Policymakers are increasingly calling for the creation of an early warning system to prevent future breakdowns of the global economy. But so far no one has answered the key questions of who would operate such a system and how it could work. If crises are to be detected and dealt with promptly, those charged with the task must be able to speak clearly and with authority.

Any forthright, disinterested assessment of the global economic system's stability requires two sorts of independence. First, the institution making the analysis and judgments must not have anything other than its own reputation riding on its assessment; in particular, its own policies or lending should not be shaped in any way by its judgment. That means it should not have any policy or lending facilities. Thus it cannot be part of the existing international financial institutions, all of which are policymaking, governmental or lending institutions.

Second, the institution must be independent of the big countries or parts of the global economic system that might contribute to future instability. Therefore it cannot be subject to interference by the board of the institution. That means that its assessments cannot be part of the IFIs in their current form, or indeed any form that may emerge from reform proposals. The fact is, main shareholders, through their board membership, always interfere in any statement that they think might be interpreted as critical of their country.

Those who have served in the IFIs at a senior level (in my case for a decade in the European Bank for Reconstruction and Development and the World Bank) will know how hard it is to make judgments about important economies without the direct and indirect intervention of board members or their staff, often on explicit instructions from their home capitals. It would be impossible to have frank and clear assessments of economic risks if such politically motivated board interventions or editing were possible.

We must not delude ourselves that this task of fostering stability could be performed by the International Monetary Fund, the World Bank, the Financial Stability Forum, the Bank for International Settlements or any future form of such institutions. John Maynard Keynes foresaw the inevitability of such interference from the board clearly as he shaped those institutions at Bretton Woods in 1944. Experience has shown the wisdom of his prescience. Separation of the assessment from board interference is crucial.

What would be required? The structure of any new institution follows directly from the need for the two types of independence. Further, the impact of its assessment of sources of instability will depend on the quality of its analysis, so it must be staffed by first-rate officials. Its task would be to warn explicitly of the key sources of future instability and, in doing so, put pressure on decision-makers in government or key private institutions.

With 100 high-quality staff and outstanding leadership, such an institution could be very effective. A budget of $20m (euro 15m, pound 14m) per annum would be sufficient. An endowment of $500m would give it the independence it needs for 30 years or more. Its board would be advisory, non-resident and meet not more than twice a year. The board would have the power to appoint the head (for, say, a seven-year term) and ensure its finances are well managed. It would have no power to interfere with, or comment on, its assessments.

It should be led by someone with outstanding economic credentials, who has strong policy experience at a senior level and who will be respected for taking a position independent of his or her country of origin. Examples of possible leadership, focusing on countries that are not part of the Group of Eight industrialised nations (this need not be a requirement, provided the individual has a strong international perspective), include: Montek Singh Ahluwalia (India), Leszek Balcerowicz (Poland), Kemal Dervis (Turkey), Francisco Gil D�az (Mexico), Stanley Fischer (Israel) and Trevor Manuel (South Africa).

Tinkering with existing institutions cannot provide the independence we need. The politically sensitive task of warning about growing systemic risks can be delivered only by a new institution. We can create one at a cost that is very modest in relation to the dangers we face. As we witness the extreme consequences of getting it wrong, now is the time for action.

The writer is IG Patel Professor of Economics & Government, LSE, and is formerly chief economist of the EBRD (1994-99) and World Bank (2000-03). His book A Blueprint for a Safer Planet will be published on April 2

Copyright The Financial Times Limited 2009