Baseline Expenditure Projections and Employment Inflexibility

The importance of good baseline expenditure projections – discussed in recent blog entries by myself and Jordi Baños Rovira – has just been underlined further by the European Commission’s recently-published proposal to make it compulsory for member countries to produce detailed medium-term expenditure (as well as revenue) projections “based on unchanged policies”. This underlines further the pressing need to improve methodologies for estimating baseline expenditure. A key underlying problem is that expenditure forecasting is much more difficult that revenue forecasting because much expenditure policy is not explicit. Revenue policy is clearly formulated in law. Expenditure policy – with the exception of mandatory social transfers – is for the most part neither formulated in law nor clearly stated. The difficult question arises of how to interpret “existing expenditure policy” when that policy is not explicit.

One question is whether to use an output or input-oriented interpretation of “existing expenditure policy”. It is often assumed that existing expenditure policy refers primarily to an explicit or implicit position taken by government about the level of services to be provided to the community. In this spirit, it could be assumed that the implicit expenditure policy is to increasing service provision in line with demand so as to maintain the same availability of services to citizens. This is an output-oriented conception of existing policy because it is the level of services to be provided which drives expenditure. The output-oriented conception is entirely appropriate in some countries, particularly those with more performance-oriented government and a high degree of expenditure flexibility.

However, the output-oriented interpretation of baseline expenditure is much less appropriate in many other countries. The degree of flexibility of the government in personnel expenditure is a key relevant factor here. If government can hire and fire at will, decisions about the level of services to be provided to the public will determine government current final consumption expenditure. But if, by contrast, government workers enjoy high levels of job security, this is not the case. Instead, current final consumption expenditure will be driven primarily by decisions taken on government employment levels, and may or may not be closely linked to government views on the level of service to be provided. It will, in other words, be input rather than output-driven.

This has a number of important implications. The first is that if government has an explicit policy on staffing levels, baseline expenditure estimates must take that policy explicitly into account. France is a good case in point. It has, at present, an explicit medium-term policy to replace only one of every two departing civil servants, in order to slim down its bureaucracy. Its medium-term projections of current expenditure (excluding transfer payments) should under these circumstances be mainly based on the implications of this policy for the evolution of the government wage bill.

The second implication is that, in countries with rigid public employment systems – which includes most low-income countries – the adoption of an explicit medium-term policy on public employment levels is an essential part of medium-term budgeting. In such countries, such a policy provides a firm basis for meaningful baseline expenditure projections, and helps to overcome the widespread problem of MTEFs with expenditure projections which have no relationship to actual budgetary policies.

The third and final implication is that technical assistance to low-income countries on the detailed methodologies to be used for projecting the evolution of the wage bill would be enormously valuable, not only to help them adopt a more medium-term budgetary perspective, but also to improve annual budgeting.

Comments are closed.