The world we live in is full of situations where evidence is created to support policies. Of course, it should be the other way around. Policy should grow out of evidence. Take, for example, attempts to address poverty. There are those who support aid to developing nations. Others state that such aid will more likely harm countries since they do not develop their economies and instead grow dependent on the aid.

As of 2013, no less than $134 billion was spent on development assistance around the globe, according to the OECD (Organization of Economic Co-operation and Development), and still many people remain in poverty. Don’t get me wrong; progress is being made. But much more could be achieved more quickly with new direction.

In 2015, the World Bank expects there to be 702 million people living on less than $1.90 per day. As this figure shows, poverty is a remarkably big problem and, with lives at stake, it makes sense to promote better alternative approaches to poverty reduction.

Recently there has been a new breed of thinkers looking for more effective ways to address poverty. Esther Duflo, a French economist now at MIT, is one of them. Instead of siding with any one ideology, she put theories to the test through experimentation. She looks at the world as a laboratory.

Instead of addressing poverty as an abstract problem, she specifically concentrates on concrete matters such as malnutrition, illness and education. If she has any biases they are probably toward approaches encouraged by behavioral economists; she is generally in favor of nudging people toward positions that will help them improve their lives. While somewhat intuitive, the “nudge concept” is a fairly new approach. Generally the nudge community’s ideas are very simple, effective and not as expensive. Let’s take a look at this approach through specific examples.

One idea to improve school attendance in a low-income village is to offer money conditioned on attendance. Duflo and her team found it’s not even necessary to condition the stipend on a specific outcome. Instead, only by labeling these payments as education money transfers, they manage to increase school participation. In fact, up to 97 percent of households got at least one child enrolled in school. In other words, no more money was spent to actually achieve a goal than when these noneducation grant programs were established.

Another nudge found to increase school participation is as simple as deworming kids and explaining the benefits of education to their parents. Without spending on conditional cash transfers, these simple solutions turned out to be effective, and the costs of deworming and explaining benefits are relatively low compared with cash transfers.

To take the idea of nudging a step further, another example relates to immunization. Sometimes this task is not easy because people often distrust vaccines. How could advocates create a cheap incentive that would encourage people to accept vaccination? Those aligned with Duflo offered lentils along with the vaccinations. Although 1 kilogram of food is not much, people were at least willing to give vaccines an opportunity when provided with this incentive. Ultimately, Duflo’s team managed to increase the immunization rate from 6 to 38 percent in parts of India.

This experimental approach eliminates the necessity of guessing and of siding prematurely with an ideological perspective. In the end, strategies like the ones above may not actually end poverty as quickly as we might like. Yet experimental proof suggests that using evidence-based measures is the right way to alleviate the consequences of scarcity.

John Hoffmire is director of the Impact Bond Fund at Saïd Business School at Oxford University and directs the Center on Business and Poverty at the Wisconsin School of Business at UW-Madison. He runs Progress Through Business, a nonprofit group promoting economic development.

Better Policies For Better Lives:  Act Now

By Mario Alejandro Mercado Mendoza