Cash on delivery aid: changing the incentive mix 
Author : Jean-Michel Severino
Date : December 22nd, 2008
I recently came across a post by Nancy Birdsall on a new instrument promoted by the Center for Global Development, called ‘Cash on Delivery’ aid. In a way (Nancy, whose Center is a partner of ID4D, might wish to correct me if I am wrong, or complete if need be) it is conceptually close to ‘output-based aid’, except that governments rather than non-governmental actors are in charge of delivering the output, and that aid comes as a form of reward upon delivery rather than a direct payment of the costs incurred. The novel philosophy in this approach is that donors are as ‘hands-off’ as possible - leaving it up to recipient countries to find the most appropriate ways to reach given objectives. In a nutshell, under this type of program donors would commit ex ante to pay a specific amount for a specific measure of progress on a specific objective. In education, for example, donors could promise to pay 100€ or so for each additional child who completes primary school and takes a standardized competency test.
I find this concept appealing in three important ways:
- first, it is serious about ownership: by targeting a specific goal but not interfering in the policy choices, it leads governments to think through the most appropriate ways to reach this objective in their specific national context. Experience has shown that donors don’t always know best.
- second, it shifts incentives away from inputs (how much money do we give?) towards outputs (what concrete results do we achieve?), and clearly places the onus of results on governments. If pilot projects are set up, they will teach us important things about how public choices are made within governments, and to what extent they influence performance. For one thing, there is likely to be more public scrutiny and debate on the decisions made to ‘win’ the challenge - which may be a positive result in itself.
- this leads me to a third interesting dimension, which is the underlying ambition of changing the structure of the incentives. Cash on delivery works on supply-side incentives. Other initiatives (targeted at young girls in India, but also at public school students in Washington DC!) have worked on education through the demand-side, by paying children to attend and get good grades. Comparing the successes of demand-side and supply-side incentives in the field of education could teach us a lot on the determinants of progress.
Yet I also have my concerns (aside from obvious technical difficulties of finding agreed and verifiable benchmarks of ‘success’, which are important to think through but can probably be overcome):
- how will the governments attain results if aid isn’t there in the first place? This is a typical ‘chicken and egg’ problem: some countries do not have the budgetary space to ‘invest’ in the big push that would be needed to attain the objectives initially. These programmes would therefore need to be complementary to existing programs (such as the Fast Track initiative in education).
- secondly, Cash on delivery is a form of (ex-post) conditional aid. As a reward to good performers, it risks imposing a double penalty on populations whose governments are incapable of delivering. And by penalizing ‘poor performers’, it makes the implicit assumption that they are responsible for their bad results. Although this may be the case, it cannot be taken for granted. Like many good ideas, it should therefore probably be consumed moderately; we cannot hope to apply it uniformly and across the board.
In any case, this is interesting new stuff, which definitely merits to be discussed in the columns of this blog. Dear fellow blog members, guest commentators and readers, the floor is yours!
JMS
German
Spanish
French
Italian
Portuguese
Japanese
Korean
Russian
Chinese
saurabh pandey
Date : March 18th, 2009 01:38:34
cash on delivery or output based aid is a promising concept for developing countries. it surely will give professional and accountable outcomes.
Carmen Nonay
Date : February 3rd, 2009 02:53:43
COD is a very promising approach and we look forward to seeing how it would be piloted in developing countries. I would agree that it is conceptually close to output-based aid (OBA) but there are some differences.
OBA in education uses explicit output- or performance-based subsidies to complement or replace school fees for poor students. This involves the contracting out of education services to a third party (service provider), with subsidy payment tied to the delivery of previously specified learning outputs. This implies that the service provider assumes the risk of pre-financing the achievement of the agreed outputs.
The main differences between OBA and COD are that OBA pays for outputs while COD pays for progress towards outcomes; and OBA usually either disburses funds directly to service providers or passes money through national governments based on an existing contract between governments and service providers. Also, OBA usually is explicitly used to bridge a pre-existing funding gap that prevents the poor from accessing basic services in education . As I understand it, COD would pay mainly to the governments and would be “hands off” in the sense that the government only has to demonstrate what it has achieved and does not have to document to donors how it used their money.
