Studying How Income Inequality Shapes Behavior

It's clear that inequality in America has grown at a fast clip in recent years. From 1980 to 2010, the top 1 percent's share of income has doubled from 10 percent to 20 percent, while the income share of the bottom 90 percent fell from 65 percent to 52 percent. Those kinds of swings have spurred public outrage and protests, as well as a fair amount of handwringing on the part of politicians.

What's less clear is how this rising level of inequality has affected the nation. Researchers have tried to determine its impact on a wide array of indicators—among them economic growth, public spending, financial stability, political representation, and average health and educational outcomes. But this research has done little to demonstrate the effects of inequality. Some researchers have found no effects at all, while others have identified opposite effects in different situations.

“The beauty of running lab experiments is that we can simulate many different types of inequality and look at many different types of decisions.”

"You'd think that with such a large change in our society, you'd see a definitive impact somewhere," says David A. Moss, the John G. McLean Professor of Business Administration at Harvard Business School, where he teaches in the Business, Government and the International Economy unit. "It's striking how much work has gone into examining this question—including by many truly outstanding researchers—and how little we have to show for it, at least so far."

Of course, it's possible that rising inequality simply doesn't exert any significant economic, social, or political effects. This would explain the lack of findings, he says. But it's far from the only explanation, and perhaps not the most likely one. In a recent working paper cowritten with Anant Thaker (HBS MBA 2011) of the Boston Consulting Group and Howard Rudnick of the Tobin Project, Moss offers an alternative perspective: Perhaps we've been looking at the question the wrong way.

Inequality And Making Decisions

In Inequality and Decision Making: Imagining a New Line of Inquiry, Moss and his colleagues propose that rather than looking at the effects of inequality mainly on the macro level, researchers should also look at the micro level, exploring how rising inequality might affect the individual decisions people make across a wide range of areas, from risk-taking to voting.

Moss offers a simple analogy: Imagine you want to test a theory that punches cause stomachaches. If you examine the data and find that some people get punched without getting stomachaches, and that others get stomachaches without getting punched, you might mistakenly conclude that there is no causal relationship between the two. The problem, of course, is that only certain types of punches-punches to the stomach-produce stomachaches. "Without greater precision about mechanism, it would be easy to wrongly dismiss the original hypothesis that punches cause stomachaches," Moss explains. "And the same may be true about inequality and our inability at this point to say much about its effects. The problem may be that we don't yet know enough about mechanism."

Moss followed the same logic working with colleagues at the Tobin Project, a Cambridge-based independent, nonprofit research organization that he founded in 2005. As he read the literature on inequality and spoke with numerous scholars in the field, Moss noted the diversity of situations in which the phenomenon has been studied. Income inequality might look very different-and exert very different effects—in one country or one city as opposed to another, due in part to differing institutional contexts. If so, then by aggregating the data, it's possible that researchers have been lumping together dissimilar cases that effectively cancel each other out.

If there was a way to tease out the mechanism by which inequality could affect this or that economic indicator, he reasoned, then it might be possible to refine the large-sample studies in productive ways-focusing only on punches aimed at the stomach, rather than all punches, in examining the connection between punches and stomachaches, for example.

One way to uncover the consequences of inequality, Moss and his colleagues propose, is to set up experiments in a lab to look at possible effects on individual decision-making. Behavioral psychologists have found that people often make decisions based on their frames of reference. Some have argued that by changing that frame of reference, rising inequality could change the way people make economic decisions. For example, economist Robert Frank has suggested that inequality might induce higher spending as we try to "keep up with the Joneses," creating "expenditure cascades" that could destabilize the economy as a whole.

In another example, researchers HBS Associate Professor Michael I. Norton and Columbia Business School professor Ilyana Kuziemko have suggested that those close to the bottom of the economic ladder may take greater risks than those above them in order to avoid tumbling down any further. They call the behavior "last-place aversion."

To test this theory in the lab, the pair teamed up with HBS Assistant Professor Ryan W. Buell and Stanford PhD student/candidate Taly Reich set up a lottery where participants were given varying amounts of money (some received larger amounts, others smaller) and then had the option to buy a ticket that, were they to win, might allow them to move up economically. The researchers found that the likelihood participants would choose the lottery was essentially the same across all rungs of the ladder-except the bottom two rungs, where participants opted for the riskier option more often.

More Research Needed

So far, these are among the first experiments that have even touched on the possible connections between inequality and decision-making. The researchers believe more experiments are required to figure out what these connections might be. "The beauty of running lab experiments is that we can simulate many different types of inequality and look at many different types of decisions," he says. "It's an excellent place to begin forming hypotheses about mechanism."

Recently, the Tobin Project brought together a dream team of researchers, from various fields and universities, to develop a novel set of experiments on inequality and decision-making. Joining Moss and Tobin program manager Howard Rudnick are Norton and HBS colleague Francesca Gino , Kuziemko and Columbia University colleague Ray Fisman (HBS PhDBE 1998l, and Nancy Adler of the University of California, San Francisco. The first experiments will take place later this year at Harvard Business School's Computer Lab for Experimental Research.

Members of the research team have already tried a dry run of one of the experiments, which involves participants being promised a sum of real money at the outset of the experiment (it could be high, low, or in between), and then answering a set of questions about the broader distribution, such as whether a more or less equal distribution would be preferable. "It's funny, even though you know it's just a game, it's hard not to notice if you receive only $2 when the person next to you receives $30 or even $50," Moss says. "The question is whether that affects the decisions you make about what to do with the money and what types of public policies you support. We'll see."

The experiments themselves will not definitively explain the effects of inequality, since they are set up in an artificial environment outside the complicated interplay of real-world economic systems. But Moss and his colleagues hope that they will point the way toward better large-sample studies going forward. "I don't think the lab experiments will tell us an answer," he says, "but they may move us in the right direction. That's the hope."