(A fuller version of this argument is in the pipeline and will appear here in due course. Meanwhile, comments and suggestions are welcome – please post your responses here or tweet @cds_Bath using #brandaid)
The world of international development aid was never simple, but it seems to become ever more complex as agencies, financing mechanisms and acronyms proliferate. Public understanding struggles to keep up, with debate often pitched at a depressingly bland level. Is aid working? The correct answer, of course, is that aid comes in many different forms and brands. More interesting questions then abound, like which sorts of aid works best, when and why, and is the mix of different forms of aid right in different contexts? My proposition is that distinguishing between different forms of aid more clearly can contribute to raising the quality of public debate about its effectiveness.
Many of us choose daily between thousands of supermarket products – too many perhaps. When John Heinz chose to market 57 products in 1896, did he imagine, that the number of products available under the Heinz brand would eventually grow to 5,700 and counting? In the aid hypermarket, how many different forms of aid can you distinguish between, and how easy is it to decide which ones are better? Branding is becoming more widespread in the aid world. One example was the 2012 decision to add the Union Jack to the UKaid logo and to emulate USAID’s “from the American People” tag line. Another is the consolidation of leading global NGO brands, entailing for example the merger of seven different Save the Children programmes in Ethiopia. Driving through the Rift Valley not long ago, I passed the time by ticking off the aid logos along the roadside: a European Union seed project, a South Korean aided model village, an Indo-Ethiopian sugar mill, a USAID sponsored dairy, and so on…
Such advertising may be viewed both as a modest step towards greater aid transparency, and as more than symbolic of the way external aid undermines country ownership of its own development. In her celebrated book No Logo, Naomi Klein railed against branding as another example of the rise of corporate power and the privatisation of public space. But branding is also an important mechanism for overcoming information overload, signalling quality, enabling us to differentiate between products and putting corporate reputations on the line. Much depends on how much independent evidence is available through which consumers can make informed judgements about which brands they like and why.
So what about aid? How can we advance understanding of the quality of its many forms? Much work is going into improving aid statistics, so it is clearer who gives what to whom (e.g. the work of Development Initiatives). There is also scope for more independent and consumer watchdog style assessment of the quality of different aid brands – for example, why does the UKaid ‘supermarket’ have IFAD’s services on its shelves but not the ILO’s? The All Party Parliamentary Group on International Development and Environment plays an important role in this area, and we now have the heavyweight Independent Commission for Aid Impact. However, with the myriad of different (and often confusing) financing mechanisms that make up aid (programme aid, project aid, challenge funds, guarantee funds, technical assistance etc.), there is also a need for more clarity about deeper characteristics that differentiate aid and aid brands.
To be more specific, I propose we start by making a stronger distinction between two different dimensions in which I believe aid can be assessed. First, there is social impact arising from how much any particular aid activity adds to the wellbeing of intended beneficiaries. Aid is nearly always channelled through partner agencies, so social impact also depends on their costs and on how much value they add, e.g. converting cash into appropriate health or education services. Second, there is financial sustainability, or the extent to which local aid partners recover some of the aid money by charging intended beneficiaries or end users for the services they are providing, and thereby reducing their own aid dependence. The diagram shows these two dimensions as the X and Y axis on a graph. In an ideal world, we would like to deliver aid that performs well on both counts (i.e. the top right of the diagram). But in practice we face a trade-off determined by the range of aid technologies available at any time.
Within this constraint, we can distinguish between two extremes of aid. Development assistance (top left) often does well on social impact, but poorly on financial sustainability – in other words aid has to keep flowing to sustain the benefits to end users. Conditional cash transfers are a fashionable example of this, underpinned by the ‘just give them the money’ line of criticism of more heavily managed aid. Development finance (lower right) in contrast, allows the partner and/or donor to recover more money (indeed it may be profitable) but with less benefit to end users. Microfinance is an example of this – somewhat less fashionable now than it was a few years ago.
So how does this help us with a Which style analysis of aid effectiveness? Consider a particular aid agency or project you know well; does it emphasis development assistance, development finance or something in between? How sure are you of this? How helpful is the donor in informing you about how this brand of aid is expected to perform in these two dimensions? Take the simple example of a charity appeal to a comfortably off person in country A to buy a goat for a poor person in country B. The primary appeal to the donor is emotional, and it is mostly assuaged by the simple act of giving (or what economists call ‘warm glow’). This appeal may also be strengthened by the suggestion that the goat will breed and that the offspring will be returned to the local partner for passing on to other poor people. However, we generally demand or receive relatively little evidence about what actually happens, relying instead on the overall brand reputation of the aid agency. This reduces the incentive for donors to invest in telling us how their activities aim to perform in these two dimensions, as well as finding out and sharing with us how they do.
Differentiating between types of aid may also help identify trends (political or otherwise) in the way aid is being disbursed. It’s no accident perhaps that the rise of development finance in DFID’s portfolio – such as increased use of funds to encourage UK businesses to invest in development projects – started under New Labour and has gathered pace under the coalition government (not without critics).
I’ll finish with a comment on measurement. The diagram may seem at face value to be encouraging greater precision in measuring social impact and financial sustainability. Of the two, financial sustainability is perhaps the easier to measure, although even that is not easy (it depends on the scale of operation of the local partner relative to the value of the aid supplied, for example). Measuring social impact is much harder, particularly when aid affects multiple dimensions of our well-being, and these impacts are durable (think of an extra year in primary education, for example). There is much more to say about this – indeed it has recently been the subject of a highly informative and entertaining policy wonk war on evidence. But the single most important point I want to make is that being clearer and more precise about definitions and meanings does not necessarily also entail having to be more precise about measurement. This is where branding comes in again. The process by which we assess the quality of a brand is complex and involves the synthesis of lots of evidence, likewise with aid. I think we do need more and better evidence about sustainability and impact, but as Ros Eyben and her colleagues in the Big Push Forward keep reminding us, it is madness to be overly fixated on precise measurement of results. Aid observers (advocacy groups, consultants, journalists, academics, official evaluators, bloggers…) have a rich array of qualitative, quantitative and mixed methods to draw upon and should make use of the full range.
To sum up my argument: aid has become more complex and we struggle to keep up. Distinguishing development finance from development assistance can help. This entails assessing aid projects for both social impact and financial sustainability. One disincentive to this is that donors may find it easier to ‘sell’ aid to the public by conflating them. Reliably assessing social impact is also difficult. But ‘good enough’ impact evaluation does not necessarily require precise measurement, and can help to improve the way the reputations of different brands of aid are established.