The Global Partnership on Output-Based Aid (GPOBA) has identified a handful of OBA schemes in the education sector (plus some conditional cash transfers which are not OBA but can be complementary to OBA). The challenge of introducing OBA in this sector has often been defining the appropriate “output” which balances achievement of results with reasonable performance risk transfer. GPOBA is looking at ways to develop more pilot OBA projects in education. We have had some interesting discussions on this - and COD - with both the CGD and the EFA-FTI Secretariat, as well as our donors.
GPOBA is planning to publish a short paper on OBA in the education sector later this year. For more information, please check our website where you can also sign up for our quarterly e-newsletter: http://www.gpoba.org.
Carmen Nonay
Senior Infrastructure Specialist
GPOBA
World Bank
Dilip Kumar
Date : January 20th, 2009 11:12:31
Cash on delivery aid: It sounds a novel concept. Many times developments are mismanaged, either due to deficiencies of the administrative system or due to lack of will… I feel, this type of ‘Cash on delivery aid’ can atleast generate the will, the will to perform, the will to prove. It is a general tendency, unless one’s performance is being monitored, no one tries to improve.
The author rightly pointed out “it is the population that is going to lose due to the inefficiency of the government” . I feel that government should be scrutinised by the public, the public should know that they are loosing their fund because of the government’s poor performance. There should be sound public auditing system. The outcome should be measured by the public and at the end the public is the one capable of questioning the Government(In Democracies Like India).
In India, a scheme was started by the government in which unemployed people who wish to do unskilled work are assured a wage of 100 days. In this scheme initially there were lot of hiccups, taking into consideration the huge amount of administration involved. But with time, in some states a profound public auditing system was established, which made the scheme work and the poor were finally able to get benifits.
When funds are meant for public benifit, they should also be part of its management. Then only I feel Inclusive development can take place.
Date : January 9th, 2009 06:46:32
Jean-Michel Severino raises good questions about our Cash on Delivery (COD) Aid idea [1]. Let me comment on each briefly in turn.
First: How can revenue-strapped governments get results without upfront aid (COD payment comes only after the fact)? Severino is right that COD Aid (let’s say for Tanzania, as President Kikwete wants to see a donor do a pilot there) has to be complementary (or additional, but I like the word complementary used by Severino better) to existing programs and aid flows – if in fact countries don’t have the upfront resources to try new ways to increase enrollment and learning (quality). At our side event in Doha [2] (at the UN conference on Financing for Development in December attended by developing country officials, NGOs, and donors), this emerged as the biggest concern of officials from developing countries. We think there is also the possibility that governments can get more bang from their current bucks, including by increasing efficiency in their public bus transportation or road maintenance for example (to enable higher secondary school attendance), or by policy or program changes that don’t cost anything (competitive cost-saving procurement of textbooks). But practically speaking, it’s true that COD Aid is meant to supplement not substitute for current aid and government spending on education. In fact in 2004 when Owen Barder and I wrote our “Payments for Progress” paper [3], we were looking for a way to help donors scale up aid without hitting diminishing returns! Sadly that view feels outdated today – there is a real risk that total official aid flows will decline [4]. On the other hand, we think that new philanthropic and other private aid flows, which Severino with Olivier Ray highlight in a forthcoming draft paper entitled “The End of ODA”, could be triggered by results-driven aid vehicles like COD Aid. The Gates Foundation for example is exploring performance-driven approaches for aid to health. Some of their findings are included in a paper by Dan Kress and Paul Shaw entitled “A New Aid Paradigm for Donor Assistance in Health? The Appeal of Results Based Financing”.
Second: COD Aid won’t work where governments are not “good performers”. On this I have to disagree at least in part. Of course it won’t work where government is corrupt or leaders have no commitment to improving education. But in many so-called “fragile states” the leadership is responsible and committed and COD Aid could create the incentive that would help jumpstart a process of sensible policy changes across government as well as healthy demand for technical cooperation from donors on measurement, testing, and implementation of program initiatives.
Still, Severino is right that COD Aid should be “consumed moderately” and not applied across the board. In fact we think practically speaking the next step almost has to be thinking through how COD Aid would work in one or two specific country settings.
One other (minor) comment: Severino refers to “outputs” as in output-based aid. I think it’s useful to distinguish outputs (such as number of schools built, number of teachers trained) from “outcomes” – which in the case of schooling would be amount of learning a child has achieved. COD Aid is ambitious in measuring and paying for an “outcome” — at least that is the idea.
