You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

Lab-Conditions Tax Avoidance

Is the American public's distaste for taxation irrational? A 2011 paper by Abigail B. Sussman, a graduate student in psychology at Princeton, and Christopher Y. Olivola, a fellow in behavioral science at the University of Warwick, suggests that it is. The paper, published last year in the Journal of Marketing Research (and flagged today by the Russell Sage Foundation's Twitter feed) found that between two ways to save money--(legal) tax avoidance or straight-up discount--more people will choose the former than the latter, even when the latter saves an equal or even greater amount of money. There's just something about not paying taxes.

Sussman and Olivola detail this phenomenon by describing three experiments.

Experiment One. One set of respondents is presented with this hypothetical:

You want to buy a new television and have a particular model in mind. Calling around, you find that only two stores, Bob’s Electronics and Tom’s Electronics, carry that model. Bob’s Electronics is located very close, about a 5-minute drive, but offers no discounts on the television set. Tom’s Electronics is located farther away, about a 30-minute drive, but offers the television set [tax-free, which is equivalent to an 8 percent discount] Where do you go to make your purchase? 

A second set of respondents is presented with a near-identical hypothetical. The only difference is that instead of offering an 8 percent discount through tax avoidance, Tom's Electronics offers a 9 percent discount through an in-store markdown. 

Unsurprisingly, a majority of both respondents said they'd bypass Bob's and drive 25 minutes further to Tom's to get a discount. But a lot more said they'd go to Tom's for an 8 percent discount attributable to the TV set not being taxable (76 percent) than said they'd go to Tom's for a 9 percent discount directly from the seller (59 percent). The no-tax inducement was significantly more effective than the (larger!) markdown inducement.

Well, some people get suspicious when they hear a store has marked something down. Maybe that means it's a floor model into which small children have poured sticky fruit juice. That brings us to ...

Experiment Two.  New hypothetical. The first set of respondents is asked:

Imagine that you are walking through the mall looking for a particular jacket that you have seen advertised. You come across two closely located stores that carry it. The first store offers no discounts, but has no wait to purchase the coat. The second store is having a special [“axe-the-tax” sale, with the store selling all items tax-free, equivalent to a 9 percent discount]. However, due to the popularity of the sale, there is a wait to purchase items there. How long would you wait in line to receive the discount? 

A second set of respondents is presented with a hypothetical that's identical except that instead of a tax-free discount that saves 9 percent on the coat, the discounting store offers a "customer rewards" sale that discounts all items in the store--not just the coat--by the same 9 percent. Then both set of respondents are presented with a variation on each of their hypotheticals asking not how long they'd wait in line to receive a 9 percent discount, but rather, how deep the discount would have to be to persuade them to wait 15 minutes.

In both instances, the 9 percent discount attributable to a tax exclusion was deemed more compelling than the 9 percent discount attributable to a store-wide sale. The tax avoiders were willing to wait longer, on average, for their 9 percent discount (32 minutes), than the customer-rewards-seekers (26 minutes). Similarly, the tax avoiders required a smaller discount to wait 15 minutes (6.7 percent) than the customer-rewards-seekers (7.3 percent).

Well, people aren't necessarily very smart about how they purchase consumer goods. They're probably smarter when it comes to making investments. That brings us to ...

Experiment Three. The first set of respondents is asked whether they prefer to invest an inheritance in their bank account or in a corporate bond. The bank investment carries no risk and pays $75 per year in interest (after $25 in taxes is removed). The bond investment carries some risk and requires that the principal not be withdrawn for 10 years, but it will pay $120 per year (after $40 in taxes is removed). 

The second set of respondents is asked to respond to a hypothetical that's identical, except this time the bond is a municipal bond, not a corporate one, and therefore is not subject to taxation. In spite of that difference, the net annual interest from the bond remains an identical $120.

This time the disparity was huge. Eighty-two percent of the tax avoiders chose the risky $120 interest from the municipal bond investment over the non-risky $75 interest from the bank investment. But only 18 percent of the non-tax-avoiders chose the corporate bond investment, even though the difference was the same $45.

Well, maybe a government-bank bond seems less risky than a corporate bond. That brings us to ...

Variation On Experiment Three. This time there is only one pool of respondents. It is given a choice between two bonds, which are not identified by type at all. Both bonds have the same risk and both require the principal not be withdrawn for 10 years. One bond pays $400 per year in interest, on which $100 is removed for taxes. The other bond pays $300 per year in interest and is tax-free. Want to guess how many preferred the tax-free bond, even though it netted precisely the same amount? Seventy-seven percent! People would much rather make $300 per year knowing they aren't contributing anything to the government than make $300 per year after taxes. It makes no financial difference to them, and a $100 difference to the Treasury. They hate paying taxes so much that they'd rather deprive the government of revenue than not, even when there's no benefit to themselves.

This, apparently, explains Grover Norquist's existence. Norquist, I suppose, would dispute that this behavior is irrational. He would likely say that people's wish to see the government shrink is entirely divorced from financial self-interest. They aren't cheapskates, they just hate big government. But of course poll after poll shows that people don't want the government to shrink--not when "the government" is described to them not as an abstraction but as specific government programs. What people want is big government that they don't have to pay for.