Medium-Term Expenditure Framework Debate

The objectives of the debate among others were to:

• define the medium-term expenditure framework (MTEF) and clarify any misconceptions;
• ascertain the Fund’s perspective on the framework; and
• learn from Uganda and other countries successes and failures in implementing a MTEF.

Vinaya Swaroop opened the session by explaining that the MTEF is a tool for linking policy, planning and budgeting over a medium-term (i.e., 3 years); it consists of a top-down resource envelope and a bottom-up estimation of the current and medium-term costs of existing policies; and it involves rolling over this exercise each year to reflect shifts in policy. The framework matches these in the context of the annual budget process;

If successfully applied, Mr. Swaroop explained that the MTEF can (1) improve a country’s macroeconomic balance by developing a multi-year resource framework; (2) assist in improving resource allocation between and across sectors; and (3) improve predictability of funding for line ministries. Mr. Swaroop also clarified some common misconceptions: that the MTEF will facilitate sound budget planning only if given credible policy choices; the link between inputs and outputs may remain unclear; and the MTEF will not increase accountability (i.e., how resources are being spent). The audience was given a checklist of five points that should be considered when attempting a MTEF:

• Does the country have reasonable three-year projections of expenditures and revenues?
• Is the bottom-up exercise of three-year cost projections (capital and recurrent; program and subprogram costs) feasible?
• If capacity does not exist, is there a sequenced process of building it?
• What country preparations are needed before a MTEF can be successfully adopted?
• Will the MTEF solve all service delivery problems?

Mr. Swaroop closed with an example that highlights the importance of the checklist above. Ethiopia, the second largest country in Africa (after Nigeria), has limited technical capacity as is the case in most sub-Saharan African countries. A federal country with 11 provinces, it does not have a plentiful supply of planners, accountants, budgeting experts, especially at the provincial level. As a result, most provincial governments are unable to plan one year ahead much less implement a medium-term expenditure framework. Currently, there is no possible way to support the MTEF concepts in most of the provinces. At the federal level, some progress has been made in budgetary planning particularly in the area of the capital budget, but little work has been done to get a medium-term perspective for recurrent spending.

Barry Potter began by explaining that the Fund has done less work on the MTEF than the Bank. There is generally a positive view of the framework within the Fund so long as it is implemented in a stable macroeconomic setting. He suggested that it is probably not suitable for all 182 Fund members, but acknowledged that some proponents of the MTEF consider that governments must have an MTEF to have a decent annual budget.

Mr. Potter acknowledged that the MTEF is initially very difficult to implement and that one must be prepared for countries to make mistakes. In Ghana, the MTEF was striking because it existed as an external exercise grafted onto the system. When the country faced a large trade adjustment, the annual budget was quickly overtaken and hence the MTEF; this had a very negative impact on the credibility of the MTEF for line ministries. Therefore, one lesson might be that if a country cannot adhere to an annual budget, it should be wary of putting too much emphasis on developing a medium-term expenditure framework.

There are three reasons a country should plan for an MTEF: (1) it increases macroeconomic control; (2) it leads to a more logical allocation of resources than an annual budget; and (3) it offers all levels of government greater confidence to effectively plan services. The speaker proceeded to describe six features important to the implementation of an MTEF:

• It should not be attempted without a clear statement of fiscal policy;
• An objective macroeconomic framework is necessary;
• It requires spending estimates for years two and three (they are neither budgeted nor arbitrary numbers);
• Estimates should receive "formal" status;
• Out-years have to be limited by a top-down constraint; and
• Out-years for years two and three should ideally create a hard budget constraint.

Many countries incorrectly implement certain components of the MTEF, because of failure to set, or take account of, one or all of the following:

• An objective macroeconomic assessment;
• Cyclical factors;
• A sensible time horizon (three or four years, but anything beyond is not worthwhile);
• Inflation, particularly different relative price movements;
• Distinction between existing and new policies;
• Planning reserve.

Mr. Potter addressed the situation for heavily indebted poor countries (HIPCs), where the HIPC program will take a number of years to complete. Careful thought will have to be given to how the Bank and Fund measure the change in public spending. In the meantime, he urged that governments not make short-term fixes that will not positively impact longer-term plans.

Allister Moon presented a case study on Uganda. The Uganda MTEF began in 1992 and has undergone three major phases since then: The first, from 1992-1994, focused on the macroeconomic framework and selective treatment of medium-term allocation (e.g. wage bill, defense, roads, etc.); The second, from 1995-1997, examined the macroeconomic framework in addition to comprehensive sector allocations, which linked to sector policy objectives (e.g., UPE, PEAP); The current phase, which began in 1998, focuses on annual consultations with donors, legislatures, and civil society.

The speaker listed the preconditions for beginning a MTEF. They include: macroeconomic stability; a stable resource envelope; commitment from the cabinet early in the MTEF cycle; core capacity of the finance ministry and central agencies; and supportive donor behavior. Other conditions need to be in place to enable progress, such as: capacity to enforce a hard budget constraint at the sector/ministry level (systems capacity and political will); executive commitment to having a more transparent budget process; capacity in sector policy analysis and expenditure planning; and continued donor support. Mr. Moon noted that in Uganda, there was strong commitment from cabinet to put these conditions in place early in the MTEF cycle.

More information is available in the attached Powerpoint presentations.

• MTEF Objectives and Preconditions (40KB)
• MTEF: What is it? (80KB)

Questions

1. Is the predictability of resource flows impacted by the MTEF?
2. How sustainable is the MTEF in Uganda?
3. How does the Bank plan to address the issue of capacity building?
4. How can the Bank plan for poverty reduction if a country conducts planning on an annual basis?
5. How do you link inputs to outputs to outcomes?
6. To whom does the Bank sell the MTEF? What type of political commitment do you need?
7. Is the ‘project’ approach compatible with the MTEF?

Responses

1. It is essential to have a MTEF for there to be any resource predictability.
2. If the country is prone to shocks, then setting up a highly elaborate MTEF process is not recommended. In Uganda, it took six years before the government felt confident about the process. Therefore, it is very important to consider design when implementing a MTEF. Ghana may be an example of where donors have tried to go too far too fast. A Fund team that assessed the MTEF examined the line items in year one and year three – they were exactly the same, suggesting that the MTEF had not been updated to reflect shifts in policy priorities.
3. Doing a MTEF is a difficult process even when the Bank takes on a participatory approach. TA operations led by the Bank have not been very successful at building capacity. The most sustainable solution is for the government to increase its own capacity.
4. It is certainly far more difficult to plan for poverty reduction when a country does annual planning. Hence, the MTEF is an extremely important instrument in the context of the PRSP.
5. It is possible to map from inputs to outputs, but it is more difficult to map from inputs to outcomes. One suggestion would be to do an ex-post exercise to assess impact. There are some examples of robust analyses done in Malaysia, but having good data is a prerequisite.
6. Emphasis has to be placed on the top-down process to begin with. Discussions should start with the Mof F, but requires the cooperation of the line ministries. It would be very difficult to succeed unless there is a high degree of commitment for the top level of government.
7. General budget support is compatible with the MTEF, but project funding is probably not.