Kiva. In fact, here is a Kiva promotional picture that is a perfect example: a loan to buy a sewing machine. Such an investment raises personal productivity a little and may help generate income above the poverty level, but such investment might also be in lieu of going to school.
Whether these microloans are effective poverty alleviation programs is still a hotly debated question, thought the balance of the evidence seems to be that they do increase consumption of recipient households. My new colleague at OSU, Elizabeth Schroeder, has a very important paper that finds significant effects.
My concern is that these loans, which tend to be small, high-interest and where repayment begins immediately, can distort incentives and cause recipient households to concentrate on small-scale entrepreneurial activities rather than investing in education. In fact, there is some good evidence to suggest that this is exactly what is happening.
What concerns me the most is how this trade off affects the long-term growth trajectory of an economy in which microloans are prevalent. So I teamed up with my macro colleague Bruce McGough and we built a model of just such an economy and with it we show how microloans can actually lead to lower growth (lower future GDP) and lower welfare. This is a paradoxical result because we model microloans as in injection of new money from outside the economy (rather than a redistribution) and the GDP and welfare results hold despite this fact. The mechanism that causes this is the well-established empirical fact that average, not just individual, levels of education are beneficial to growth. By suppressing educational investment, you can lower the overall productivity of the country and leave it worse off than before.
Now the trade off I illustrate with the Kiva picture might still be the right one to make - after all school is no help if you are starving - but it is important to understand the implications. Our paper is, therefore, a cautionary tale not a condemnation.
Read all about it here.