Fast Company

J-Pal’s Most Provocative Inquiries

J-PAL members hope their findings will inspire smarter anti-poverty policy. Here, a look at some of their most provocative inquiries.

Question

Is the prospect of top-down (government) monitoring or grassroots accountability better at preventing corruption on a local infrastructure project?
Subjects: 600 villages on Java, Indonesia

Answer

Grassroots participation had negligible effect, while a top-down system reduced missing funds by 8%.

Question

Does removing group liability -- the mutual accountability common in microfinance -- raise loan default?
Subjects: 3,285 clients of a microfinance bank in the Philippines

Answer

No. Counter to conventional wisdom, defaults did not rise after group liability was removed. But loan officers were less likely to expand to new areas without group liability becauseof the lingering belief that it helps.

Question

How do couples pool financial risk and respond to temporary changes in income?
Subjects: 142 married couples in Kenya

Answer

Inefficiently. Husbands increase spending when they get an income boost but not when their wives do. Wives' spending does not change in response to an income boost. Wives share their boost with husbands, but husbands do not sharetheirs with wives.

Question

What incentives most effectively encourage distribution of female condoms?
Subjects: Hairstylists in Lusaka, Zambia

Answer

Study ongoing with four groups: one of volunteers; one with small financial incentive; one with big financial incentive; one with nonmonetary incentive. Results may apply to other public goods.

Question

Does computer-assisted learning cost-effectively boost test scores in math and language?
Subjects: 111 primary schools in India

Answer

For math, yes. For language, there was negligible impact.

Question

What effects do cash incentives have on poor students and their families?
Subjects: 10,000 students in Bogota, Colombia

Answer

Students receiving incentives improved attendance and passing rates but showed little change in academic effort. Incentives affect siblings negatively -- families may favor children who receive cash.

Question

Do merit-based scholarship programs for girls affect the performance of both girls and boys?
Subjects: 128 primary schools in western Kenya

Answer

Yes. Test scores rose significantly for both boys and girls in a community, even though only the girls were eligible to win scholarships.

The J-PAL members who authored the above studies are Jonathan Robinson of UC Santa Cruz (Kenyan couples); Ben Olken of MIT (Indonesian corruption); Dean Karlan of Yale (Philippine microfinance); Michael Kremer of Harvard, Edward Miguel of UC Berkeley, and Rebecca Thornton of the University of Michigan (Kenyan scholarships); Marianne Bertrand of the University of Chicago and Leigh Linden of Columbia (Colombian incentives); Abhijit Banerjee and Esther Duflo of MIT, Shawn Cole of Harvard, and Leigh Linden of Columbia (Indian computers); and Nava Ashraf of Harvard and Oriana Bandiera of the London School of Economics (Zambian condoms).

Add New Comment

0 Comments