Global problems are too complex for simple solutions


15 May 2004
Financial Times

(c) 2004 The Financial Times Limited. All rights reserved

Next week Copenhagen will host an intriguing intellectual exercise: a debate between competing saviours of the world.

The Copenhagen Consensus is a project of Bjorn Lomborg, the "sceptical environmentalist" who used cost-benefit analysis to undermine the rationale for the Kyoto carbon emissions treaty. The Copenhagen exercise extends that approach to a range of global problems. Given the limited resources available - US$ 50bn or so a year of development aid - it asks a panel of economists to choose between 10 bids from experts, each quantifying the return from spending more money in their field. Is hunger and malnutrition the most deserving cause? Infectious diseases, anyone? Or perhaps clean water?

The project bills itself as a clear-eyed, scientific look at an area dominated by arbitrary choices and influenced by woolly emotion. But while it is certainly thought-provoking, comparing costs and benefits sector by sector obscures the fact that such problems are not only highly uncertain but also heavily interrelated.

The Consensus claims a new approach. But international development has always attracted what might be termed "CEO solutions": selective, top-down, data-driven command-and-control. It recalls Robert McNamara, who brought the strategic planning management techniques of Ford (and the Vietnam war) to the World Bank, without great success; and Paul O'Neill, the former aluminium executive who, as US treasury secretary, spent his tour of Africa in 2002 working out a back-of-the-envelope scheme to supply clean water to all Ugandans and wondering why everyone else could not see that it was as simple as he thought.

Sadly, few great advances in development owe much to such one-dimensional solutions. Some do: the postwar "green revolution" radically increased crop yields and reduced hunger, to the credit of the Ford and Rockefeller foundations, which bred new crop strains, and the developing countries that adopted them. But few of today's problems respond to isolated and concentrated technocratic effort. An AIDS vaccine would be one, but that does not make it on to the Copenhagen list.

As one Save the Children aid worker bravely told Mr O'Neill in Uganda, villages all over Africa have useless wells. Donors who decided that clean water was the priority would build a well without ensuring spare parts to repair it, skills to maintain it and local support to keep it going. The obsession with one dimension obscured the importance of others.

The history of aid and health is another example of the limitations of the Copenhagen approach. It biases the outcomes towards easily identifiable projects such as treating a single disease whose costs and benefits appear spuriously precise. The history of such interventions in poor countries, very popular decades ago, is not encouraging. With some obvious exceptions, such as smallpox eradication, such programmes acquired a bad reputation for pulling key staff out of other parts of the health system, creating serious - but probably unquantifiable - costs elsewhere.

The pitch to Copenhagen to spend money on combating communicable disease, by two academics at the London School of Hygiene & Tropical Medicine, is interesting. They point out that there is considerable uncertainty about both the costs and the benefits of such an initiative. Whether the money is spent effectively depends heavily on the public health systems that will carry out the treatment. Their costs are hard to quantify and probably understated in the bid. Donors to HIV-AIDS programmes in Africa have discovered that they can buy as many anti-retroviral (ARV) drugs as they like but a treatment network is vital. And multiplying the price of ARVs by the number of AIDS sufferers is far easier than working out how to rebuild a broken health system or the cost of doing so.

These days, the more progressive aid donors are edging towards a different approach. They decide which countries are likely to use the money well, let them fix priorities and then either fund their budgets or provide technical assistance. States, almost by definition, have many dimensions. Honest states that function well, govern under the rule of law, refrain from fighting their neighbours, provide or encourage health and education services and integrate their country into the global economy will tend to be rewarded with growth that reduces poverty, pays for clean water and prevents mass emigration. Such an approach can address all the Copenhagen aims - except, perhaps, climate change, which, as a global public good, is rather different from the rest.

The progressive aid approach has its own drawbacks. It looks uncomfortably like picking winners among governments and betting on them, which could easily go wrong. But it is predicated on a more helpful view that development goals complement rather than compete with each other. For instance, Susan Rose-Ackerman of Yale, who wrote the Copenhagen submission on governance, notes that the continued existence of government corruption will undermine attempts to solve any other problem by making it difficult to implement.

Ensuring that aid money is spent well is an admirable goal. Even if the project does no more than stimulate research into what actually works, and the wider costs and benefits of different solutions, it could be worthwhile. But problems as complex and multi-faceted as those the world faces are not amenable to such a simple choice.

The writer is a member of the FT's editorial comment team