15 December, 2007Issue 7.1EconomicsPolitics & Society

Email This Article Print This Article

The Economist’s Burden

Seán Mfundza Muller

Paul Collier
The Bottom Billion: Why the Poorest Countries
Are Failing and What Can Be Done About It

Oxford University Press, 2007
224 pages
ISBN 978-0195311457

Imagine the journey that human beings embarked on with the development of the first Stone Age tools. It led us toward a lush-looking mountain which, with progressive social and technological evolution, we then began to climb. Since that beginning, the peak—if there is one—has remained permanently shrouded in cloud, but at progressively higher levels we find that our lives are richer than before. Certainly there are very few at any level who would prefer to return to a lower one. This is the Mountain of Development, and the populations of all the countries in the world are at different positions in the climb. That is, the human race has become segregated in its ascent. Some countries are mapping previously uncharted territory of prosperity and well-being on the higher reaches of the mountain. Others are still working their way around the lower contours. And even within countries there are differences; some have their citizens clustered close together, others are strung-out over great distances or even split into two or three groups that are barely linked at all.

Paul Collier argues that as a result, one billion of us (a sixth of the world’s population) have reached an altitude where levels of prosperity are beginning to approximate some kind of ideal. Another four billion are some distance behind, but will reach these higher slopes in the near future due to their growing strength and experience. But the final billion are struggling. In fact, they are trapped on the lowest slopes, and in some cases sliding further and further down. Most of these countries—approximately 70%, excluding the likes of Laos, Cambodia, North Korea, Haiti, and others—are African. Based on a concern for our fellow human beings and an interest in our societal future, Collier asserts that the primary development objective of the wealthy OECD (Organisation for Economic Co-operation and Development) countries and international development agencies should be to help this ‘bottom billion’ in their ascent.

Collier, head of the Centre for the Study of African Economies (CSAE) at Oxford, is one of the most important and influential authors in development economics. In addition to his work with co-authors on aid, civil war, development, poverty, capital flight and the interactions between these, The Bottom Billion builds on two papers he co-authored in 1999 with Jan Gunning that are now primary references on sub-Saharan Africa’s poor post-independence economic performance. The book is also the continuation of a more than decade-long academic debate about why some countries have experienced little or no economic growth. The two other main positions are represented by The End of Poverty written by Jeffrey Sachs—perhaps best known to most as the Special Advisor to the current and previous Secretaries General of the United Nations—and William Easterly’s The White Man’s Burden. Towards the end of The Bottom Billion Collier explicitly situates his analysis between these two extremes. One the one hand, he agrees with Easterly that aid is not the primary solution, and that therefore the excessive focus of Sach’s proposals on external financial support is misplaced. However, unlike Easterly he does believe, like Sachs, that there is still much which can be done by outsiders at a fairly broad level. In that sense his is a message of optimism.

Collier’s own approach in The Bottom Billion, captured by our mountain analogy, begs a number of questions which are worth noting before we proceed. First, the idea that all countries are aiming at the same notion of a developed society (i.e. climbing the same mountain) might seem a little more plausible following the end of the Cold War, but there are many cogent voices which suggest the traditional Western notion of development is flawed. Second, the premise that it is possible, or desirable, to determine what is best for other nations is one that has been historically controversial. Third, while Collier acknowledges that the nature of growth matters (rather than simply its extent), it is an issue that he otherwise ignores. Finally, and perhaps most problematically, Collier assumes that the middle four billion’s ascent toward prosperity is assured.

The book begins promisingly by arguing that a one-size-fits-all approach to development is inappropriate. But although one is deemed insufficient, apparently four sizes are enough, because in the second part Collier argues that every country in the bottom billion can be characterised by at least one of four traps: the conflict trap, the natural resources trap, the ‘landlocked with bad neighbours’ trap, and the ‘bad governance in a small country’ trap. These categories seem somewhat arbitrary and akin to methodological gerrymandering. Why not the ‘surrounded by one country, mountainous and AIDS-afflicted’ trap, which would crudely characterise the Southern African country of Lesotho? Though whether Lesotho would even fall into Collier’s bottom billion is unclear, because he chooses not to publish the list of fifty-eight countries on the basis that this might ‘stigmatise’ them. (An odd decision, since he appears happy to give individual examples.)

