Sadly there was no randomized evaluation, so we can't be sure. But there are good reasons to believe that Malawi's national fertilizer subsidy scheme has been successful in increasing agricultural production.
If you have looked something up on Wikipedia during the past few weeks you may have noticed a banner across the top of each page with some kind of "personal message" from Jimmy Wales.
This choice of message was no accident. What is fascinating about this fundraising campaign is that every banner is being tested against alternatives to maximize clicks and donations, and all of the tests all being documented online.
I had an op-ed published yesterday in the FT with Abhijit Banerjee, Pranab Bardhan, Esther Duflo, Erica Field, Asim Khwaja, Dilip Mookherjee, Rohini Pande and Raghuram Rajan, discussing the recent crisis in microfinance in India. You may need to register (for free) with FT.com to read the whole thing.
This past week, wrangling over the Bush-era tax cuts has riveted Washington... If only there were a scientific way to determine the real impact of taxation on industriousness, labor supply, and innovation.
According to some scholars, there is. Randomly assign a representative sample of the population — say, 10,000taxpayers — a lower tax rate, and see what happens...
It has long been the standard practice in medical testing: Give drug treatment to one group while another, the control group, goes without.
Now, New York City is applying the same methodology to assess one of its programs to prevent homelessness.
But some public officials and legal aid groups have denounced the study as unethical and cruel, and have called on the city to stop the study and to grant help to all the test subjects who had been denied assistance.
IPA Research Affiliate David McKenzie put together this Chart based on some of his research for the World Bank blog. The migration study wasn't randomized, but was based on 4 surveys and the next-best statistical methodology. And with impacts that big, you have quite a lot of room for bias.