Two Very Modest Proposals for Making Technical Assistance More Effective!

Posted by Richard Allen1

Good Government Means Different Things to Different Countries” is the title of a recent article by Matt Andrews, Associate Professor of Public Policy at the Harvard Kennedy School of Government. It is an important paper, for its insightful analysis, but especially because it challenges the conventional wisdom of PFM experts in both the World Bank and the IMF. The central message is that there exists no one-size-fits-all approach to good governance. Effective government means very different things in different countries. Local conditions and institutions are hugely important. Matt tests this proposition through a study of selected PFM practices in a set of OECD and non-OECD countries, such as the development of fiscal rules and internal audit.

The challenge for the Bank and the Fund is that so much of their technical assistance and advisory work on PFM is based, explicitly or implicitly, on a model (implicit or explicit) of best practice. As Dani Rodrick has explained, this reflects a more general bias towards best practice models in the economic work of the Bank and Fund, which allows for cross-national comparisons and benchmarking institutional performance, but takes scant account of local constraints and opportunities. Matt rightly quotes the example of the PEFA framework, the implicit benchmarks of which (though they are not called benchmarks) represent good practice in advanced countries, especially countries with an Anglo-Saxon tradition of public administration. The Bank’s CPIA rating system is based on a similar premise. The implicit assumption is that less developed countries should move towards the advanced country model in a series of stages. The end-point is clear; all that is open to debate is the sequence of steps that moves the country to this desired state. However, as Matt demonstrates, this approach is seriously flawed.

Countless technical assistance reports tell a similar story. The actual practices in the low-income country being reviewed are compared with the advanced country benchmarks, and a set of recommendations are made which, if adopted, would rapidly move the country to this advanced state. Of course in practice, such a transformation rarely if ever happens. A subsequent mission, five years later reveals that in practice, little has changed in the country concerned, and a similar set of recommendations is made which in turn are demonstrated some years later to have had little or no impact. Such a story is all too familiar to those who have worked as advisors in the PFM field.

What can be done to resolve this dilemma? I would suggest two modifications of current IMF/Bank practice which should help a lot, but neither is straightforward. The first is for technical PFM missions to become less purely technical. A natural tendency for technical experts is to recommend technical solutions to problems that are not fundamentally technical, or at least have a critically important non-technical component. Take, for example, a recommendation that I found in a recent report on a middle-income country with a distinctively Southern European approach to public administration. This was to establish a top-down budget preparation process, in which the cabinet of ministers would play a key role in debating fiscal policy options, and setting multi-annual ceilings on budgetary expenditure. A related recommendation was to establish a system of periodic expenditure review, as used quite successfully in countries such as the Netherlands and the U.K.

Of course, in a perfect world, such recommendations might make sense. Unfortunately, in relation to the country concerned they make very little sense because (i) the cabinet system is weak and would require a complete overhaul to make it work on the lines suggested, an outcome which seems politically implausible; (ii) the government is a weak coalition, unlike in many Anglo-Saxon countries which have one-party governments, and has no proven capability to make hard choices about public expenditure; (iii) the ministry of finance is politically fragmented, weakly led and has limited capacity to undertake the technical analysis which would be necessary to improve the efficiency of the expenditure prioritization process.

What is clearly needed in such circumstances is for the authors of the technical report to have a much greater understanding of the political drivers of decision-making in the country concerned, and the powerful institutional constraints on implementing change. If this work had been done – and it requires skills in political economy analysis that are not usually found among the members of a technical PFM mission – the suggestions for reform might have been much closer to the mark and easier to implement. My suggestion would be that technical PFM missions mounted by the Fund or Bank should include a political economy expert whose job should be to carry out the necessary institutional analysis and subject the recommendations of the PFM technicians to a rigorous reality check before they are put onto paper. Greater use could also be made of local consultants with an in-depth knowledge of both public finance and the institutional environment.

My second suggestion is that technical assistance reports should never be the end-point of a process of intervention by the Bank or Fund. To prevent such reports being quietly put on a shelf in the minister’s back office, and left to gather dust, they should always be the starting point of a discussion and dialogue with the finance ministry on the analysis undertaken by the mission, and its findings and recommendations, Other stakeholders (for example, the president’s and/or prime minister’s office, the central bank, the ministry of planning, the line ministries, the national audit office, the parliament, and civil society groups) should ideally be included in such a dialogue. Instead of reports containing 50 or more recommendations compressed into a complex multi-annual action plan, the outcome of such a dialogue should be, for the average developing country, a much smaller subset – say three or four, at most -- of carefully selected actions that could realistically be undertaken during a 4-5 year period, and which stand a reasonable chance of being implemented.

China has adopted this approach for many years, with some success. Rather than asking the Bank or Fund to mount a full-blown technical assistance mission, the authorities propose a seminar on accounting or budget classification or budget preparation to which they invite specialists from a range of countries. The emphasis is on learning, engaging in a dialogue with experienced counterparts, and translating the experience of others into ideas which might (or might not) be relevant to the Chinese experience, their institutions and decision-making processes.

1 Richard Allen is currently working on a project with the World Bank that is analyzing the role and responsibilities of central finance agencies (CFAs) in low-income countries. A database covering 70-80 countries is being prepared, along with case studies of selected countries. The project includes a strong emphasis on the analysis of political economy and institutional characteristics of CFAs.