Truth in advertising: ChildFund, Kiva, and Bolsa Família

I had the pleasure of attending a “tweetup” in NYC on Tuesday with a bunch of development and ICT4D* bloggers/tweeters. It was good to put faces with the writing. There were a number of interesting conversations, and at one point Linda Raftree asked a simple question:

How is the new interest in direct cash transfers different from child sponsorship programs?

I decided to dig a little deeper, but I’m taking the question in a different direction. This post ties together ChildFund, Bolsa Família and Kiva because I think these organizations tell us something about funding, marketing and accountability for programs serving the poor.

1. Sponsorship model disaggregated: fundraising v. programming

You may remember that American television sets were once swamped with ads from programs like the Christian Children’s Fund (now ChildFund) urging donors to change a child’s life with a small monthly sponsorship. This was a hugely popular charity model for decades, but you don’t hear much about it anymore.

Recognize this guy? ChildFund spokesman Alan Sader

The child sponsorship model worked in the sense that it brought in money. Donors felt emotionally connected a particular child, which improved the chance that they would keep giving. ChildFund and other NGOs facilitated this emotional connection by supplying photographs and grateful letters from the child.

Did I say “facilitated”? Some NGOs actually fabricated this emotional connection. Of course, I’m skeptical that any “real” emotional connection can be made in this sort of relationship — but more importantly: the updates from the sponsored child were sometimes made up. A Chicago Tribune exposé in 1998 revealed that the letters from children were often faked. I guess that’s necessary if the child dies but you want the donor to keep giving, which happened in a number of cases. Unsurprisingly, these revelations lowered the public’s support for such programs. (Hat tip to David Roodman, and meta-hat tip to Tim Ogden)

While the fundraising operations suffered, the programmatic side underwent changes as well. The practice of giving money to an individual child is fraught with problems, such as the high transaction costs for the distributing organization, as well as the way it singles out individual children within a community full of children in need. So the NGOs modified the programs: the funds no longer go to particular children, but to support the community where the sponsored child lives. Here’s how ChildFund describe the current model:

“When you sponsor a child, ChildFund will use your monthly sponsorship donations, combined with donations from sponsors of other children in your child’s community, to support community-wide programs.” (How Sponsorship Works)

Similarly, Plan International states:

“Your money does not go to the individual child that you sponsor. So that Plan can make efficient use of funds, the money is pooled with contributions from other sponsors to support programmes benefiting communities worldwide.” (Sponsorship FAQs)

I should be clear that I’m not entirely sure of the timing or causality on this, i.e. whether NGOs made this shift before or after sponsorship lost favor with donors, and whether the one caused the other (if anyone has any information on this, I’d appreciate it). In any case, NGOs have now disaggregated the original sponsorship model: fundraising continues to manufacture the donor-child emotional connection, while programming takes a community-wide approach.

2. Sound familiar? Kiva’s model runs on a similar principle

There was a hullabaloo last year over whether Kiva was misleading its lenders. The short version: David Roodman at CGD kicked off a debate by noticing that Kiva doesn’t actually connect lenders to micro-entrepreneurs in need capital, as its website and hype suggested; rather, Kiva supports partner microfinance institutions that post profiles of their borrowers on the Kiva site, generating capital that Kiva lends to the partner institutions. The debate revolved over whether Kiva’s lenders fully understood the process, and it led Kiva to redesign some of its webpages.

David Roodman, Saundra Schimmelpfennig and others noted how the Kiva model is similar to the present child sponsorship model: create an emotional connection between the donor/lender and the child/borrower in order to bring in the money, but run the program on the beneficiary end in whatever way makes the most sense.

3. And the conditional cash transfer model?

More recently, one of the hot topics in the development industry has been conditional cash transfers (CCTs). The approach gained notoriety with Brazil’s Bolsa Família program, which provides funds to mothers who keep their children in school and take them in for regular medical check-ups. The CCT model has since become a darling of the development industry and has been tried in a number of other countries. (For more, see these recent Economist articles on Bolsa Família and conditional cash transfers.)

CCTs flip the disaggregation that we see in the current ChildFund and Kiva models. CCT funds typically come from a government or other impersonal donor, while the program execution distributes them to individual beneficiaries. The model acknowledges the key benefit of the market: a government/NGO can’t determine or provide many of the services that poor people need, so it’s much better to just give money to the poor and let them choose for themselves. It may also be a recognition that many poverty alleviation efforts had become biased toward community-wide public goods, and that more funds should be directed toward private, household-level goods.

What’s the key takeaway from these three organizations?

The takeaway lesson is this: there need not be an inherent connection between the way an organization gets resources and the way it spends them. This is a bit counterintuitive, as we tend to think that funding and accountability mechanisms should be linked. To give a high-profile example: one of the failings of the Iraq Oil-for-Food Program cited by the Volcker Committee was the fact that funding for UN humanitarian operations came from the Iraq government’s oil sales, while the accountability was supposed to come from the UN itself — but didn’t (see page 114 of this doc).

On the other hand, isn’t the alignment of funding and accountability a major problem for international development organizations? Program implementers should be accountable to the intended beneficiaries, but they’re actually accountable to the funders. The principal-agent problem between funder and funded is compounded by misinformation and misaligned interests between the funder and the ultimate beneficiary. We tend to think funders should hold the funded organization accountable — but who holds the funders accountable? If the NGO is closer to the beneficiaries than the funder, maybe the NGO is better positioned to determine how well the program is meeting the beneficiaries’ needs.

What about truth in advertising?

This begs the moral question: Is it okay for an NGO to deceive its funders? That was at the heart of the Kiva debate. To put it less provocatively: Is it okay to use a simplified story to raise money, and allow the funders to believe what they want to believe? The real trouble comes when many organizations engage in such practices. As Saundra Schimmelfpfennig put it:

“While deceptive advertising may help that individual charity attract and keep donors, it hurts the aid industry at large by setting up unrealistic and even harmful donor expectations and requirements.”

This is absolutely true. The concern is not limited to any particular program model. Another of the bloggers I met on Tuesday, Tom Murphy, had a recent exchange with Smile Train over their marketing. Also, there was a similar debate recently about simplified marketing in political advocacy over conflict minerals legislation and the eastern Congo (see part 1 and part 2). Of course, such concerns are not limited to the development industry. All marketing involves simplified stories. The car won’t really let you to drive like that, the candidate won’t really be able to bring more jobs to your area, and the girl won’t really tear your clothes off if you use the body spray.

I share the concerns about the negative impact of many charities using simplified marketing messages. However I also don’t blame a single organization for using such strategies. It’s very easy to sit outside the organization (or in the programs office) and say that the fundraising staff should make different decisions. But they’re in a competitive context trying to do what they honestly believe is right. They’ve done randomized trials to compare marketing messages and found what works. Wouldn’t we make the same decisions?

If there’s a remedy for the negative impact of simplified marketing, it doesn’t lie with a single organization. It’s a collective problem requiring collective action. There should be broader efforts to educate the public about the developing world and create more accurate perceptions about how change happens. This has to start with high school history classes, undergraduate economics courses, and study abroad programs. Obviously the news media, TV and Hollywood have an impact on public perceptions too. This is an opportunity for a foundation or other large funder to create a better context. Then NGOs that fundraise from individuals will be able to deliver more accurate marketing messages and still be effective.

Of course the flipside — running the actual program and creating mechanisms for accountability to the beneficiaries — is a whole other beast…


* ICT4D = Information and communication technology for development. Like Venn Diagrams? Check out this helpful one from Wayan Vota.