For more on all these issues we hope ID4D readers will order our forthcoming book: “Cash on Delivery: A New Approach to Foreign Aid with an Application to Primary Schooling” by Nancy Birdsall, William Savedoff, and Kate Vyborny, as well as visit our COD Aid initiative site – including for updates on the progress toward practical application [1].
Nancy Birdsall
President
[1]http://www.cgdev.org/section/initiatives/_active/codaid
[2]http://blogs.cgdev.org/globaldevelopment/2008/12/cash_on_delivery_aid_developin.php
[3] http://www.cgdev.org/content/publications/detail/11550/
[4] http://blogs.cgdev.org/globaldevelopment/2008/10/history_says_financial_crisis.php
Silvia Mancini
Date : December 29th, 2008 11:31:32
This could be a winning approach.
I also share your two concerns with regard the lack of cash to achieve the objectives and the incapacity of weak governments to delivery.
Potential solution to your two concerns are
1) The achievement of objectives can only be based on steps within a programme log frame. The first objective (not too expensive!) will be designed according to the budgetary capacity of the government. Once the first small objective is achieved government gets the cash back plus a reward which could range from a small grant (from any IFIs) to a facilitated access to financial markets. These additional incentives would support the achievement of further objectives already planned in the programme log frame.
2) The mutual accountability framework (http://www.oecd.org/dataoecd/36/53/40459701.pdf) could be an answer. However in the context of the mapping of accountability framework for donor and partners performance, great importance has to be given to the high quality regular monitoring of the government performance. I suggest the latter would be the responsibility of a third party, most likely a private organization, which would provide an on going counselling on results achievements and evaluation of results.
The costs of the third party would initially be charged to the donor community.
Phillip Huggan
Date : December 8th, 2008 09:21:27
This approach has much potential. It has been used in Mexico in some application, food programmes for families that send their kids to school, or something like that. In the USA, there is a doctor paying obese patients a dollar for each pound of weight lost. It works. For the latter I guess the reward is psychological. Winning a prize is fun no matter the actual value. For the developing world the “prize” is much more tacit: basic human goods.
In the USA somewhere, high school students are giving $50 for earning a high grade.
This approach is capitalism at its best given funding. I’d think philanthropists would jump on board if they knew it was effective aid, because it allows one to cherry pick recipients and applications.
Che Thuy Nhu
Date : December 7th, 2008 10:39:15
Dear readers,
In the global financial crisis, Cash on delivery aid: changing the incentive mix is a necessary problem that will be discussed on BLOG.
In Vietnam we have 2 ways to use money
1 – The donors let Vietnamese PMU team spend money and collect bills (Cash in hand). PMU have to answer for expenditure before the Donors and the law inside and outside of Vietnam. This way often happens with budget from Vietnamese government. In TA project and ODA.
2- The donors, base on the agreement, send money directly to working teams .The Vietna mese team carry the independent monitoring and registries the progress. The donors answer for directly payment. In LOAN and other projects.
Some time 2 ways are using in one project: part by donors, part by PMU Vietnmese.
I think, in this new post Mr Severino give example in education, it is the only concrete case
The question about Cash on delivery aid: changing the incentive mix need to be discussed in other projects.
In Forum http://edu.net.vn of MOET Vietnam I suggest an idea to help the pupils in poor area. The poor families can’t pay for fee in school. So why the drop–out rate is high? When I was in Kunminh ( China ), they had an experimentation: Principal of school established the list of poor pupils, that can’t pay for cost fee. The need is about 50 USD for 1 child/1 school a year. For example: In Vietnam it is about 1 USD/1month x 9 months for class 6 + other needs (I am still collecting concrect) but I think this can’t be higher than in China.
The support should be send directly to account of School as payment for education. Parent give food for children. Principals have to answer for the effectiveness and send the results of pupils to donors ( Vietnamese and other form abroad ) at the end of the years (By Email – Internet will be in all schools in Vietnam in future or use the Internet of the Cultural–Post office in the commune). The report is also send to Provincial and Ministerial levels by E- Government (one click). Principals answer for correct of request. If something goes wrong he or she have to payback from his or her salary. One child receive only 1 support.
Delete all other Funds for cost fee in schools. Small sum of money will ensures one child to pass 1 class, ensure teachers to receive payment for their work, reduce the drop–out rate in remote area.
I think with Internet the fund can be go directly to Schools,with no need through other channels, other local funds.
Thank you