Do Credit Cards Actually Increase Spending?

There are surprisingly few studies that explicitly examine the impact of credit cards on spending. Several studies have noted a correlation of higher spending with credit cards, but there are several possibilities that can explain the correlation.

For example, credit card users as a whole may be more affluent, which might account for higher spending when compared to cash users. And consumers may segment their purchasing behavior, paying for smaller ticket items with cash simply because they have enough cash in their pocket to cover the cost of the smaller purchase, but will pay for more expensive purchases on a card out of necessity or convenience.

Additionally, cash buyers may simply spend less due to liquidity constraints: buying fewer items or less expensive items simply because they do not have enough cash on them at the moment. All that goes to say that the correlation observed in prior studies between higher spending and credit cards doesn't necessarily point to credit cards causing that change.

I stumbled across a Carnegie Mellon paper that attempted to specifically examine the impact of credit cards on spending, and that controlled for many of the pitfalls from earlier studies. Two economics professors, Elif Incekara-Hafalir and George Loewenstein, ran an experiment in a cafeteria over several trials to test the impact that credit cards had on spending.

Before entering the cafeteria, consumers in the treatment group were offered a choice of two different coupons: one offered an $8 Amazon gift card if paying with credit, and a $5 Amazon gift card if paying with cash. (The control group was not offered a choice.) To redeem the coupon, the consumers had to provide the receipt and fill out a survey.

The study was unique in several ways: for one, it was a randomized experiment being run in a real life consumer situation. And liquidity concerns were mitigated by surveying participants' financial situation, and also because the purchases were so small. It was unlikely that cash buyers were paying less for lunch simply because they had insufficient funds available, and that could be teased out from one of the survey questions: ("About how much cash do you have with you right now?").

The findings from the experiment are surprising and run contrary to the belief that credit cards induce higher spending. For the short version, here is the abstract:

In a field experiment, we measure the impact of payment with credit card as compared with cash on insurance company employees’ spending on lunch in a cafeteria. We exogenously changed some diners’ payment medium from cash to a credit card by giving them an incentive to pay with a credit card. Surprisingly, we find that credit cards do not increase spending. However, the use of credit cards has a differential impact on spending for revolvers (who carry debt) and convenience users (who do not): Revolvers spend less when induced to spend with a credit card, whereas convenience users display the opposite pattern. 

For one, I was somewhat shocked to see that credit cards didn't increase the overall spending when compared to cash buyers. Most of what I've read from personal finance blogs indicates that plastic is likely to change purchasing behavior, and for the worse. Older studies point to the fact that simply seeing a credit card logo might increase the figure a customer would be willing to pay for an item by several hundred percent.

But as someone who does nearly all his purchases on a credit card, it was a relief to see this result. I might actually be pulling one over on the credit card companies, and getting free money via rewards without increasing my baseline spend. Go me!

But within the overall "equal" spending of credit card users is a strange anomaly. Again, quoting from the paper:

The numbers just reported hide an interesting pattern, a difference between participants who report carrying a credit balance (“revolvers”) and those who do not (“convenience users”). Figure 1 illustrates the nature of the difference; it suggests that the treatment, which encouraged diners to pay with credit cards, increased the spending of convenience users, but decreased the spending of revolvers.

What this means is that the study found that credit card users who actually pay their credit cards off each month spent more: 15% more! Even more curious: those credit card users who carried a balance actually paid less.

It's completely counter-intuitive. The consumers that we'd typically think of as responsible users of credit were impacted negatively in their spending by using credit. (The authors cite another study as well, finding that convenience users, those who carry no balance, are more sensitive to cash back offers -- this points to the Amazon gift card potentially inducing these consumers to spend more.) And those credit card users we'd typically think of as irresponsible, those who keep a balance on their card, spent less than cash users. (The authors posit that "revolvers" might be spending less due to previous, negative experiences with credit.)

Caveats
The authors caution that broad conclusions shouldn't be drawn from this study, since it is one of the first randomized experiments on the impact of credit cards on spending. They cited the possible effect of the coupon: that credit card users might have felt manipulated by the offer of the Amazon gift card and might have been especially careful not to spend more as a result.

Additionally, the timing of the experiment in 2008 and 2009 may have impacted the results as it took place in the middle of the recession, with many consumers having a negative view of consumer debt. So there are reasons to hold off on making broad conclusions from this study, and for waiting for additional research.

But if you can't jump to hasty and unreasonable conclusions on an anonymous personal finance blog, then where can you? As someone who pays off his credit cards each month (which I imagine is typical of the readers of this blog), is it possible that my behavior mirrors that of the "convenience users" in the study, and that I may be especially prone to spend more with a credit card?

Or is the general finding of the study more applicable? That credit cards don't actually induce higher spending than cash, so we should confidently move forward, churning rewards cards and reaping the benefits.

What say you, readers? What do you take away from this experiment?

Photo is from shawnzrossi at Flickr Creative Commons.