The basis for the creation of the categories as the four main predictors of poverty was ostensibly the econometric analysis (the use of statistical methods in economics) conducted by Collier and his co-authors over the last decade. On one issue at least, the consideration of natural resources, this type of analysis appears to yield some important insights. A simplistic assessment of the matter might conclude that countries do better when they receive resource windfalls, but in fact this seems to be more the exception than the rule. Collier does a good job discussing the reasons for this: the resource boom displaces other economic activities, reduces the competitiveness of manufactured exports, and creates dangerous political incentives. The discussion regarding the impact of being land-locked, on the other hand, is vaguely informative but yields nothing new. In terms of the other two traps, it is hardly surprising to find a relation at the national level between poverty, conflict, and poor governance. But do we need econometric analysis to tell us this? More specifically, does an econometric analysis reach conclusions that we could not have reached otherwise, or succeed in giving these conclusions greater validity?

This is one of the main problems of The Bottom Billion: Collier uses econometric analyses to give his arguments weight that they wouldn’t otherwise have. He regularly makes the point that his claims are based on ‘a mass of technical papers published in professional journals and subjected to blind refereeing’. Two issues arise.

First, as any professional economist should know, publication of papers based on empirical analysis doesn’t imply their conclusions are true. It just means the analytical process by which they were reached is not obviously flawed. That both Sachs and (particularly) Easterly are widely-published should make that clear. For instance, consider the question of whether aid makes a difference to economic growth. Some published papers say it increases growth, some say it doesn’t and more recent ones say ‘it depends’ what other factors you control for. Second, it is all too easy to weave a plausible story around an econometric result. This occurs in a number of places in the book. One example is the somewhat counterintuitive result that the propensity for oil-related violence in Niger Delta regions decreases in association with higher numbers of oil wells. Another is the finding that aid appears to reduce the chance of wars, but does not have the same effect on coups. That selectivity in explanation carries over to inferring causality from correlations. In one case, for instance, Collier makes the causal statement that a resource-rich country shifting ‘from autocracy to intense electoral competition would lower [its] growth rate by nearly 3%’, but in other places he explicitly refrains from this type of assertion.

Such definitive statements are typically beyond the bounds of what can be determined at the country level when employing the statistical tools of economics in a responsible way. In part this is due to the actual economic theory of growth being unsatisfactory, particularly with respect to developing countries. To return to our earlier analogy: there isn’t even agreement on whether poorer countries not stuck in the quagmires of the bottom billion should be able to climb the mountain faster because the lower slopes are easier to ascend, or whether by virtue of having developed superior techniques and tools the rich countries are destined to stretch their lead inexorably into the future. (This is referred to as the ‘convergence’ issue in growth theory.)

Given the credibility derived from using an econometric approach, is it true that Collier’s diagnosis and recommended cures are fundamentally a function of econometric evidence? The short answer is that they are not. Many of the problems could be diagnosed in a way that combined qualitative insights with much more simple analysis of the statistics. In addition, the policies proposed as solutions rarely make recourse to statistical analysis of the kind that was apparently so important for diagnosing the problems. As a result Collier appears guilty, in both a methodological and practical sense, of what the African economist Thandika Mkandawire once called ‘the pessimism of the diagnosis and the optimism of the prescription’.

Neither can Collier’s arguments be rescued by appeal to solid qualitative evidence. The first half of the book contains many examples of poor attempts to illustrate the ‘statistical evidence’. For instance, is De Beers an example of corporate social responsibility because they have worked to reduce the sale of blood diamonds? Or did they realise that they could strengthen their diamond cartel by implementing a system which excluded the entry of non-licensed stones? Is the Mozambican rebel group Renamo the best example of a movement started to fight for social justice that became increasingly dominated by opportunists, when it was actually started in part by the apartheid government in South Africa to destabilise the post-independence government of Mozambique? Is it not a little dishonest to cite Botswana as a shining example of economic development when, despite sustained GDP growth on the back of diamond production, most citizens remain unemployed, or in subsistence agriculture, and HIV prevalence is among the highest in the world? (Neither of which, one must add, is due primarily to bad governance, but simply the limited success of attempts to diversify the economy and combat HIV infection.) Doesn’t recent research suggest that civil wars in Africa may not increase the prevalence of HIV in affected countries, contrary to what Collier claims and also contrary to what had been previously thought?

