business’s competitors start accepting cards, then jumping on the card acceptance bandwagon might increase revenues, or at least offset some of the loss to competitors.

So does card acceptance really benefit merchants from increased sales? The jury is still out. But if anyone wants to have a better answer, here’s my bleg: I’m looking for a data set. If your company has recently started accepting cards and is willing to contribute its data to a study (with reasonable confidentiality) please contact me.

Comments

[Chuchundra]

I don't have any data to give you, but you might want to inquire about what happened when supermarkets started taking credit cards.

I pay with a credit card for just about everything these days. Any business that doesn't take credit cards might have a hard time getting any money at all from me. I suspect that there are a lot of people who are similar.

[Adam Levitin]

Chuchundra, would you still pay with a credit card if it cost you an additional 3% at point of sale? (That'd be $720 annually on a purchase base of $24,000.) What if there were an annual fee of $500? I'm curious how much the convenience, security, legal protections, rewards, etc. are worth to you (and other readers).

[Bob Lawless]

My instincts on this are directly the opposite of yours, Adam. Because it's obviously an empirical question in the end, I admire your call for data. Still, I think the instinct here is that a payment in cash--or maybe even check requires the consumer--to feel the sting of the loss of money in a way that payment by credit card does not. Any money substitute works that way. Gambling chips, in addition to being more convenient, also would work in the same way.

[Chuchundra]

If there was a 3% surcharge on credit card purchases, I'd certainly make fewer of them. I'd use cash, checks and other types of electronic payment whenever possible, saving the credit card for online transactions, emergencies and other times when it was unavoidable.

On the other hand, right now I actually get 1% back for all my credit card purchases. I use my card for just about everything from grocery shopping to bill paying. In fact, I got married this year and put a good portion of the cost of the reception, ceremony and honeymoon on the card. The Rabbi, sadly, would only take cash.

Of course I'm a credit card "dead beat". I pay off the card in full every month and I haven't paid a penny of credit card interest in years.

[Adam Levitin]

Bob, I think there is an endowment effect associated with cash that there isn't with credit cards. But I suspect there is one with checks--I know that when I have to write out the amount of a large check, I feel the pinch in a way I don't making the same purchase on a card. The gambling chips are an interesting comparison.

Chuchundra makes my point--consumers would avoid credit cards if they realized their costs and had to internalize them. But for now Chuchundra is doing exactly what a savvy consumer should do in the current situation: try to get a card with the most generous rewards program and use it without revolving a balance.

But consider this. Only 45% of the interchange fee (which is the majority of the fee the merchant pays its bank) goes to fund rewards programs. So you've really paid the other 55% yourself. If the merchant pays 2.22% and you get 1% back, you've paid an extra 1.22% on the transaction. You'd do better if you paid cash (and the merchant didn't take credit cards). For small transactions, you might say "who cares?" but recall the fractional penny rounding scam in Office Space--pennies add up.

[Bob Lawless]

Adam, I was trying to distinguish the positive from the normative claim. I'm not sure I disagree with you about the benefits of making consumers feel the sting of the interchange fee as you discussed in a different post. As a positivist claim, my strong instinct is that people do spend more with credit cards than with cash.

[bob]

I know you asked for empirical evidence, and I don't have it. But it is interesting to think a bit about the logic. Suppose there's only one consumer and one merchant. If using a card induces a consumer to spend more today, it must mean the consumer will have to spend less tomorrow. In this example, the merchant is simply "borrowing" sales from the future.

Now suppose there are many merchants, however. Now it is possible that the merchant is borrowing sales that OTHER merchants would have made in the future. And of course if all merchants behave this way, it is at best a zero sum game for the merchant. That is the intuition of one of Bob Chakravorti's important papers.

One other way to think about this is that having access to a credit card may lead consumers to reduce their precautionary savings (here savings and cards are substitutes). If that is the case, there is a one time increase in sales enjoyed by merchants, but thereafter we return to the tradeoff between consumption today and tomorrow.

