Cash on Delivery: a big new aid idea? Actually, the EC’s been doing it for years!

August 6, 2009

     By Duncan Green     

One of the more exciting proposals in the UK Conservative’s recent Green Paper on development (see previous post here) is the idea of making aid ‘Cash on Delivery’ (CoD). ‘We will commit to pay a certain amount to a recipient government for a specific measure of progress – for example £100 for every extra child who attends school, or for every extra mother who gives birth in a proper medical facility’.

CoD has a number of potential advantages: it makes aid easier to sell to a sceptical public back home, and by focussing on results, it gets away from the ineffective and discredited practice of donors prescribing specific policies via ‘conditionalities’ on their loans – a government can choose to reduce maternal mortality any way it likes, provided it achieves results.

A couple of alarm bells, however. Rewarding success could further polarize the gap in resourcing between ‘donor darlings’ and the rest. And it would not make sense only to pay on results, not least because governments need some cash upfront in order to achieve the improvements. Thankfully, the Tories recognize this, stating ‘we will need to ensure that developing countries are able to finance the up-front investments necesary to achieve the desired outputs’.

A lot of the thinking behind this proposal has been done by the Center for Global Development in Washington DC – kudos to them. But it turns out that for some years now the European Union’s aid programme, traditionally viewed with considerable scepticism (not least by many eurosceptic Tories) for supposedly epitomising everything that’s bad about aid, has actually been doing some pioneering work in this area. These days the former laggard of international aid looks more like a candidate for ‘most improved player’.

In May 2008, the European Commission (that’s Europe’s official machinery, as opposed to its national governments) introduced ‘MDG contracts’. These link at least 15% of the EC’s direct budget support (i.e. money that goes straight to governments, rather than to specific projects or ministries) to performance in achieving the Millennium Development Goals. Not only that, but they’re doing it at scale – the EC is the world’s biggest multilateral aid donor (bigger than the World Bank), and a third of all its aid is in the form of budget support – tens of billions of $ a year.

The MDG contracts build on the previous use by the EC of what they call ‘outcome conditionality’ (see here for a review by Eurodad). The big difference is that MDG contracts run for 6 years, rather than the 3 years previously covered. That’s a real victory for predictability: once a government signs one, it can go ahead and hire those 10,000 extra teachers it needs, without fears of an unfunded wages bill hitting it in a couple of years’ time when the aid runs out.

A 2008 paper summarizing some pretty detailed research for Oxfam was surprisingly complimentary by the grudging standards of us hard-to-impress NGOs:

‘Oxfam’s research shows that over half of the performance indicators tied to the Commission’s general budget support agreements call for direct improvements in poor people’s health and education, in particular for girls and women, who often carry the heaviest burden of poverty.

Oxfam’s research also suggests that this aid does help to make a change in poor people’s lives. Government spending on education has increased by nearly a third (31 per cent) in eight of the countries that receive some of the largest amounts of the Commission’s general budget support. In all but one country (Rwanda), this has resulted in an increase in the number of children enrolled in primary school. In Madagascar, the proportion of children enrolled in primary school increased from 69 per cent in the period 2001 to 2002, to 92 per cent in 2005. Of course the Commission is not exclusively responsible for these positive results, but the evidence does show that where it is giving large amounts of budget support, headway is being made in reducing poverty.’

A case study from Malawi by my colleague Max Lawson shows how this works in practice: a small amount of aid was cut due to immunization coverage not being up to scratch. The amount was small, but the political attention generated large – a good result. According to Max, what governments complain of is that they currently have to jump through different sets of hoops: policy conditions set by the World Bank and outcome hoops set by the Commission. That absorbs yet more staff time.

So how is CoD different from MDG contracts? It talks in terms of $ per student rather than overall amounts, but that’s just packaging. MDG contracts combine a CoD element with more traditional budget support, and it would be important to ensure that a similar half-way house approach is replicated in any CoD scheme. Can people point out any other major differences? In the end, of course it doesn’t really matter where the idea came from, provided it works. After all, as the saying goes, success has many fathers, (but failure is an orphan).

Update: see here for an instant response from Nancy Birdsall, William Savedoff and Ayah Mahqoub at CGD – they’re quick workers these guys – setting out their view of the differences between CoD and MDG contracts.

August 6, 2009
Duncan Green