A Couple of Provocative Thoughts

Don't provoke him. You wouldn't like him when he's angry.

A few weeks ago, Nick Eubank wrote a piece about Somaliland in the Guardian’s Poverty Matters blog. Somaliland is an interesting one – it’s a state that is not officially recognized and therefore in receipt of less aid than would be expected (though it is incorrect to assert, as he does, that it receives no foreign aid), but is performing well above reasonable expectations. They have a pretty well-functioning democracy, biometric passports – like Lee, I still can’t get over this – and pretty good economic performance.

Eubank suggests that it’s the very lack of aid and subsequent development of tax structures that has led to Somaliland’s good performance, a view to which I have some sympathy. I’ve written in the past about why developing taxation systems is crucially important for all developing countries and why it tends to be difficult. If a side-effect of Somaliland’s lack of access to aid is being forced to be more representative and more developmental, could there possibly be a benefit to restricting rather than increasing aid volumes?

It’s a provocative idea, because it flies in the face of almost everything we’re led to believe about development and levels of development funding, but as a thought experiment, it’s worth considering. It would be likely to yield specific benefits:

  • Most obviously, it incentivizes the creation of a viable tax regime with the benefits of accountability, representation and responsiveness that issue from one
  • It also creates far more pressure to achieve concrete and valuable results than any system of performance assessment – strictly limited aid volumes makes failure far more costly
  • Ownership would take a big boost. The biggest boon to local ownership of the aid agenda would be the ability and necessity to refuse aid packages that are not wanted. If bad aid or aid out of line with the local development vision replaces good aid, the incentive to exercise ownership of the agenda and refuse the aid is that much higher
  • A limited volume of aid would mean that the limited management capacity of local institutions is not stretched over hundreds of projects and programmes (as is the case now), but focused on a few tightly prioritized programmes, with consequent benefits for implementation.

Offsetting these benefits is the obvious drawback that much of what needs to be done could not be financed with a strict aid ceiling. The value of the thought experiment is not to assess whether we should cap aid, but to see what benefits it would bring so we can work out how to generate these benefits without limiting aid volumes. The hardest two to incentivize without some resource constraint are the improved ownership (which really requires that local institutions to refuse aid far more than they actually do) and the management effects. Suggestions welcome.

Staying in the region, Chris Blattman also linked to a paper suggesting that Somalia was better off without a state. While the writer of the original paper seems to be a libertarian (and somehow the headline ‘Libertarian Dislikes Government’ isn’t all that shocking), it does make an interesting basic point: the correct counterfactual when we consider poorly governed or chaotic states is not a ‘good’ Government, but a ‘different’ one. This is something to bear in mind when we consider countries with corrupt, dictatorial or ineffectual Governments. Very often it’s not the specific leadership or political party that is at the root of bad governance or decline, but the structural characteristics of their rule: their relationship with the opposition, their relationships with the public, their ability to govern effectively and the options open to them. In very different contexts I’ve argued this for Malawi and Zimbabwe. In Malawi, the same leadership was fine in one political context and dictatorial in another. In Zimbabwe, the temptation to lay all blame at the feet of a very convenient (and abhorrent) villain blinds many to the complex roots of his crimes and Zimbabwe’s decline. The saddest thing is that this point is often neglected even by immensely powerful decision makers – the invasions of Afghanistan and Iraq both show marks of this error.

Going back to the original article, is there a case to be made for statelessness when the Government is predatory? The argument is strongest in the static view, comparing a bad state and statelessness. The more dynamic the view taken, where we consider how tyrannical or corrupt regimes can become effective ones, the less this view makes sense, though it’s also true that plenty of tyrannical regimes didn’t evolve but were overthrown. Again, the thought experiment of whether some developing countries would be better off stateless is most valuable in focusing our attention on the long term needs and real dynamics of state building – not just in terms of building institutions but understanding how bad or predatory institutions can evolve into good ones. It’s happened in many countries, and links back to the questions of accountability, representation and taxation discussed above.

9 thoughts on “A Couple of Provocative Thoughts

  1. Sam Gardner

    September 16, 2011 at 6:17am

    Indeed a very provocative idea. It puts the need for internal structures for accountability squarely on the table.

    The ” do no harm” principle, such as developed in the Sphere standards concerning supporting local systems should be better investigated and, applied.

    However, I am doubtful whether for the unarmed poor roving bands of warlords are better than even a predatory state.

  2. Justin

    September 16, 2011 at 12:39pm

    Why do you dismiss the idea of actually capping aid? If we are confident that doing so would bring the benefits you outline why should we be looking for other ways to bring those benefits? Is it that you think aid-capping is simply politically impossible or do you really think that without aid developing countries have no means of financing “what needs to be done”? It would seem that if through aid-capping development projects were made more effective that the resulting stronger country economies through taxation could then start taking on the burden of paying for “what needs to be done?”
    Without meaning to cause any offense, aid professionals sometimes sound like overbearing mothers of adolescent children who can’t imagine that “their babies” (developing countries) could possibly make it on their own out there in the big scary world without the guiding hand and support of their mother.