In the same vein, the book suffers from significant historical omissions and revisions. Most notably, pre-independence colonial history is completely ignored. This omission is particularly egregious in the African context, where most of the countries in question were once colonies, delineated and structured around natural resource extraction with minimal local development. One could argue that the current economic problems facing the bottom billion are separable from anything other than their recent pasts, but it is a doubtful claim and Collier doesn’t even attempt to make it. At one point he does say, ‘In retrospect, it was perhaps a mistake for the international system to permit economically unviable areas to become independent countries…’. Would that be the ‘international system’ under which Africa was carved-up among the colonial powers for the exploitation of slave labour and natural resources?

Were such revisionism simply tailoring the argument to an audience uninterested in historical baggage one might overlook it. However, it also results in flawed analysis. Can one discuss ethnic strife in countries like Rwanda without considering the colonial influence? Is it coherent to decry the role of aid in financing military expenditures, but ignore the hypocritical role of many OECD countries in securing arms deals with dubious states? Is Tanzania really an example of Uganda’s ‘bad neighbours’ because it invaded the latter, when in fact this was only because the brutal dictator Idi Amin first sent his troops to invade northern Tanzania? Relatedly, though he steers clear of ideologically-based free-market economics, Collier feels the need to defend the structural adjustment policies (SAPs) ‘negotiated’ with developing countries by the IMF and World Bank in the 1980s. This may not be a coincidence, given that Collier has worked for the World Bank, but it is unfortunate. SAPs have been used as excuses for economic failure by some governments, but there are documented examples—such as the Mozambican cashew nut industry—where components of the policies destroyed whole sections of various countries’ economies.

Many of these mistakes or selective inaccuracies do not radically harm Collier’s arguments, but they make one very sceptical of the soundness of the story being sold to relatively uninformed audiences.

Despite this, many aspects of the book are praiseworthy. Collier is frankly critical about the role of Western countries in resource-related corruption, and relative to many economists, he manages to consider the impact of globalisation on poor countries in a non-ideological manner. (Concluding, incidentally, that it has done them more harm than good.) In considering possible solutions to the identified problems, he does not exclude certain options merely for the sake of political correctness. Furthermore, unlike some of his predecessors—such as Joseph Stiglitz, the Nobel Prize winner who authored Globalisation and its Discontents—Collier does give some consideration to how the various changes in OECD policy might be enacted within the existing power structures.


For those already familiar with the issues raised, the book is worth reading because of the spectrum of policy proposals it makes and the fact that it will almost certainly be influential. Those unfamiliar with the facts and debates will no doubt find it interesting, but would do well to approach it with a sceptical eye.

Overall, where the The Bottom Billion succeeds is in characterising the failure to engage with the development challenges it outlines. On one hand it successfully critiques many past approaches to growth and development: ill-informed military interventions, excessive rigidity and myopia in development agencies, well-intentioned but misplaced actions by NGOs, and excessively ideological positions on globalisation and its effects on the poorest countries. In these respects it is a breath of fresh air. On the other hand, the book itself serves as an example of the very serious problem with the way in which economists inform policy.

Contrary to Collier’s assertion in his introduction, no intractable questions are cracked. Despite great and admirable efforts to the contrary, economics is simply not a science at such a broad level. Economic theories of growth remain contested and largely of abstract intellectual interest. The result, one of the lesser-known ironies of economics, is that those practitioners most eager to make sweeping, definitive statements based on econometric analysis are often those with either the least understanding of its theoretical underpinnings and practical constraints, or the most willing to overlook them. Thus the most useful policy recommendations and much of the preceding discussion in The Bottom Billion are, I would suggest, more a function of Collier’s broad qualitative experience and are only tenuously linked to the cited statistical analysis. This, then, is the economist’s burden: that the tools exclusive to the field are incapable of answering the social and economic questions that are most important. Hundreds of years on and with the benefit of hindsight there is still debate about what precisely enabled the current crop of developed countries to scale the slopes of the development mountain at such speed; it would be foolhardy to think that people who cannot even answer this question can tell us how the rather different countries of the bottom billion can emulate them.

Seán Mfundza Muller is a South African student reading for an MPhil in Economics at Oxford.