E-mail updates

Sign up to receive updates from CGD:

  
Buy CGD books

Global Development: Views from the Center

« Nicholas Kristof Reviews Easterly's White Man's Burden – and Summarizes Foreign Aid Conundrum | Main | Airline Tax for AIDS Drugs Takes Off: Time for the U.S. to Sign On »

September 22, 2006

NYC Cash Transfer Plan and the Power of Impact Evaluation

Posted by Ruth Levine at 10:50 AM

When New York City Mayor Bloomberg’s anti-poverty commission recommended this week that the city pay poor people to send their kids to school and keep up-to-date on immunizations (see Increasing Economic Opportunity and Reducing Poverty in New York City the idea had an oddly familiar ring to it.) [Correction: The idea of introducing conditional cash transfers in NYC was articulated by Mayor Bloomberg as he accepted the Commission's report, not by the Commissioners themselves.] This is just the type of "conditional cash transfer" that's been implemented at large scale in Mexico, Brazil and elsewhere, often with strikingly positive results. There's little doubt that the commissioners were inspired by those experiences, whose impact has been convincingly demonstrated through careful evaluations, and documented in a veritable library of research papers and policy reports (for a start, see: From Social Assistance to Social Development: Targeted Education Subsidies in Developing Countries, by Sam Morley and David Coady, a short write-up in Millions Saved, and a new Brookings publication by Santiago Levy, one of the principal architects of Mexico's PROGRESA program, Progress Against Poverty). The south-to-north migration of this policy innovation represents a fascinating case study in the diffusion of ideas when they are backed by strong evidence.

Wisely, the commission also stated that the impact of this and other anti-poverty programs should be assessed systematically:

Evidence-based practice is the hallmark of any accountability structure. Developing an evaluation strategy must be an integral part of the design process itself, as it also aids in clarifying underlying assumptions. When evaluating initiatives New York City must strive for the highest standards and most rigorous methods feasible.

Such evaluation will be tremendously valuable to New York City policymakers, but also to those beyond the city's - and the country's - borders. As each new experience is evaluated, a body of evidence is built about what works, what doesn't work, when and why. In other words, multiple studies of similar programs contribute to the global public good of knowledge for policymaking.

It’s certainly likely that the impact of conditional cash transfers in New York will differ significantly from the effects in the very different environments of Mexico, Brazil, Nicaragua or Honduras. Some hints about how results may differ are emerging from work we are doing under the auspices of the working group on performance-based incentives in health. The working group is looking at supply-side “pay for performance” schemes and demand-side approaches in both developed and developing countries, and drawing lessons about both the programs’ impacts on health-related behaviors and the factors that should be considered in designing them. The early stage of our review suggests that targeted cash incentives can be powerful in changing behaviors in many settings, but may be somewhat less potent (particularly in affecting provider practices) in the U.S., where a broad range of other financial and non-financial incentives are at play. We’ll have a full set of findings and recommendations in a book coming out in 2007 - perhaps just in time to get on the reading list of those who are charged with designing and implementing New York City’s anti-poverty initiative.

Trackback Pings

TrackBack URL for this entry:
/mt/mt-tb.cgi/630

Comments

Thanks for bringing this to our attention. It is so easy for people to think that good ideas - for good social programs or improved governance - only flow North-South. This example shows that it can flow in both directions and that the US has much to learn from social policy in developing countries.

I would note that the example frequently cited, PROGRESA, is often described as "World Bank-supported" when, in fact, its origins and funding are almost entirely domestic. Such characterizations just further the impression that good ideas only come from developed countries.

A final comment: the NYC case is apparently going to be funded by private contributions. Why does the US increasingly turn social policy over to private philanthropists? It is an absurdity that the US cannot make and fund social policy through our normal civic and democratic processes.

Posted by: Bill Savedoff at September 22, 2006 12:12 PM

Funny -- I read the report from the anti-poverty commission, and couldn't find anywhere where it recommended paying poor people to send their kids to schools. Then I read the Mayor's remarks and he notes explicitly that the idea to try "conditional cash transfers" was _not_ directly recommended by the commission. So who did come up with the idea?

Posted by: Andrew Leonard at September 22, 2006 04:54 PM

Oops. Andrew Leonard is 100% right. The idea appears in the Mayor's remarks, at some length, but not in the Commission report itself. They do, however, make a big push for evaluation, as noted above. Apologies for the error, and thanks to Andrew for correcting the record.

Posted by: Ruth Levine at September 22, 2006 05:52 PM

Hey, no worries. This is a great site/blog. I'm delighted to have found it today.

Posted by: Andrew Leonard at September 22, 2006 08:14 PM

Conditioning welfare payments is politically attractive, especially on motherhood and apple pie interventions such as school attendance and vaccination, but monitoring and enforcing these conditions is also expensive so it is important to assess the costs and marginal benefits ex ante. Are NYC vaccination rates particularly low among the poor? Or are you paying people to do what they would do anyhow and increasing an entitlement unnecessarily?

Posted by: Amanda Glassman at September 25, 2006 01:07 PM

Great post. Levy's Progress Against Poverty Book is available free online. Also there are more links about Progressa and CCT programs here.

Posted by: Azad Amir-Ghassemi at September 26, 2006 02:29 PM

Today's Financial Times (9/29, Wheatley and Lapper) credits Brazil's conditional cash transfer program Bolsa Familia with Lula's probably re-election this Sunday with this sub-title: "hand-outs have helped put the president on course for re-election..." These programs are indeed politically powerful, but I am troubled to see them described as hand-outs. Rigorous impact evaluations in 4 Latin American countries provide unambiguous evidence that CCT reduce poverty, increase schooling and improve nutritional status. With respect to Brazil, according to older papers on the program, the program performs less effectively with respect to nutrition (Morris et al 2003) but has recorded gains in schooling for poor children (Bourguignon et al 2003) and these human K gains are essentially irreversible -- another year of schooling completed is associated with higher returns in the labor market. And the Brazil program is now targeted to the poor using a more transparent mechanism. Newer evaluations being undertaken by IFPRI should tell us more, but providing conditioned transfers that improve the human capital of the poor should not be described as "hand-outs".

Posted by: Amanda Glassman at September 29, 2006 05:23 PM

Great idea this - CCT - Can this be tried on a pilot basis, in some of the developing countries in Africa?

Posted by: Richard Ngetich at October 6, 2006 01:29 PM

Richard, great question. In our analyses so far of CCT programs in Latin America, we've found that it's often desirable, if not necessary, to have a certain level of supply of services (both quality and quantity) available to respond to the increase in demand created by the cash transfers. This may be a dilemma in particulary poor or remote areas where there are serious supply constraints. On the other hand, it may create the desired incentives for a positive supply response. There is little concrete evidence so far to tell us which way it would go--yet another reason to pilot and rigorously evaluate!

Posted by: Jessica Gottlieb at October 19, 2006 11:32 AM

Post a comment




Remember Me?

(you may use HTML tags for style)