ABSTRACT
The critical role of macroeconomic and fiscal modelling is well established in both developed and developing countries. Where quantitative tools are inadequate, macroeconomic management may be undermined, in turn heightening the risk of economic imbalance. Weaknesses in this domain played an important part in the economic crises which beset much of Africa during the 1980s (Tarp 1993). Despite the known limitations of formal models, they are useful, if only to make core assumptions explicit and to help think through the inherent trade-offs with respect to stability, growth and development. Even when quantitative tools are simply absent, policy choices always refer to some form of implicit model – whether logically consistent or not. Quantitative tools typically also have a central place in budget management, providing the basis for identifying government financing needs and setting budget ceilings. In the current environment, robust tools to inform policy-making are growing in importance. The Economic Commission for Africa (2005), for example, notes that the forward-planning orientations of both MTEFs (Medium Term Expenditure Frameworks) and PRSPs (Poverty Reduction Strategy Papers) demand increasingly robust planning tools. The prospect of scaled-up aid flows represents a further source of pressure on macroeconomic policy-making. Given the above, it is useful to assess the extent to which modelling
approaches at the country level are adequate to economic management challenges. Although one might expect this to be a fairly routine activity, one notes a distinct lack of published material providing guidance on best practice in macroeconomic and fiscal modelling in developing countries. In other words, it is difficult to identify broadly accepted grounds on which country modelling practices may be assessed. In response to this gap, an initial
objective of this chapter is to propose a general set of principles for the evaluation of macro-fiscal modelling at the country level. These themes are developed in Section 2, including a brief review of the relevant literature. Based on the authors’ experiences in macro-fiscal modelling in Mozambique, Section 3 applies the framework to this specific case. We find that modelling practices have improved over recent years. They are, however, at an early stage of development and a number of weaknesses are obvious. In particular, we note the absence of tools to think through the relationship between investment (allocation), external resource flows and growth over the medium term. In addition, the informality of modelling practices tends to undermine their consistent use in the policy process. Our evaluation concludes with a consideration of the deeper factors that may determine improvements in modelling practices over the near term. Section 4 brings the chapter to a close. At this point it is useful to clarify the position taken in this chapter. Our
focus is on the practical side of modelling rather than its theoretical basis. This is not to say that theory is irrelevant. Theoretical positions are fundamental to building and interpreting sound models. However, robust practical implementation is a prerequisite to inform policy-making. The point here is that models can be valid partners at the table in many key fiscal and macroeconomic policy processes (Don 2004). Modelling practices determine whether quantitative tools are adequate to policy challenges and/or have a clear ‘voice’ at the table. Our interest in macro-fiscal modelling practices thus concentrates specifically on models used within the government on a regular basis.
