I linked last week to Matthew Yglesias’s Slate piece “The Best and Simplest Way to Fight Global Poverty,” which reports on a study that gave unconditional cash grants to poor young adults in Uganda and found that four years later, recipients of the grants had more business capital and higher earnings than those in a control group. I thought about the study again over the weekend as I listened to a Planet Money podcast about a charitable school-building project in Haiti that exemplifies how hard — and potentially less effective — it can be to give poor communities assets like schools rather than just handing cash over to poor individuals.

At his Small World blog on Bloomberg Businessweek, Charles Kenney explores donors’ reluctance to give unconditional cash and describes some of the research showing that such an approach might be effective:

Worldwide, richer people express fears about handing money to poorer people. Giving poor people money is no way to stop them being poor, the thinking goes: Surely they will just waste it. Instead, we design complex, bureaucratic programs like SNAP, the supplemental nutrition assistance program (formerly known as food stamps) to help poor families buy food and only food. That way, they can’t buy a trip to Disney World with our tax dollars.

…A growing number of studies suggest that this is wrong-headed, that just handing over cash even to some of the world’s poorest people actually does have a considerable and long-lasting positive impact on their incomes, employment, health and education.

Kenney’s piece also made me think of a World Bank-funded study in Malawi (which I wrote about previously) that gave both conditional and unconditional cash transfers to girls and found that those receiving the transfers had lower HIV infection rates than girls in a control group who got no money. The researchers found lower HIV prevalance rates in both the girls whose grants were contingent on staying in school and those who received the money unconditionally.

If enough research shows that the biggest impact per donor dollar can come from unconditional cash transfers, large donor organizations might alter their project funding guidelines in response. But what about charitable donations from individuals? That’s what I was thinking about while listening to Planet Money’s “It’s Hard to Do Good” episode about the efforts of Tim Myers, a retired Colorado contractor, to turn thousands of dollars in donated money into a school for Haitian children.

The saga began in 2010, when Planet Money’s Adam Davidson and Caitlin Kenney traveled to Haiti to see what kind of impact post-earthquake foreign food aid was having on a rice-growing region, L’Artibonite, which was spared physical earthquake damage. One of their stories included clips from a visit to a school in Villard, where schoolchildren crowded into a one-room church for lessons and many students didn’t have textbooks. After hearing about the schools’ financial struggles, listeners wanted to know how they could give money to the school. The school’s principal, Enselm Simpliste, set up an account to receive donations — and when donors gave $3,000, Simpliste decided to use the money to build a schoolhouse. Months later, the Planet Money team returned to Villard and found that all the money had been spent, and all that Simpliste had to show for $3,000 was “a foundation, some concrete blocks and some rocks and sand.”

The story didn’t end there, though. Two Planet Money listeners, Tim Myers and Fred Ireland, decided to raise even more money and travel to Haiti to personally ensure the completion of the school construction. They launched a nonprofit called the Haiti School Project and began a series of trips to Villard. As a retired contractor, Myers figured he had the experience to oversee the project, but the group was committed to using local labor for the construction in order to ensure the money actually went to the community they intended to help. Myers calculated, though, that the project would cost more than $140,000.

Now, Planet Money reports that although the school isn’t as elaborate as originally planned — most notably, it’s a single story rather than two — the walls are up and construction is nearly complete. The Haiti School Project eventually raised $100,000, and has purchased textbooks and trained teachers as well as employing local residents. But Myers says it’s been far more time-consuming and complicated than he originally expected. Plus, because Haiti doesn’t have public schools, the project has essentially given a gift of a school building to Principal Simplice, and have only his non-enforceable agreement that he will use it to provide schooling to local children whose families struggle to afford school fees.

When asked in this latest episode if it might have been better just to write checks to schoolchildren’s parents, Myers says, “Definitely. I could’ve saved myself a lot of heartache and money.” But, he also doubts the group could’ve raised $100,000 if they’d been pitching unconditional cash for parents rather than school construction. “Don’t people have this feeling that they want to be involved in something specific?” he suggests to Planet Money. “They want to see where their money’s going. … People like to feel like, when they give money for something, it’s going for something specific.”

It’s important to note this school was the Haiti School Project’s first (and, I’m guessing, only) attempt at building a school. Organizations with a long track record of building schools or health clinics or other physical assets have a better grasp of what each project entails and can probably spend donations more efficiently than Myers and his colleagues did. But the most efficient use of donations may be simply giving money directly to poor people. If more research accumulates to suggest this is the case, will those of us who enjoy visualizing the specific projects our donations support be willing to give just as generously, and trust the recipients to decide how to spend the money?