The 2007 annual meetings of the World Bank and the IMF in Washington will be the first occasion that the new heads of the IMF and the World Bank have met collectively with their shareholders. Both organisations are facing a crisis of bankruptcy and irrelevance. The crisis reflects a widespread feeling among developing country members that the organisations continue to operate as extensions of western foreign policy; and that they continue to promote a core set of microeconomic policy prescriptions that have not been shown to work in practice.

As is well known, a consensus emerged over the 1980s in much of Latin America and in the multilateral development organisations headquartered in Washington. Dubbed the "Washington Consensus", it contained two central prescriptions: one, maintain a secure and stable macroeconomic regime by keeping the exchange rate competitive, the budget deficit low, the tax base broad; and two, improve markets by quickly liberalising foreign trade, abolishing restrictions on inward foreign direct investment, eliminating preferential interest rates, privatising public enterprises, and protecting private property.

Right up to the present, these prescriptions have constituted "global policy" on development, on the promise that market liberalisation would be rewarded by higher rates of growth and lower poverty. The alternative to market liberalisation is posited as state "intervention" in markets, which leads only to "rent-seeking" and stagnation. Hence the word "reform" is used only to denote change in a market liberalising direction.

There are three main problems with the Washington Consensus. First, the prescriptions emerged not out of the record of what demonstrably has worked, but out of what policy professionals in and around Washington would like to see, derived from simple economic theory. To this day, there is remarkably little good evidence that countries which adopted the Washington Consensus more have enjoyed better growth and poverty-reduction performance than those which adopted it less, at least if one excludes states which are barely able to do anything. Haiti over the past decade scores well by Washington Consensus criteria, Vietnam scores badly; but Vietnam has by far the better performance.

Second, by focusing on an entity called "the market", the prescriptions ignore the task of building the capacity of industries and firms, including the capacity to compete in international trade. They also ignore the macro task of changing the economy's production structure both to intensify input-output links and to diversify production towards sectors with faster growing demand and higher rates of profit. They assume, wrongly, that there are no serious "market failures" in these processes in developing countries, for which state action potentially can be a corrective.

Third, the prescriptions are couched as valid for countries at all stages of development. They ignore the "late development effect", the idea that countries which begin to industrialise when other countries are already highly developed have to use different policies and institutional arrangements to those used by earlier developers, in order to compensate for the disadvantages and capture the advantages of coming late - advantages such as the potential to use more advanced technologies already used elsewhere.

The most successful region of the developing world since the Second World War is East Asia. The development policy precepts followed in Japan, South Korea, Taiwan and China have a lot in common, and are substantially different in emphasis to the microeconomic core of the Washington Consensus. These precepts have the advantage of having worked on the ground. We call these precepts the Beijing-Seoul-Tokyo Consensus for development, or the BeST Consensus for short.

The BeST Consensus starts with the proposition that palliative policies, focused on poverty reduction (as in the Millennium Development Goals), must not be confused with development policies. Development policies take as their touchstone building capacities of (local or joint venture) firms, especially technological capacities; and strengthening the links from profits to investment and investment to profits.

In tackling the task of building the capacities of firms, public agencies can help compensate for deficiencies in the existing structure of markets - agencies such as export-import banks, export processing zone administrations, development banks, technology institutes, and high-level state coordinating agencies (such as MITI and Ministry of Finance in Japan, the Industrial Development Bureau and the Industrial Technology Research Institute in Taiwan, and the National Development Reform Commission and the Export-Import Bank in China). Staffing up such agencies with technically competent people is a high priority in countries following the BeST Consensus.

To encourage local or JV firms to grow and reinvest, states following the BeST Consensus assure them of preferential access to certain sectors and learning opportunities. The choice of which sectors to target is made easier in late-developers as compared to already developed countries, because they are much further back from the world frontier. Broadly speaking, the targeted products and technologies are among the ones the latecomer economy is already importing or buying at monopoly prices; or among the ones where private firms - at home or in neighbouring countries -- are already showing signs of success, perhaps in export markets. As more domestic firms become able to produce the newly substituted products or master the newly imported technologies, the assistance has to be retargeted at the new domestic frontier. Upgrading and diversification is continuous.

But the BeST Consensus does not endorse a simple import-substituting industrialisation strategy. It enjoins a combination of (a) assistance to exporters (including duty-free access to inputs and public provision of information about export markets), and also (b) temporary assistance to selected import-substituting production, in conditions of enough openness to ensure that most domestic firms are subject to close to world market prices (adjusted for transport costs) in most of their operations. Externally set prices put pressure on firms and the political leadership to be economically efficient. When state intervention weakens market incentives the state has to supplement market discipline with alternative disciplining mechanisms, such as sunset clauses on industry assistance. The East Asian economies provide abundant evidence on how to do what mainstream economists say is impossible - promote exports and import-substitution at the same time.

In the BeST Consensus the state builds markets not only by reducing uncertainties and by strengthening the profits-investment link, but also by disseminating information about opportunities at home and abroad and generating an expansionist psychology in the business community. The financial system complements this by combining competition between commercial banks with arrangements for concessional finance for targeted industries, and cautious external financial liberalisation.

The BeST Consensus gives heavy emphasis to tertiary education, especially science, engineering and maths, a priority which the Washington Consensus hardly mentions.

These policy precepts have to be adjusted in the scale and scope of application in line with the capacities of the state. It is not a matter of constructing a whole "developmental state" or else going for free markets and a minimal state. The important point is that the national (in some circumstances, regional) leadership embrace the norms behind the BeST Consensus, and then proceed to apply the precepts in line with the capacity available. The leadership can act in the knowledge that the BeST Consensus does have a proven track record, including in China today. Developing country governments and multilateral development organisations should pay it careful attention, and use the occasion of the annual meetings to stimulate a debate.

Kuen Lee is professor of economics, Seoul National University; John Mathews is professor of strategic management, Macquarie University; Robert Wade is professor of political economy, London School of Economics.