With the temperature outside falling, the plight of the homeless during the winter months is coming into sharper focus. New research by a University of Pittsburgh professor - showing that competition can spur the homeless to save more - offers a possible insight into helping the homeless improve their future prospects.

The paper, titled "Competition As a Savings Incentive: A Field Experiment at a Homeless Shelter" and written by faculty members at the University of Pittsburgh and Arizona State University, appears in the November 2013 issue of the peer-reviewed Journal of Economic Behavior and Organization.

The research is the first to look in detail at the saving behavior of the working homeless. The authors believe that better understanding of this population may provide clues to the problem of chronic homelessness - and also provide insights into how the very poor reintegrate into the economic mainstream.

Sera Linardi, an assistant professor in Pitt's Graduate School of Public and International Affairs, and her colleague Tomomi Tanaka, an assistant professor at Arizona State University, worked with residents at an Arizona homeless shelter who were deemed to be at the highest level within the shelter by complying with community rules and honoring responsibilities. The shelter provided food, toiletries, and transportation expenses on the conditions that residents would find jobs and save a portion of their income each month. Shelter residents who participated in the study had jobs and were close to obtaining permanent housing.

Linardi and Tanaka categorized 123 residents into baseline and competition groups. The baseline group engaged in their ordinary saving behaviors, depositing a portion of their income in savings accounts or giving it to the shelter's case manager for safekeeping. In the competition group, however, residents were offered a prize of $100 for the individual who saved the largest percentage of his or her pay during a one-month period.

The results were impressive, with the average rate of savings increasing from $127 to $207 during the month of the competition - a 33 percent increase, from 53 percent to 86 percent of earnings saved by participants, whose average monthly income was $240.

Repeating the competition for a second month, Linardi and her coauthor no longer found any difference in saving between the competition group and the baseline group. They speculate that this is because those residents who responded to competition in the first month were able to leave the shelter soon afterward. Linardi also speculated that the experiment "may have worked too well, spurring a burst of unsustainable effort" and leaving residents in the competition group unable to save at the same high rate for a second month.

"If it was the case that people who wanted to leave the shelter sooner used the competition as a way to spark a last burst of effort, and that increased their total savings in the end, then it's useful," said Linardi, who also is an assistant professor in the Department of Economics within Pitt's Kenneth P. Dietrich School of Arts and Sciences. "If, on the other hand, it exhausts them, then that's not good."

"If you think everyone has an equal chance of winning," Linardi said, "you'll put in higher effort. So there's a higher chance they'll put in more effort here. Previously, in other environments, the homeless might have felt that they could not get ahead of others in terms of their finances. But competing among clients of a homeless shelter levels the playing field. "

The authors conclude that focusing on savings might be premature, and leads to only short-term results, if such initiatives are not paired with efforts to improve the job skills and income of the homeless over the long run.

Linardi is optimistic, however, that the experiment and its findings could be part of a broader solution to homelessness. Pointing to the lower-level residents at the Arizona shelter, who have difficulty finding jobs with regular pay and may be facing addiction or health struggles, Linardi said, "Maybe something similar to this competition, that is not directly focused on saving, would be appropriate." She cited the example of Back on My Feet, a national group helping the homeless improve their confidence and self-esteem through physical exercise challenges.

Linardi's choice to focus on the homeless was motivated, in part, by her experience as a volunteer math tutor at a California shelter. Watching students leave the shelter for temporary housing, only to return later because of unemployment or medical bills, impressed upon Linardi firsthand the difficulty of escaping homelessness.