Gerald F. Hyman: Focus on the particular: yes and no

Development Policy11 Mar 2010Gerald Hyman

Two opposite spectres hang over the article Aid for development can be better by Peter van Lieshout, Monique Kremer and Robert Went (hereafter ‘the article’) and, to a lesser extent, over the WRR Report 84 Conclusion (‘the conclusion’). These explore the interaction of what the article calls ‘three concepts of development, a country-specific approach, and a broader perspective’. The first spectre: the success of East Asia in general, especially the relatively recent spectacular growth of China; and second, the relative failure of sub-Saharan Africa but also other regions, especially parts of South Asia and (but for the huge oil incomes in the Gulf) the Middle East.

‘Why does Africa not follow the path taken by Asia?’ ask the authors of the article. Why indeed? Their answers are not persuasive, in my view. They centre on what the authors see as a failure of donors to direct their assistance properly, especially because of the authors’ unease with ‘sweeping statements’ and ‘ready-to-go recipes or big answers’, rather than specific country contexts.

Without at all denying the importance of local context, there are some simpler, more obvious reasons for the discrepancies in performance. Unfortunately, they are not politically correct reasons. They go to endogenous rather than exogenous factors: the history, culture, psychology and, most importantly, the political economy of the poor countries, rather than the policies of rich, primarily Western donors. In many cases, that local, endogenous political economy consists of a small, ruling, too-often predatory elite and its set of patronage networks and personal greed.

Unfortunately, most African countries fit that description, but Africa is hardly alone. And under those conditions, the policies that have regularly achieved success – what the authors dismiss as ‘replicable examples’ – do not get implemented. These include (but are not limited to) relatively free, competitive markets with open competition and low costs of entry; low levels of corruption; the rule of law; good governance; free flows of information, and; a competitive political system.

Why? Primarily for political economy reasons: because those well-worn, general policies would disadvantage the kleptocratic elite and their clients. Not because the policies are really in doubt. Of course, there are other reasons as well. Paul Collier has enumerated some in his Bottom Billion book. But, as I read the evidence, those other answers do not account simply or mostly for the variation in performance.

Take China. To oversimplify, after re-securing power in 1980, Deng Xiao Ping inaugurated huge economic growth by changing China’s policies towards the market. In doing so, he created new incentives. China’s enormous underlying potential – including a willingness to work hard, delaying gratification, disproportionately high savings rates, investments in education, and so forth – kicked in. Of course, there were other factors as well, including local context. But those other factors did not change much from pre-Deng to post-Deng, and certainly not enough to account for China’s subsequent growth. One factor definitely cannot explain China’s performance: the role of the donors and their assistance programmes.

In Africa, on the other hand, donors have ploughed in billions of dollars in assistance without much to show, yet regularly agonize about their own performance and policies. They are encouraged, even driven in good measure, to do so by the self-justifying rhetoric of precisely the kleptocratic elites, whose self-interests are at the root of the poor policies and performance that cause the problems in the first place.

So the article and the conclusion pose a number of suggestions, not for Africans and their governments, but for the Dutch. What’s wrong with the Dutch, they ask? How can the Dutch improve their assistance?

Part of the answer, they say, is that Dutch assistance has been has been applying general theories of development and consequent programmes, when the answer lies in the specific contexts of particular countries. Generalized solutions and policies are misguided, these two pieces argue. It is precisely like a physician, they posit. Each patient is different and needs not a general treatment or prescription, but a set of idiosyncratic medications peculiar to his or her personal maladies.

Yes and no, in my view. Medicine proceeds precisely under the assumption of commonalities mixed with peculiarities. Doctors do not begin anew with each patient, as if that series of symptoms had never been encountered before. Instead, physicians begin with, for example, generally applicable patterns of physiology and the germ/virus theory of pathology. This begins under the assumption that the causes of illness are common and can yield common prescriptions. With those and other tools, physicians diagnose (the article’s appropriate term) the illness(es), sorting them into sets that have been seen hundreds of times before and for which medical research has (hopefully) come to some ‘generally applicable recipes’ and remedies.

Of course, the particular complex of maladies, especially among the elderly, may well call for an idiosyncratic mixture of remedies because of their interactions. Some illnesses have no such cure. A few have not yet been encountered. Indeed, the article and the conclusion reinforce that same approach, if only inadvertently. If some level of generality were beyond reach, what would be the purpose for establishing the ‘knowledge infrastructure’ in Africa and Asia that they both call for? Or for ‘learning from mistakes’, or putting ‘learning first’? Or for ‘right order’? Such ‘information banks’ might have separate, isolated units for each country and sub-country, but that does not sound like what either document envisions, nor would such a limited set of institutional approaches be very interesting or productive.

So, the spectacular growth in China exemplifies some general principles that have also been illustrated in Taiwan, South Korea, Chile, Ghana, and so forth. Many issues are, of course, highly idiosyncratic: how China put them together; how it overcame the policies and institutions that inhibited growth under Mao; how deeply entrenched party interests, procedures and prerogatives were overcome; how its historical and social structural impediments were surmounted, among other issues. But there are generalities about what needs to be done, even if the path to doing so is particular. In some ways, it would be absurd to suggest that Kenya follow the Chinese path because the political and social contexts are so different. In another sense, that is exactly what Kenya needs to do: follow the general policies and prescriptions of China, South Korea, Chile and Ghana.

In my view, the commonalities are critical, but if assistance is to be at all useful it will need, as the authors say, to be grounded in the particular ways that prevent some of the time-tested remedies from being put into place – and these are deeply rooted in the specifics of local society, culture, history, politics, economics and psychology. The key, then, is precisely to look for generalities while respecting the peculiar expression of those generalities in the particular country context.

None of this gainsays that donors should regularly examine their own policies and improve their performance. How can donor performance and assistance policies be improved? These two documents raise many interesting issues and reach several other interesting conclusions about Dutch assistance – the role for the Netherlands in general, whether its geographic and topical commitments have been too dispersed, whether it needs a better, ‘non-confetti strategy’, the relative efficacy of investments in the social sector rather than some others (particularly productive sectors), and so forth – but these can wait for some later comment. For the moment, it would be better not to abandon the search for some commonalities among the welter of specifics: generalizations like paying more attention to the middle class or getting the ‘right order’, as the authors of the article suggest. Nor, however, would it be wise to apply blindly some general prescriptions without an in-depth understanding of the particular contexts.

WRR report: conclusions

  • Strong governments in Asia but not in Africa (1)
  • Impossible to say in general about what works best and why (2)
  • Prescriptions have been too general (2); there are path dependencies (3) but a ‘right development process’ (3)
  • General answers but always exceptions (3) [Of course, but they are exceptions to the general]
  • Substantial professionalization (6)
  • Extensive local knowledge and contacts (6)