Development and Application of a New Zealand General Equilibrium Model

Abstract

Whether a country gains or loses from dismantling protection is a question which has received much attention in overseas studies; studies which deal both with the relevant theory and with actual measurement. The topic has not been well analysed in the New Zealand context. Discussion amongst economists and other interested parties has certainly occurred but this has been based more on philosophical and political considerations than on applied economic research. Since questions of protection reform affect the whole economy it is inappropriate to study such problems in a partial or selective framework which cannot capture the interdependencies between each and every sector in the economy. A multi-sectoral general equilibrium model overcomes this deficiency. This thesis is concerned with the development and application of such a model. The model (named JULIANNE) is a medium term policy model designed to answer 'what if' type questions, particularly questions about trade and structure. It is not a forecasting model. Its role is rather like that of a laboratory in the natural sciences, where experiments can be conducted in a situation where certain aspects of the (economic) environment can be controlled by the researcher so that it is possible to measure the relationships between the variables of interest. The closer the environment is to the 'real world' the easier it is to apply deductions from the experiment to reality. But even quite artificial experiments can yield useful insights. The thesis comprises eleven chapters, the first three of which introduce and develop the model, examining some of the overseas general equilibrium models and assessing some of the problems which need to be addressed when constructing such a model for New Zealand; a model with an emphasis on trade and structure. The following three chapters present the JULIANNE model including its equations, a detailed explanation of its features and routines, and its method of solution, which for general equilibrium models is a most important consideration as it distinguishes the purely abstract Walrasian model from a model which is actually computable. Chapters 7 and 8 apply the model to various problems, especially to protection reform, but also to other interesting topics such as export subsidisation, relative occupational wage rates and medium term projections. The issue of model validation (in a general sense) is also covered. In Chapter 9 the model is extended from a single period snapshot model into a multi-period dynamic model, essentially introducing another variable; time, that can be controlled by the experimenter. Some of the results from Chapters 7 and 8 are then reassessed with the extended model, as described in Chapter 10. Results from the application of the model to questions about the effects of changes in protection enabled one to conclude that under flexible factor prices with fixed factor employment, the gains from freer trade vary directly with the values of the export price elasticities of demand, with the potential for economies of scale arising from specialization, and with the time horizon under consideration. They vary inversely with the values of the elasticities of substitution both between domestic and imported goods of a given type, and between goods of different types. Under a different labour market asumption, namely fixed real wage rates and flexible employment, the case for free trade is much stronger (that is, for a given set of parameter values). The profile of protection across sectors can also be important with the not improbable chance that a low uniform level of protection is superior to complete free trade, again depending on parameter values and the characteristics of the labour market. In this connection the observed uniformity of the current protection regime is very dependent on the degree of sectoral disaggregation identified in the model. As the degree of disaggregation increases, the potential for specialization also increases, as does the potential for substitution between different commodity types. Just how important these issues are, is a question for future research.