Social accounting matrice and computable general equilibrium modelling in South Africa with reference to the financial sector

Abstract

A significant challenge facing South African economists is the formulation of appropriate policies to redress the unequal distribution of income and wealth. In many other developing countries, a similar challenge has led to the use of Social Accounting Matrices (SAMs) and Computable General Equilibrium (CGE) models to give quantitative support in the formulation of income redistribution policies. A SAM can be defined as a numerical representation of the economic cycle with emphasis on income distribution. A CGE model can be defined as an economy-wide model that includes the interaction between demand, income and production structure, and where all prices are adjusted until decisions made in production are consistent with decisions made in demand. SAMs and CGE models are particularly well suited for analyses of issues such as income redistribution, since they impart a general equilibrium approach to analyses, as opposed to the traditional partial equilibrium approach. A SAM and CGE culture is lacking, however, in South Africa. This is implied by the fact that almost 35 developing countries already had SAMs before South Africa did, that the most recent South African SAMs, for the year 1988, are "preliminary" and that applications of the SAMs are limited. One of the reasons noted for the limited use of the South African SAMs is their omission of the financial sector. Since South Africa is characterised by a "first world" economy alongside a "third world" economy, the interactions between the developed financial sector and the rest of the economy may be complex. Because these interactions could affect income redistribution policies they should be taken into account, preferably through a general equilibrium framework.

The primary objective of this study is to contribute towards a suitable framework for analysing income redistribution and growth in South Africa by extending the present South African SAMs to include the financial sector. The secondary objective of this study is to contribute towards establishing a SAM and CGE modelling culture in South Africa. The primary objective is achieved by illustrating how the national financial accounts can be rearranged and incorporated into a SAM. This method is used to compile four different financial SAMs for South Africa. Two financial SAMs are for 1988 and contain only flow of financial variables, while the other two are for 1990 and contain both initial and end of the period stocks of financial variables in addition. The secondary objective is achieved by first tracing the origins and development of SAMs and CGE models and arguing that SAMs and CGE models are a culmination of developments on respectively the inductive and deductive sides of economics. Second, the principles of SAMs are discussed and a basic SAM derived from fundamental economic principles. It is shown that a SAM represents the economic cycle, and is therefore suitable to serve as a consistent database for a CGE model. Third, South Africa's existing SAMs are critically discussed, and the results for income redistribution policies from simple CGE models based on these SAMs are presented. Fourth, it is shown how the South African SAMs can be used as a basis for an extended CGE model. Fifth, the SAMs' weakness of omitting the financial sector is illustrated by discussing the role of the financial sector in an economy. Special reference is made to the role of the financial sector in developing countries. This discussion also serves to determine the theoretical foundations of the national financial accounts and to determine the possible issues that can be investigated through a financial SAM. Finally a simple CGE model, based on a financial SAM, is used to investigate income redistribution policy in South Africa. The results are combined and

contrasted with those obtained from CGE models based on the existing South African SAMs. One contribution made by using a financial SAM is to warn of the danger of income redistribution policies turning into a "populist" experiment over the short term. The results also suggest that redistribution in South Africa should be a long-term strategy that should be accompanied by a better provision of education and a restructuring of the economy. Indications from the models on the form this restructuring can take are noted. The most important issues identified as worthy of further research are the (i) construction of large, multi-sector models based on the financial SAMs; (ii) the provision of a comprehensive manual on compiling and using SAMs in South Africa; (iii) the compilation of a SAM for South Africa for 1990 that includes a distinction between rural/urban households; (iv) the construction of a "structuralist" CGE model for South Africa to be contrasted with the Walrasian, neo-classical models that were discussed in this study; (v) the compilation of updated real as well as financial SAMs for different regions in South Africa; (vi) the use of regional financial SAMs to investigate the effects of a "regional liquidity reserve" for banks; (vii) the compilation of more accurate data on the stocks of financial assets and liabilities owned by the major institutions in South Africa; (viii) the investigation of ways through which liquidity preference in South Africa could be lowered; and (ix) the investigation of the effects of diminishing the high degree of concentration in South Africa's financial sector. It is concluded that although SAMs and CGE models are capable of providing fresh insights on policy options, they remain auxiliary instruments to assist policy makers, and cannot replace human ingenuity and compassion in the search for appropriate policies to redress the unequal distribution of income and wealth in South Africa.