Poor Logic or the Logic of Being Poor

What if we've had it all backwards: people make bad decisions because they're poor, rather than that they're poor because they make bad decisions.

Recent findings to that effect may be the most important social science research in decades because they could influence public policy for millions of people.

I first heard news of the studies back in late August, such as this story from The Atlantic. Researchers did two studies, one in the U.S. and one in India. In the U.S. study,

low-income people who were primed to think about financial problems performed poorly on a series of cognition tests, saddled with a mental load that was the equivalent of losing an entire night’s sleep. Put another way, the condition of poverty imposed a mental burden akin to losing 13 IQ points, or comparable to the cognitive difference that’s been observed between chronic alcoholics and normal adults.

That's over 10 percent of IQ lost because of stress imposed during the experiment to think about having to get your car repaired. Subjects who were well off, given the same prompt about the car bill, were unaffected. Here's a more detailed description (from Princeton University's news site) of the prompt:

The first set of experiments took place in a New Jersey mall between 2010 and 2011 with roughly 400 subjects chosen at random. Their median annual income was around $70,000 and the lowest income was around $20,000. The researchers created scenarios wherein subjects had to ponder how they would solve financial problems, for example, whether they would handle a sudden car repair by paying in full, borrowing money or putting the repairs off. Participants were assigned either an "easy" or "hard" scenario in which the cost was low or high — such as $150 or $1,500 for the car repair. While participants pondered these scenarios, they performed common fluid-intelligence and cognition tests.

Subjects were divided into a "poor" group and a "rich" group based on their income. The study showed that when the scenarios were easy — the financial problems not too severe — the poor and rich performed equally well on the cognitive tests. But when they thought about the hard scenarios, people at the lower end of the income scale performed significantly worse on both cognitive tests, while the rich participants were unfazed.

And that's folks with incomes only as low as $20,000. I wonder if the effect would have been greater if the subjects were below the poverty line?

In the Indian study, researchers gave intelligence tests to sugar cane farmers at two different times of year -- first when they were in the lean months, far from harvest time, and the other after they had been paid for their annual harvest. They found the same results -- about a 10 percent decrease in cognitive scores during the no-money times.

What do the authors recommend, in terms of changes to policy or practice? Again quoting the Princeton story:

The researchers suggest that services for the poor should accommodate the dominance that poverty has on a person's time and thinking. Such steps would include simpler aid forms and more guidance in receiving assistance, or training and educational programs structured to be more forgiving of unexpected absences, so that a person who has stumbled can more easily try again.

Eliminating poverty might be an even bigger step, such as by increasing the minimum wage. We have to get rid of out-dated ideas about what causes poverty and how to get out of poverty so we can make some useful and realistic decisions.

Other recent articles on the topic:

• The Psychology of Poverty and Welfare Reform (from the Inequalities blog)
• Details on how the studies were conducted
• One of the studies' authors, Sendhil Mullainathan, was on Marketplace Money today, so if you like to hear these things explained rather than read about them, check that out.