Show ’em the money! – Trialing conditional cash transfers in schools

Across Latin America, governments are turning to conditional cash transfers to overcome poverty and inequality. In a recent post, Andrew Leigh asks whether we should trial the approach in Australia.

Conditional cash transfer programs attack poverty in two ways. Like income support programs, they provide poor families with money. This alleviates poverty in the short term. And by using that money to leverage changes in behaviour, the programs develop human capital to support greater self-sufficiency in the long term. For example, under Mexico’s Oportunidades program parents can only receive payments if they send their children to school, get regular health checks and attend health and nutrition classes. Policy makers hope that by improving the health and education of the next generation, disadvantaged communities will be better able to overcome poverty.

So far the results have been positive. New York Mayor Michael Bloomberg was so impressed that he started his own pilot program — Opportunity NYC. "Historically, the rest of the world has often looked to America for leadership in social policy" said Bloomberg, "but there is no reason that we cannot also learn from the experience of others."

Bloomberg’s program has three components — Family Rewards, Work Rewards and Spark. Like Oportunidades , Opportunity NYC programs make payments to parents that are conditional on their children’s attendance at school. But Family Rewards and Spark also make payments to children. Opportunity NYC has a stronger focus on student test results than most Latin American programs.

Spark works with students in grades four to seven. Students are take ten tests over the course of a year and earn cash payments if they do well. Fourth graders can earn $25 for each test and seventh graders can earn $50. There are also rewards for groups to encourage collaborative learning. But as Andrew points out, not everyone is convinced that the approach will work. Some worry that it focuses on student effort when the real barrier is the quality of teaching. Others argue that paying children to perform on tests will undermine their intrinsic motivation to learn.

But despite these concerns, Leigh argues that offering incentives for school attendance and performance is worth a try:

… in places where educational attainment is low and stagnant (eg. US inner cities), it’s worth rigorously testing on a small scale. The downside risk isn’t zero, but if the intervention works, the societal benefit is potentially very large.In Australia, conditional cash transfer programs could theoretically be implemented in a low-income school anywhere in the country. But the most likely context is in Indigenous education, where we know that Indigenous children are one year behind when they arrive at primary school, and two years behind when they finish primary school. Anyone game for a randomised trial?

At least two components of Opportunity NYC are being tested using a randomised trial approach. MDRC, a an independent research organisation, is conducting a five year impact evaluation of the Family Rewards and Work Rewards components of Opportunity NYC. The American Inequality Lab and Harvard University Professor Roland Fryer are evaluating Spark.

Adding insult to injury?

Opportunity NYC is based around the idea that disadvantaged Americans fail to take advantage of readily available opportunities. In City Journal, Nicole Gelinas argues that the program is based on the assumption: "that poor New Yorkers are so unable to act in their best interest that they will not even take advantage of an existing, well-publicized government program without the promise of additional short-term cash." According to Gelinas the initiative insults the people it is supposed to help.

Some students felt the same way. According to the NYC Student Union, many are unhappy with Spark’s emphasis on test results and cash payments. "It insults hard- working, low-income students by conveying the message that they could not possibly value education in itself and must need some sort of incentive in order to perform better in school,” one student said.

Roland Fryer, the Harvard economist who created Spark, has heard this criticism before. Even before Opportunity NYC was on the drawing board, Fryer was running his own pilot experiment in the Bronx. According to Stephen Dubner of the New York Times, the idea took off from there. The chancellor of New York’s public schools, Joel Klein met Fryer at a dinner party, and when he heard about the experiment he suggested expanding it to include more schools.

Fryer met with a group of 15 school principles and pitched his idea:

The principals began to grill him. Even if the kids do respond to the cash incentives, one principal asked, what happens next year, when they aren’t getting paid? Won’t students in other grades be resentful? What will parents think when their kids start receiving cash in the mail every few weeks?

