More Than Good Intentions

by Dean Karlan

  First, it strengthens the case for incentives, which proved again to be a powerful motivator. Second, it’s a prime example of the value of finding out what doesn’t work. As I said way back in the introduction, new ideas are essential. Rather than being discouraged that this particular approach wasn’t a home run, we should be gladdened that we can now rule out some ideas. Thanks to Thornton’s creativity and insistence on testing, we know more now than we did before, and future efforts will be better for it. Maybe the next HIV program will continue to use incentives to encourage the retrieval of test results, but will find a better way to sell condoms.

  Third, and most important, this project highlights the importance of understanding exactly how programs work, of tracing the links between the behaviors we can see and influence and the outcomes we care about. Sometimes those links are tenuous; frighteningly often they are altogether unexplored. The most obvious and ubiquitous figures that charities and development programs advertise—dollars spent, people enrolled—are only signposts. If we don’t understand how those things lead to the welfare of recipients, we are losing sight of what matters—helping people make real improvements in their lives. In the big picture of poverty, that just won’t cut it.

  12

  TO GIVE

  The Takeaway

  I hope that, by this point in the book, you’ve been inspired by some of the ideas we’ve seen that work to fight poverty. If you’re like me, you might also be struck by the daunting challenge of figuring out what to do.

  Odds are you don’t work for an aid organization, and you most likely won’t be called upon to design or deliver a development program. For most people, the “what to do” is about giving. As I touched on briefly in the introduction, we have more ways to give now than ever before—not just by putting checks in the mail, but also on the checkout line at Whole Foods Market, by text from our mobile phones, and online through Web sites like Kiva.org. As the avenues for giving have multiplied, individual donations have spread to fill them. In the United States today, individual donations exceed those of corporations, foundations, and bequests by three to one.

  So individuals have a lot of clout, and we can use it if we act together. What we have to realize is that every donation really is two things: First, it is a contribution to a particular organization, which helps them to run their programs; second, it is a vote. Giving to one organization means choosing it over another (actually, over thousands of others), and it sends a message. That message gets amplified whenever we discuss our favorite causes with friends, families, and co-workers; or when we talk about them on Facebook, Twitter, or MySpace.

  The development world is listening. Indeed, given that individual donations account for the majority of funds available to charities (not to mention the fact that Americans’ private giving to other countries exceeds the U.S. government foreign aid budget), it really doesn’t have any other choice. And that means that we have a unique opportunity to make our voices heard.

  The question is: How should we choose what to say, whom to support? This book has been mostly about ideas that hold promise for the poor. But ideas don’t just happen; someone—usually some organization—has to go out and make them happen. That means that when we consider contributing, we really need to know two things: First, is there good reason to think this idea will work to solve this problem in this setting? Second, will this particular organization implement this idea effectively and efficiently?

  This book has had a lot to say about the first point, but has not talked about the monitoring of organizations to determine which are efficiently run and which are not. I will leave you here with some thoughts and recommendations, and will continue the conversation through the Proven Impact Initiative Web site (about which more below), as well as through partnerships with groups that focus on aid effectiveness.

  First, one important note of caution. Discussions about aid effectiveness often hinge on the issue of administrative costs: Put pejoratively, what proportion of each dollar donated gets sucked away to overhead and fund-raising expenditures? Organizations with low administrative costs are generally thought to be better, as they spend more of their money on delivering products or services. But, truth be told, this is a really bad metric. Based on the evidence, it’s unclear whether administrative costs have any correlation at all with aid effectiveness. Some interventions simply cost more to manage than others.

  More important, administrative cost figures are fairly arbitrary; many items can be counted either as overhead, or as program services. The gray area of nonprofit accounting is just that—gray. So rewarding those with low administrative costs often amounts to rewarding those with aggressive accounting practices, nothing more.

  The question we should be asking when we give is simpler and more focused on what actually matters: For each dollar donated, how much good will be done?

  In the hopes of coordinating our voices behind that question, I want to leave you with seven ideas that have me particularly excited. These are some of the programs and products we’ve seen in this book that stand out. Each of them has been subjected to at least one rigorous evaluation, and has stood up well compared to some alternative solutions to the same problems. That does not mean they are all conclusively proven and above reproach. Some are closer to “proven,” while others are still “being proven,” which is just to acknowledge that the spectrum of evidence is wide. It’s a long journey from having no concrete knowledge to making confident prescriptions, ready to scale up without further testing. The ideas below are all somewhere along that path, but they’re not all at the end of it.

  I also want to be clear about what is not included here: a lot of promising ideas—many of which we have encountered in this book as well—that also have evidence behind them. The list below is not intended to be exhaustive.

