Can aid promote democracy and improve human rights? This is an important question and some ingenious statistical sleuthing from researchers at Yale University has produced evidence suggesting that its answer may well be yes, albeit with qualifications.
While it’s important, the question of whether aid helps, hinders or has no impact on human rights and democracy is not an easy one to answer. In theory, aid could be helpful – perhaps though donors using the threat of aid withdrawal to force improvements, or maybe through the funding of civil society – but there are good reasons why it might not be. Where aid is given for geopolitical reasons donors may simply ignore rights abuses when doling out the cash. And even when aid is given with good intentions, leaders on the receiving end might ignore donors’ demands. Aid might even make things worse by reducing recipient governments’ dependence on domestic tax revenues, assisting repressive leaders to stay in power by freeing them from the constraints of their domestic social contract. All this theoretical uncertainty contributes to the practical dilemma of whether we should be giving aid to countries with poor human rights records at all (something that we’ve debated before on this blog: see Dinuk’s excellent post here and my contribution here).
Empirically, one way of answering questions about aid’s impact on democracy and rights is statistical analysis using multivariate regression on cross-country data. In an era of increasingly good data on aid, and on levels of democracy and human rights protection, this is a potentially powerful too, but also one with limitations. Most critically, distinguishing causality from correlation. Simple regression analysis may be able to show us that aid and human rights (or democratic governance) are correlated but it can’t tell us what’s driving the correlation. If, for argument’s sake, we find that aid levels are positively correlated with the protection of human rights, we still don’t know whether this is because aid improves human rights or whether it’s merely because donors give more aid to countries with better rights records. This is a vexing technical problem that bedevils all manner of quantitative research on international development.
Yet there are potential ways round it. And in the case of aid’s impact on democracy and human rights, the Yale researchers (Peter Aronow, Allison Carnegie and Nikolay Marinov) have come up with a clever piece of analysis focusing on aid from the European Union.
In a recent working paper (and caveat lector – it is only a working paper, yet to be published in a peer reviewed journal, but it seems sound) they use an instrumental variable approach to isolate causality amongst correlations between aid, and democracy and rights.
The paper focuses on EU aid and uses the rotating presidency of the EU Council as its instrumental variable, which works as follows. The presidency of the EU Council rotates amongst member states, and when a country is president of the Council, for no reason other than the fact than the country is president, its former colonies get more aid. This increase has nothing to do with the existing state of democratic governance and human rights in these countries, it’s completely independent – merely a product of the rotating council presidency. And ascendancy to the head of the council is unlikely to have an impact on democracy and rights in former colonies other than through aid. For these reasons, the relationship between ascendancy of any particular EU country to presidency of the council and any subsequent changes in human rights and levels of democracy in former colonies almost certainly has to have been caused by changes in aid.
Making use of this the paper’s authors then undertake regression analysis which produces good evidence to suggest that increases in EU aid lead to improvements in Human Rights (as measured on the CIRI Human Empowerment Index) and democracy (as measured by recipient countries’ Polity IV score). The results aren’t overwhelming, but the impact is positive and of a meaningful magnitude (see the bottom of page 14 on the working paper). EU aid very likely helps promote human rights and democracy. That’s the good news.
The bad news? Well the first bit of bad news is surely the fact that this study could ever have been undertaken. Don’t get me wrong, it’s a fascinating paper, but it is depressing to learn that, at least up until 2007 when the rules of council presidency changed (and quite possibly since then too), in part, receipt of EU aid wasn’t a question of need, or impact, but rather happenstance in the form of EU Council Presidency. This is most definitely not best practice.
The next bit of bad news is that while the paper shows aid causing improvements it also shows improvements waning once EU Council Presidency induced aid inflows taper off in former colonies. Aid seems to have an impact but it’s not transformative in that its legacy after the aid has gone is negligible. (This is particularly for human rights but also for democracy albeit more slowly.) This is sobering but not that surprising, particularly given that the increases studied are only short term bursts associated with Council presidency. And while it should provide caution against expecting too much from aid, it is not the same as saying aid doesn’t work. As I’ve written previously on this blog, we have and will be giving aid for a long time, and improvements in people’s quality of life are still improvements even if they require on-going assistance.
The final bit of bad news is that, thanks to Europe’s economic travails, EU aid budgets appear set to stagnate or decrease in coming years. Which could well mean worsening human rights and less democracy in those countries that loose out.
The paper isn’t perfect. There are lexical errors: the authors conflate good governance with democracy yet the two concepts, while related aren’t the same thing, and at one point they conflate the concepts of development and economic development. And it’s a little surprising that the study doesn’t include economic growth and governance (as conventionally defined) amongst its outcome variables (the data are available). Also, in the long run it would be excellent to see some case studies using process tracing to shed more light on exactly how aid helps improve rights and democracy. Nevertheless it’s an important contribution, albeit one that suggests a whole new set of sad consequences of Europe’s economic malaise.
Terence Wood is a Research Officer at the Development Policy Centre and a PhD student in the State Society and Governance in Melanesia Program at ANU. Prior to commencing study he worked for the New Zealand government aid program.
Leave a Comment