October 29, 2010

A Tax by Any Other Name

New research reveals how labels trigger cognitive and emotional processes underlying decision making.


Politicians and advertisers have long understood that labels trump substance. Researchers have confirmed this truth again and again: in one study where consumers were presented with a choice between identical ground beef labeled “75 percent lean” or “25 percent fat,” most people described the “lean” label as better tasting. Similar examples abound.

Professors Eric Johnson and Elke Weber often collaborate on questions of how people weigh present-term options against long-term outcomes, and were interested in exploring how people respond to labels in the context of climate change and political affiliation.

“So much of our national dialogue centers around whether taxes are bad and what to do about climate change. And there’s this big debate about the virtues and vices of a carbon tax,” Weber says. “At the same time, the voluntary carbon offset market is doubling every year; people voluntarily pay for carbon reduction by purchasing carbon offsets.”

But just as 75 percent lean ground beef and 25 percent fat ground beef are the same, carbon taxes and carbon offsets are fundamentally the same — both fees fund efforts that limit or reduce carbon emissions. Weber and Johnson wondered: are some of the same people who oppose carbon taxes also purchasing offsets? Or do people who oppose taxes (presumably Republicans) oppose offsets while those who are less averse to taxes (presumably Democrats) embrace offsets? Or would other demographic characteristics predict how people responded to each of these labels?

To answer these questions, Johnson and Weber worked with doctoral student David Hardisty. Rather than simply offering subjects the options to choose between two labels, subjects had to choose a trade-off: whether or not to pay a fee that would fund carbon reduction efforts when purchasing gas, electricity, computers, or airline tickets.

The fee — which costs about the same as an actual carbon offset, such as adding $7 to a $345 airline ticket for travel from New York City to Los Angeles — was presented to half the subjects as a carbon tax and the other half as a carbon offset. The researchers expected that Republicans would be the most sensitive to the tax label — which they were, but to an even greater degree than expected: Republicans chose to pay the fee only 27 percent of the time when it was labeled a tax, but were statistically indistinguishable from Democrats when the fee was labeled an offset; 65 percent chose the offset rather than forgoing a fee. Democrats, less sensitive to the tax label, chose taxes and offsets about 65 percent of the time. Independents supported the offsets at about the same level as Democrats and Republicans, and fell between the other two groups when the fee was labeled a tax.

The researchers then presented the same set of options to a new group of subjects, but this time asked the subjects to complete a thought-listing exercise by typing what was going through their minds as they made their decisions. Again, subjects chose taxes and offsets at roughly the same rates as they had in the first study. The thought-listing revealed how much evidence subjects mustered for each option, and the emotional intensity of each decision: consistent with Republicans’ more typically negative view of taxes, these subjects listed many more negative thoughts when their choice was presented as a tax than did Democrats or Independents, but listed almost as many positive thoughts in support of offsets at the other two groups. Democrats had a generally positive reaction to the tax label, listing, on average, several more positive thoughts for taxes than for offsets. Independents listed slightly fewer positive thoughts for taxes than did Democrats.

“For Republicans, the first things that came to mind when thinking about taxes were the negative associations: government is inefficient, the money goes away and doesn’t produce any benefits,” Weber explains. “But when the same people were presented with offsets first, they listed the positive benefits they associate with the label: carbon might be a problem for our environment, it gets reduced by offsets, and the offset is dedicated explicitly to carbon reduction. The good things, rather than the bad, came to mind first.”

In past studies, Johnson and Weber were able to influence subjects to override their natural preference by simply switching the order in which they presented each of the options and asking subjects to think of all the reasons they could to choose the first option presented. To see if this mechanism held regardless of political affiliation, the researchers ran another version of the thought-listing exercise, this time asking half the subjects to first list reasons for supporting the tax and subsequently asking them to list reasons to buy a cheaper ticket. The other half of the subjects were asked to first list reasons to support the offset, and then to list reasons to buy the cheaper ticket.

Republicans had a relatively easy time generating arguments in support of carbon offsets, on par with Democrats and Independents, but they had a much more difficult time mustering arguments in support of the carbon tax than the other groups (half did not list any reasons for supporting a tax). The order of the query didn’t make much of a difference here, suggesting that though changing the order of a query can sometimes induce people to override their natural preferences, the effect is dampened when options are labeled in emotionally charged terms.

Weber and Johnson don’t attribute these results to any fundamental differences between people of various political affiliations. For example, Republicans didn’t report significantly less concern for the environment than Democrats or Independents reported. “It’s not that Republicans don’t care about the environment,” Weber says. “It’s that they think differently about taxes and are much more emotionally engaged with the idea of taxes. When a label or an idea really, really attracts, people have a hard time fighting it. And when it really, really repels, people have a hard time approaching it.”

But if Democrats, Independents, and Republicans so frequently agree on the merit of the idea that paying to reduce carbon emissions is a good choice, doesn’t that call into question whether Americans’ positions on key policy issues are truly as divergent as they have seemed in recent years?

Johnson and Weber think so. “Our political discourse is very polarized right now,” Weber says, “but our studies suggest that the recipe for greater rapprochement and constructive dialogue may be to get people to rethink issues and their components in a different order, setting aside the usual labels.”

“Labels keep us apart,” Johnson says, “but substantive consideration of these issues from different perspectives can bring us together.”

Eric Johnson is the Norman Eig Professor of Business in the Marketing Division and codirector of the Center for Decision Sciences at Columbia Business School.

Elke Weber is the Jerome A. Chazen Professor of International Business in the Marketing Division, codirector of the Center for Decision Sciences, and senior scholar at the Jerome A. Chazen Institute of International Business at Columbia Business School.

Read the Research

D. Hardisty, Eric Johnson, Elke Weber

"A Dirty Word or a Dirty World? Attribute Framing, Political Affiliation, and Query Theory"

View abstract/citation  Download PDF  

Have You Read Columbia Business?

Columbia Business School’s alumni magazine connects alumni with each other and the School; celebrates alumni milestones and accomplishments; and chronicles the impact of Columbia Business School alumni, faculty members, and students on the global business landscape.

Read Columbia Business>