  Finally, funding should not be confined exclusively to “proven” (or even “being proven”) ideas. We do need to take risks. As I said in the introduction, there is always a need for creativity, for trying brand-new and unproven tactics. That is the essential process that keeps us moving ahead. Organizations that innovate—and innovate thoughtfully, testing their new ideas rigorously—also deserve our support. That said, innovation without evaluation does not help the world as much as innovation with evaluation. I’d much prefer to give to an organization that conducts rigorous evaluations of its own programs because that gives me the confidence that in five or ten years they will likely be making better choices. They will adapt better to new contexts, technologies, and ideas. Many of the groups discussed in this book—notably Pratham (remedial education), Seva Mandir (teacher and nurse attendance), and Freedom from Hunger (microfinance)—are shining examples of organizations that continuously strive to improve through rigorous evaluation, that learn from their failures and their successes, and that publicize both so the world can learn too.

  Enough caveats. Here are the seven:

  Seven Ideas That Work

  Microsavings

  In chapter 7, we began with Vijaya, who was stuck in the borrowing “rotation” (and losing money hand-over-fist in the process) because her husband made it impossible to save at home. Then we saw women in Kenya signing up in droves for a costly bare-bones savings account—and improving their lives as a result. In fact, while claims of gender empowerment have long been made by microcredit advocates, the evidence suggests that women’s status and authority is elevated by saving, whereas credit has yet to exhibit such impacts. The need and the desire for savings are already in place; now we need to provide people with options. How many borrowers (like Vijaya) might be better served with savings products than with credit products? In all the enthusiasm over microcredit, we seem to have forgotten that basic lesson we learned from our parents and grandparents: Saving is important!

  Reminders to Save

  Saving is good, but it doesn’t come easy. We all know that. With all the things we have to spend on, saving rarely grabs our attention. There always seem to be
more pressing—or more tempting—options. Ultimately, most of us save less than we say we’d like to. As we saw in chapter 7, little reminders, like the text and direct-mail messages banks sent to their clients in Peru, Bolivia, and the Philippines, help cut through the cacophony of voices calling out for our money. They have proven to be cheap and effective ways to mobilize savings among the poor.

  Prepaid Fertilizer Sales

  Of all the efforts to increase fertilizer usage—subsidies, outreach from agricultural extension officers, demonstration plots—this simple solution is a no-brainer. For a vendor, selling fertilizer costs roughly the same no matter what time of year you do it. But, as we saw in chapter 8, timing can make a huge difference to customers. Kenyan farmers who had a chance to pay (full price!) for next season’s fertilizer at harvesttime, when their pockets were full from selling this season’s crop, bought 50 percent more. That’s a huge boost in productivity and agricultural output, practically for free.

  Deworming

  Sometimes the numbers really say it all. In chapter 9 we saw that deworming at Kenyan primary schools generated an additional year of attendance for about $3.50; the next best solution, providing free uniforms, cost about twenty-five times that much. And that doesn’t even consider the simple health benefits of being worm-free—they’re icing on the cake. In areas where intestinal worms are prevalent, fully subsidized deworming at schools is a dirt-cheap and tremendously powerful intervention. No wonder it’s catching on. Efforts by Innovations for Poverty Action and its partner organizations have led to millions of children being dewormed, but much more remains to be done.

  Remedial Education in Small Groups

  The balsakhi remedial program and Pratham’s reading camps, both of which we saw in chapter 9, are prime examples of a new direction in educational solutions for the developing world. They both found ways to work around dysfunctional school systems—the former by hiring private teachers, the latter by training volunteers—to promote genuine learning. Where existing schools are understaffed or overcrowded, outside programs like these may be the most effective path to education. Annie Duflo, research director at IPA, is working with the Ghanaian government, local teachers, and the UK Children’s Investment Fund Foundation to launch a large replication of this approach in Ghana; if all goes well at a large scale, IPA will march on to see it scaled up in many more countries.

  Chlorine Dispensers for Clean Water

  Two million people do not need to die of diarrhea each year. But they do—and solutions have eluded us for decades. Treating drinking water with chlorine is a cheap and highly effective preventative measure—so let’s get people using it. Despite the benefits of protection, distributing chlorine to households, even for free, has not proved effective enough. Yet providing free chlorine in an easy-to-use dispenser at water collection points—like we saw in chapter 10—has. Programs may become even more efficient and self-sustaining as different subsidy schemes are tested, but dispensers are a big step toward solving the basic problem: getting more people to drink clean water.

  Commitment Devices

  We saw in chapter 7 for savings, and in chapter 10 for smoking, that commitment devices can be an effective tool to help people reach their goals. This applies to everyone, rich and poor alike, although the examples in this book focus mainly on applications in developing countries. The SEED commitment savings account helped women like Sunny make significant improvements to their households. Targeting smoking, a major health problem in much of the world, we saw how a commitment account in the Philippines was strikingly successful in helping individuals quit. The basic principle behind both of these examples can be applied to many facets of life: Commitment devices let people make their vices more expensive and their virtues cheaper. In doing so, they help people make more virtuous choices.

  The List Goes On: Proven Impact Initiative

  The downside of the Seven Ideas That Work is that they are static. The ink has dried; they’re not going anywhere. But solutions in the fight against poverty are going somewhere. They are moving and changing all the time, driven by innovation and research. Hopefully these Seven Ideas will someday be eclipsed by Seven Better Ideas. I’ll be happy when they are.

  But how will you know when it happens? Unless you read economic journals or attend development conferences, it’s hard to say. The latest research doesn’t always make its way to the general public.