Fryer addressed each issue as best he could. But one question kept coming back at him: if we start paying students to test well, aren’t we sending the message that learning is not its own reward? Although the exchange flustered him, Fryer had by meeting’s end persuaded the principals to take part.

Fryer later told Dubner, "’I’m troubled by the fact we’re treating kids as inanimate objects … They have behavior, too. They respond to incentives, too.” In an article for Time, Fryer argued that many parents reward their children for good grades but that some families can’t afford to. He writes: "I think that kids will find that doing well is fun, and they will be inspired to work harder for their whole education."

But will it work in theory?

Interestingly, some of Fryer’s other research casts doubt on this theory. In a paper titled: ‘A Model of Social Interactions and Endogenous Poverty Traps‘ (earlier version here), Fryer explores another reason why children from disadvantaged ethnic minorities might fail to take advantage of educational opportunities.

According to Fryer, young people have a choice between investing in general human capital (the knowledge and skills taught in class) or investing in cultural capital specific to their group. Because of the advantages education offers in the labour market, students who focus on human capital are less likely to remain in the community. As a result, other young people in the community are less likely to include them in relationships that involve long-term reciprocity. In terms of long-term cooperative relationships, students who are likely to leave the community are a poor investment.

In order to avoid being excluded from the community, it makes sense for students to invest in group-specific cultural capital and avoid investing in general human capital. If this is right, then it’s not obvious that financial incentives will solve the problem.

Last year Andrew Leigh argued that policy makers should follow Franklin Roosevelt’s example and keep trying out different policies until they find something that works. But a more effective strategy might be to test the theories behind the policies. Too often, policies are a ‘black box’. When they succeed, we don’t know why. And because we don’t understand how policies work, we don’t know how to adapt them to new client groups or social contexts.

Experimenting with policy takes time and money. The MDRC’s evaluations of Opportunity NYC will take five years to complete. Add this to the time and effort it takes to build up a consortium of donors, policy makers and service providers, and you’ve got a significant investment. Leigh is right to argue for more experiments in policy, but he also needs a way of identifying those policies which are most likely to succeed.


‘Veronica White (CEO), Andrew Phillips (Seedco), James Riccio (MDRC) ‘Testing a Conditional Cash Transfer (CCT) Program in New York City‘, Work Bank Seminar September 17, 2007.

The Education Innovation Laboratory, Harvard University (includes video of Roland Fryer explaining Spark — from CNN’s Black in America).

Michelle Morais de Sá e Silva, ‘Opportunity NYC: A performance-based conditional cash transfer programme. A qualitative analysis.’ International Poverty Centre.

Club Troppo:

US welfare reform — beyond sticks and sermons, October 2006.

The Rise of the Supernanny State, September 2007.

This entry was posted in Uncategorised. Bookmark the permalink.
Notify of

Newest Most Voted
Inline Feedbacks
View all comments
Joshua Gans
15 years ago

I assume the rewards are for absolute scores in the tests rather than imnprovement. In that case, the scheme is likely to reward ability as much as effort (and parhaps more so in fourth grade). Kids from the better resourced of the households that qualify will tend to bring more skills to the test, and get more than their share of the rewards. So it would be good if rewards were based on improvement from one score to the next. Only, that would create an incentive to do bomb out in the first test.

In general, I have a gut reaction against material incentives for children. But I suppose that’s a culturally specific, romantic prejudice. Many if not most of the asian kids in my son’s Year 6 class last year were promised $500-$1000 if they got into selective schools.

Don Arthur
Don Arthur
15 years ago

James – My understanding is that Opportunity NYC components are trialing both rewards for improvement and rewards for proficiency.

The Family Rewards component pays parents of elementary and middle school students when their children achieve proficiency or improvement. High school students can earn a payment if they achieve a passing grade in the Regents exam.

I think Roland Fryer’s Spark component pays for performance rather than improvement.


[…] Show em the money! – Trailing conditional cash transfers in schools […]


[…] Update: