Karlan is in the country implementing a commitment program in 170 schools. The scheme sets up savings accounts for children that can only be spent on educational expenses. Early results hint at an answer to one question that has long plagued him: can commitment devices actually change important decisions, like whether to attend school?

“It could be that such mechanisms are only used by those who would have accomplished their goals anyhow. But we want to know if the commitment device leads to behavioral change that would not have happened otherwise, which is why we test these theories with a randomized trial where a group of other people are not offered the product,” he shouts over the static of a police radio.

“The study is not over, but initial results indicate both savings and school attendance are rising compared to a control group. You need basic items to go to school – pens, paper, books – so if you can’t afford them, why go? But if you have them, it could change your life by getting you to school.”

A PROFESSOR OF ECONOMICS at Yale University, Karlan is on leave for a year as he visits the Ugandan project and others around the world. He first came to experiment with commitment devices when he and a colleague hit on a novel idea to lose weight. Noticing that they both set goals they never reached, the pair made a financial contract to lose 38lb (17kg) each.

With half of their annual income at stake and a no-renegotiation clause, both men hit their target – and held it for several years. Realizing the power of such arrangements, Karlan eventually helped found stickk.com, a free website that enables people to reach lifestyle goals, such as exercising, quitting smoking or studying through binding “Commitment Contracts.”

By committing to a monetary penalty, users significantly increase the cost of failure. Among the incentives offered by the site to help more than 150,000 people achieve their goals are such things as “anti-charity.” This involves pledging money for a cause you do not support if you fail to meet your target. Anti-charities on offer include such potential hackle-raisers as the National Rifle Association or the Manchester United Fan Club.

“Commitment contracts are a broad term to describe any device, informal or formal, that helps someone voluntarily increase the price of their vice or lower the price of their virtue. The main aim is to restrict or limit future choices in order to help resist temptations when making decisions. Devices don’t need to be binding in a legal, financial or forceful way, but that can help for some,” he explains.

FOR THE LAST FEW YEARS, Karlan has blazed a trail through the United States and developing countries seeking to understand the power commitment contracts may have to transform the life of the impoverished. But his work goes beyond this and also examines other behavioral and social interventions. Along with colleagues, he has examined, among other projects: crop price indemnifying loans for farmers in Ghana, charitable giving in the United States, a school incentive program in India, health education in Peru and aspects of microfinance and microsavings in various African and Asian countries.

Although the resulting papers are freely available, his book More Than Good Intentions (with Jacob Appel, 2011) provides a compressed version of the results and insights he has gained as to how new behavioral tools can assist the poor and how development money can be better invested. The book shows that by taking into account human irrationality, small changes in banking, insurance, health care, and other development initiatives can improve the lives of people everywhere.

“Everyone has different temptations. I may need help to lose weight, you to stop smoking, while someone else may need a self-control for savings. These are issues common to all people,” Karlan says, explaining the universality of some of his research.

“In a general sense, issues of temptation transcend cultural differences, in that you are not going to find one country full of self-control and one that is devoid of it. This isn’t to say some may have stronger savings cultures and others have bad eating cultures. However, the biggest differences are across people, not across cultures. We all have our own special weakness!”

IN GREEK MYTHOLOGY, Sirens were creatures whose irresistible music lured sailors to shipwreck on nearby rocks and drown. Ulysses, returning home from the Trojan wars, wanted to hear the music, but took the precaution of having the ears of his crew filled with beeswax to block out the sound. He then had his men tie him to the mast and ignore his pleas to be released until they had passed the island of the Sirens.

The plan worked. Ulysses heard the music and survived because he committed to a rational course of action at a neutral time (before he could hear the Sirens’ songs) and ensured that he stuck with his decision. In behavioral finance terms, his action was an example of a pre-commitment strategy that helped him avoid poor decisions.

Such behavior is known as a “Ulysses strategy” and has even been suggested as the basis of a new type of contract between clients and financial advisors (Behavioral Finance in Action, 2011).

In one of his projects, Karlan designed a savings product (SEED – Save, Earn, Enjoy Deposits) with a Philippine bank for people who wanted to restrict access to their savings, because they knew they could be tempted to access them before they reached their goal. “Traditional economics assumes people to be perfectly time-consistent,” explains Karlan. “That is, the tradeoffs I make today and tomorrow are estimated the same way that I plan to make all my future tradeoffs. But there are things you say you’re going to do in the future that, when the time comes, you don’t do. All that has changed between ‘now’ and ‘then’ is that the ‘then’ is the ‘now’ and you get tempted. We think we can resist temptation, but when it becomes ‘now’ we often can’t.”

From a traditional economic point of view, SEED made little sense. It offered no benefit over a normal savings account and actually involved a liquidity penalty. Account holders could only gain access when they achieved a self-set target. Yet, the product proved surprisingly popular with individuals who knew they had time-behavior issues. After 12 months, the average bank account savings of a SEED account holder increased 82%. Positive though the results were, as often with research, they opened up many more questions. Not the least of these is can such a device change long-term behavior? And so Karlan finds himself pondering the answer to this and other behavioral finance questions while his driver assists Ugandan police with their inquiries.

“If there is one thing I’ve learned, it is that we need to understand the needs of individuals better to develop better products” he muses. “Sometimes we are selling savings when credit would be better; sometimes we sell credits when we should be promoting savings. We tend to get caught up in the false promise of the product, without thinking about what is really helpful for people.”

The roadside transaction complete, America’s economic action man heads off and, as the telephone waves weaken and die, falls into silence.