[F. Javier Arias Varona]

I have no data to share eihter, but I'd lke to point out that me and the people around me spend more with credit card. From my point of view it is true what Bob said: you feel much more the money you are spending when using cash. So in mid-priced items (let's think in a PS2 game or a nice shirt), where the "meditation" time (I hope you understand what I mean...) is obviously shorter than in more expensive items, I would buy them with the credit card if I see them when having a walk on the street, but would probably think again with cash, considering that I carry enough with me at that time. Would it be better to think in different terms for basic items (you will not buy more milk), luxury items (you will probably use credit card as the best payment method, your decision is previously made) and a whole bunch of intermediate things where probably the credit card makes it easier to spend and so pushes the sells up? If it were the case, which kind of merchant will have a stronger positive impact of the use of credit card? I should clear that I am thinking in terms of NO revolving (I have seen myself in the words of Chuchundra, actually...)

[Rob Rickner]

I think there is a simple answer. If your business doesn't accept credit cards, then consumers who aren't carrying cash are going to go to a different store. Worse for the merchant - it seems the customers they lose are making larger purchases as well.

[Adam Levitin]

I think that's right Rick--given the current constellation of credit card network rules, a merchant really has no choice about accepting cards if a competitor accepts cards. Either everyone accepts them or no one does. Sort of a prisoner's dilemma.

I agree entirely with Bob-merchants are "borrowing" against future sales, and as Bob Chakravorti of the Chicago Fed has noted once all merchants (or nearly all) are doing this, the effects of cross-merchant "borrowing" cancel out.

I know that classical economics sees saving as simply an alternative consumption choice to spending, but societally we should, perhaps, value them differently--given that people tend to be poor estimators of future expenses and risks, we should want people to have some safety cushion, and payment devices that encourage (non-necessary) spending work against that.

[bwe]

I am a retailer of men's and women's classic clothing, higher end, older customer, with a current average sale around $175. I have taken credit cards from the day we opened 31 years ago so I can't give you the comparison you're looking for. Nonetheless, I have lots of data and here are some trends I've noted since 1976 from my perspective:

1. Cash and check sales have steadily diminished as a percentage of total sales.
2. MasterCard led both Visa and American Express in total sales until about 25 years ago when Visa took over. MC is in second place, Amex in third.
3. Visa has steadily increased in percentage of total sales over the 31 years.
4. American Express has consistently had the highest average sale since day one with VISA closing in. Amex used to have an average sale over 40% higher than Visa and MC. That has steadily eroded to less than 5% higher than Visa.
5. From the beginning, I believed accepting Amex cards was a competitive advantage. Far fewer merchants accepted it, particularly restaurants, because of the higher charges to merchants and longer payment time as compared to Visa/MC. At the same time, Amex was the "prestigious" card, signifying to Amex card holders that those merchants accepting it were legitimate, high quality, and long lasting. That still exists with older customers; younger customers like to use it for the airline miles. Customers were and are very pleased that we accept Amex.
6. In-house charge account customers have steadily migrated to credit cards.
7. Discover Card is hardly ever used.

[Martha Sherwood]

Another factor which many merchants take into account is the percentage of the purchase which represents markup. When I go into the supermarket I have to go past the deli, the wine racks and upscale packaged food to get to the milk and eggs and flour - then the generic brand is either on the bottom or the top shelf. Are credit card shoppers more suceptible to this sort of marketing than cash shoppers like myself? If they are, the electronic records from today's high tech checkouts would confirm it.

[Brian]

I have long suspected that the expanding use of credit cards has increased the price elasticity of consumer products. In my view, credit cards have decoupled the link between the amount of cash in pocket and the amount of the purchase, much the way the use of casino chips most likely increase the amount of money a gambler will risk. (I would assume this would be the case since casinos seem to spend a great deal of money on the chips themselves).
I am always intrigued by reports that consumer spending seems to have slowed very little despite historically high increases in prices of consumer goods. I would suspect that these consumers are increasing their debt levels, although that would most likely be a good bet even if inflation was relatively low.

[Jane Carter]

We are wondering if it makes sense to accept American Express at our Prague hotel. The rate is 4.5% as opposed to 2.5% for Visa and MC. How many people are likely to stay at our hotel rather than another Prague hotel if they can't pay with Amex?