Is cold, hard cash more or less helpful to those in need than conditional aid? Photo: jtyerse (flickr)

Economics is about choices. What I spend money on is what I value most, because it comes at the cost of something else. People are rational, and thus make rational decisions—right?

This is the premise, says NPR, behind a new charity called GiveDirectly, which has removed the conditions from the money it gives its beneficiaries. While most charities and NGOs provide financial assistance, they do so by specifying what the money can be used for. GiveDirectly runs on the assumption that, given money through mobile banking, people will prioritize for themselves where that cash is most needed — and spend it accordingly, whether that be children’s education, health care or business startup costs.

And yet the economic principle of "rational" decision-making is oversimplified. People of every income bracket do not always spend money in the ways that would most benefit their families. According to a study on the spending habits of the poor, families that earn less than $2 a day in the average country studied spent 2 percent of their income on education, with about 5 to 7 percent going toward alcohol, tobacco, festivals and other entertainment. Spending patterns for the average American, according to a 2008 US Department of Labor study, are almost identical.

GiveDirectly is still in early stages, and is conducting a study of its own to see how its money will be spent under real life conditions. Unconditional aid undoubtedly shows a higher level of respect for beneficiaries' decision-making, and there is reason to hope it succeeds. Giving money without condition lowers administrative costs because it requires no follow up, meaning that a higher percentage of every dollar donated goes directly to those who need it.

But if a lack of conditions means a choice to spend more money at the local bar, not to put more children through school, GiveDirectly may be, at best, ineffective.

Correction: Aug. 16, 2011

A previous version of this post misstated the amount of money an MIT study found poor families spending on tobacco and alcohol.