He says he became a researcher to understand the world. Abhijit Banerjee wishes people would be more willing to really get to the bottom of problems and to and look for a tailor-made solution. At the Massachusetts Institute of Technology (MIT), he created the "Poverty Action Lab," a comprehensive database for development aid.
Banerjee says it's difficult to express the overall effect of development aid in numbers. This, he says, is one of the greatest misconceptions he encounters in his work. Instead, what can be measured quite accurately is the impact of a very specific action - such as subsidizing seed for a particular village.
For years his team has performed field work around the world, choosing villages or families at random and putting them into control groups. It's a practice taken from medicine, where it is known as randomized controlled trials.
The mosquito in the net
Banerjee's experiences in the field have led him to dispense with numerous cliches. News reports about free antimalarial nets distributed among the poor that were put to other purposes such as fishing nets and bridal veils caused indignation in industrialized countries.
But Banerjee doesn't see the problem. "You have to sleep night after night in there. In a hot place, the air doesn't go as quickly through the mosquito net. You're worried there is a hole in it and the mosquito will go in and then the mosquito is making a sound and you're always hearing it. It's very hard to sleep."
Of course Banerjee is not opposed to mosquito nets. But he says he wants to see more understanding for people in poor countries for whom a vaccination appointment is just as much a chore as a visit to the dentist is for others.
More civil servants, fewer microcredits
Another important issue for him is microcredits. These were once considered a powerful tool to turn the poor into entrepreneurs. But studies Banerjee conducted with his colleague Esther Duflo have tarnished this silver bullet.
When they asked what kind of job parents in the developing world wanted for their son, 80 percent wanted him to have a job in government. Another 15 percent suggested a job in the private sector. But no-one wanted an entrepreneur.
In fact, many of the respondents themselves ran a business. "They would love to work for someone else, but they couldn't get a job," Banerjee said. "And that's reflected in the fact that even if you offer them a loan they don't take it."
Too many goals, too little development
Banerjee was in one of the expert commissions that spelled out the United Nations' new sustainable development goals. These are meant to replace the more modest Millennium Development Goals when they expire next year.
He says the process is out of control. Nearly every participating country and organization added goals, to reach the point where there are now 169 goals and sub-goals. That's something he says no one can measure, let alone achieve: "I think somewhere the process has to give and we have to go down to 30 or I think this whole thing has become pointless."
What is now needed, Banerjee says, is a strong personality like former UN Secretary General Kofi Annan to take charge. Someone who would put his fist down and made sure that the list is kept to the essentials.
"People want to think of grand shifts in theory, but the world is different," Banerjee said. "I think there are many more things we could change if we were willing to tweak things."
And the most important of these, he said, would be to listen to people to find out what they really need.