  That’s why IPA has launched the Proven Impact Initiative. This initiative will help ensure that donors both large and small have access to ideas that work, and an easy way to support them. We are now working with several partners to communicate this information to the general public, in the hopes that we can help donors be as effective as possible.

  So stay tuned. We are just at the beginning of our search for the right questions to ask, and for the answers that follow. We look to the countless people working in development for inspiration, but we also remember that good intentions are not enough. To make a difference in the fight against poverty, we need more than good intentions, more than what sounds good, and more than what looks good anecdotally. The answer isn’t always what we want it to be, and frankly that does not matter.

  We need to think clearly, ask tough questions, and set up objective processes for learning the answers. This book is a glimpse of the work that has already been done, and a mere dent in the work that remains. But the exciting and encouraging fact is that we do have some clear answers, and are on a path to many more.

  NOTES

  Chapter 1

  11 and a hero of mine. Peter Singer. 2009. The Life You Can Save: Acting Now to End World Poverty. New York: Random House. Singer uses slightly different facts. But his writing, and conversations with him, inspired this directly. I made tweaks simply to rebut a couple of standard objections to the analogy. For instance, Singer traditionally has ruining a fancy pair of shoes as the “cost” of saving the drowning child, but then one may object, “Why don’t you just take your shoes off?” For other (funny) challenges to the analogy, watch Stephen Colbert’s interview of Singer. It is available online at http://www.colbertnation.com/the-colbert-report-videos/221466/march-12-2009/peter-singer (accessed 4/26/10).

  14 areal Facebook page: http://failbooking.com/2010/02/05/funny-facebookfails-texts-cost-money (accessed 3/28/10).

  16 how it really works. Kiva is clearer now about its inner workings than it used to be, thanks largely to the urging of David Roodman from the Center for Global Development. But the truth is still in the small print. As of this writing (March 2010), the Web site says, “Your funds will be used to backfill this loan.” Then three lines below, in much larger type, it says: “Raised so Far” and “Still Needed.” Needed for what exactly? To finish the backfilling? The impression they are trying—successfully—to create is that your funds go straight to the person you clicked on. And the interest rate is not called interest rate, but rather “portfolio yield.” I wonder how many Kiva investors/donors realize that portfolio yield is the same as interest rate?

  16 over a hundred million as of November 2009. http://www.kiva.org/about (accessed 3/28/10).

  17 three times as much as the sum of all corporations, foundations, and bequests. Giving USA, a publication of Giving USA Foundation, researched and written by the Center on Philanthropy at Indiana University.

  Chapter 2

  28 here in the United States. Two of the seminal papers on randomized evaluations of social programs are (1) Ashenfelter, O. 1978. “Estimating the Effect of Training Programs on Earnings,” Review of Economics and Statistics , Volume 60, 47–57, and (2) Gary Burtless and Jerry A. Hausman, 1978. “The Effect of Taxation on Labor Supply: Evaluating the Gary Negative Income Tax Experiment, The Journal of Political Economy, Vol. 86, pp. 1103–1130.

  37 effectively in society.” http://www.un.org/esa/socdev/unyin/documents/ydiDavidGordon_poverty.pdf (accessed 2/22/10).

  Chapter 3

  39 the Slanket was born. For a discussion of the battle to
launch and spread the sleeved blanket, check out the New York Times article (among many other Internet write-ups): http://www.nytimes.com/2009/02/27/business/media/27adco.html?adxnnl=1&adxnnlx=1 269796090-dAy7Jkx4 XGUQxoRpQwit0g (accessed 3/28/10). For an amusing comparison of the Snuggie and the Slanket, and two other similar products (the Freedom Blanket and the Blankoat), see http://gizmodo.com/5190557/ultimatebattle-the-snuggie-vs-slanket-vs-freedom-blanket-vs-blankoat (accessed 3/28/10).

  41 $412 billion. http://www.intenseinfluence.com/blog/how-much-moneyis-spent-on-advertising-per-year (accessed 4/26/10).

  42 every year from diarrhea. http://www.who.int/water_sanitation_health/publications/factsfigures04/en/ (accessed 3/28/10).

  43 ashamed to eat [them].” http://people.ischool.berkeley.edu/~hal/people/hal/NYTimes/2006-06-01.html (accessed 3/3/10).

  44 Sendhil uses in many of his talks. One of the talks where Sendhil explains the Last-Mile Problem is available to the public thanks to TED. You can find it at http://www.ted.com/talks/sendhil_mullainathan.html (accessed 3/28/10).

  45 designed an RCT. Marianne Bertrand, Dean Karlan, Sendhil Mullainathan, Eldar Shafir, and Jonathan Zinman. “What’s Advertising Content Worth? A Field Experiment in the Consumer Credit Market.” Quarterly Journal of Economics, 125(1), February 2010.

  49 did a choice experiment. S. S. Iyengar and Mark Lepper. 2000. “When Choice Is Demotivating: Can One Desire Too Much of a Good Thing?” Journal of Personality and Social Psychology 79:995–1006.