  3. Ranil Dissanayake

    September 16, 2011 at 3:02pm

    Hi Justin – no offense taken at all – I think that’s a very valid criticism, one I try and stay aware of. That’s why it’s so import to subjugate our own agendas to those of the local leaderships. The circumspection about aid capping comes from the following: 1) Political difficulties, as you suggest; 2) the fact that it would be a second best solution to an un-capped aid portfolio which manages to achieve the results I mention; 3) The fact that very few developing countries (regardless of corruption levels) would welcome a cap on aid; 4) That it’s all speculation without clear evidence.

    If there’s reasonable evidence to suggest aid capping does bring these benefits in a way no other method can, and that it generates more economic and political development, I’d be a vociferous advocate. You need look only at the world cloud of tags on the side bar of our blog to recognise that both Matt and I lean towards aid cynicism in a lot of cases.

    Sam – when I was writing this blog ‘Do No Harm’ was actually the working title. I agree with what you say in your comment.

  4. scepsec

    September 16, 2011 at 3:25pm

    Hi Ranil
    I think two points are missing from your piece. Which I’ll give is an important debate though.

    1. the cost to the Somali people in the process leading to the current situation. Pulling the plug on a country might or might not produce a good end result. The question is how many is left to enjoy it.

    2. the Real political opportunity costs to donor- and other countries during that same process. (e.g. int. security and immigration)



  5. Ranil Dissanayake

    September 16, 2011 at 6:05pm

    Soren (sorry, no idea how to produce that o on my keyboard)

    These are good points, but I’m not seriously advocating statelessness. The more interesting questions that the paper can throw up are firstly, whether there are realistic state-led outcomes that are not likely to be massively predatory (and if not, how do we widen the range of possible outcomes), and secondly whether we can say anything about how the likely regimes can be expected or induced to change in positive ways over time.

  6. Paul Clist

    September 17, 2011 at 10:01pm

    Agreed – it was an interesting article, and some interesting reflections here too. I have one point of contention and one point of agreement for you…

    My contention – you seem to assume aid inflows undermine tax revenue collection. We showed in a recent paper (Clist & Morrissey, 2011. “Aid and tax revenue: Signs of a positive effect since the 1980s”, Journal of International Development) that both grants and loans have had a positive effect on tax revenue collection over the last 25 years. You also find a larger effect if you allow for aid to affect tax over a more reasonable period of time (e.g. not simultaneously as in some studies). Building a tax base is an expensive thing, and many governments want to do it anyway, they just have other priorities. In short, the idea that cutting aid would increase the incentives for governments to undertake investments in tax collection that will be unpopular and unlikely to yield benefits in the short term, is wide of the mark.

    As for the hollowing out of state institutions and capacity, it is very difficult to measure but unfortunately I agree. An optimist would counter that different types of aid impose different costs on the recipient, but the problem is real.

  7. Ranil Dissanayake

    September 17, 2011 at 10:18pm

    Paul, many thanks for the comment – I hadn’t come across your paper before, but unfortunately I can’t access it as it’s gated. Very interesting findings though – what sort of magnitudes are you talking about (assuming you did a regression or number of them, what are the values associated with the loan/grant variables?). Would be really interested to hear a bit more about this, and also about what the mechanism you postulate for this positive effect? Is it because aid is supporting the development of a tax regime and taxable population? or because aid is increasing the ‘formalisation’ of the economy? or simply increasing the size of the economy and as such absolute tax revenues.

    Thanks for bringing the paper to my attention.

  8. scepsec

    September 18, 2011 at 11:26am

    No worries about the letter ‘ø’..

    Yes, I did get that you’re not advocating laissez faire here and like Paul, I’d agree it’s a real problem. My point is just that debating the matter without context is close to absurd.

    I think that if you’d want to isolate the effect of aid on state institutions, you’re bound to isolate it to the extend, where it is removed from your problems in the first place … creating a worthless solution.


  9. Paul Clist

    September 19, 2011 at 11:42am


    The magnitudes were always fairly small (but see an early ungated version here
    http://www.econdse.org/seminar/seminar26.pdf ). But I think the important point is that we don’t find evidence to support the popular assertion that ‘aid leads to lower tax revenue’ in recent times – we find positive effects!

    There are plenty of possible mechanisms, and its difficult to conclusively come down on one side or another, so I’ll speak just for myself. I prefer 3 explanations, which will apply in slightly different contexts. One, recipient countries want a larger tax base anyway, and loosening their budget constraint will mean they will invest more in tax revenue collection. Two, donor pressure does work – it provides small monetary incentives and, more importantly, know-how. Three, simply by talking a lot about tax revenue, donors create the idea that this is a ‘norm’, and over time recipients adopt this aim for themselves. During the period of structural adjustment, explanation 2 seems apt.

    The effect of aid on state capacity is a little more open for debate, unfortunately!

Comments are closed.