Optimisation of a sixteen sector model of the New Zealand economy

Abstract

The purpose of this paper is to report on the empirical application of a linear programming model of the New Zealand economy, and to draw some conclusions which may be of use to national policy makers. Although the mathematical structure of the model and its relationship to economic theory have been described in detail elsewhere, a general description is given in Section II for the sake of completeness, and this should be sufficient for readers primarily interested in the quantitative results and their implications. The principal conclusions that can be drawn from the analysis regarding medium and long term economic development in New Zealand are as follows: (1) Given that expansion of markets for manufactured goods will be difficult, it is important that markets for primary products exported from New Zealand should be maintained and that investment in farming should continue at a high level so that the current high standard of living in New Zealand should not deteriorate. This policy should apply even when agricultural export prices are low. The economy is particularly sensitive to the loss of markets for processed primary products. (2) Expansion of the economy and improvement in living standards are more dependent on the encouragement of new products and new industries than on traditional ones. In particular, serious efforts should be made to find further export markets for processed forestry products and for secondary manufactured products. (3) An intensive immigration policy would not be of immediate benefit to the economy. Additional labour is required, but it is only after export markets have expanded that the economy is likely to be able to stand the additional pressure on foreign exchange as well as on other resources. At the same time a small amount of immigration is recommended. It stands to reason that, where possible, the immigrants admitted should have skills in industries which have export potential. (4) There should be some encouragement of import substitution industries which will significantly ease the balance of payments. This should not, however, at least on existing evidence, be a major policy priority. Perhaps the most important comment which should be made with respect to this investigation concerns the limitations of the study. A number of theoretical and practical shortcomings associated with the linear programming technique are cited elsewhere. However, the author’s view is that the most serious difficulties are related to the quantity, quality and timing of data available. By comparison, many of the theoretical objections to the use of programming techniques (and input-output models in general), are small in importance or can be assuaged by making appropriate approximations. It goes without saying that more detail and more accuracy of data would be helpful, but perhaps these problems are overshadowed by the considerable time lag that exists between the publication of data that is available and the moment at which that data is in a suitable form to speculate about the future. As things stand the future is more often the present and in some cases the past by the time analytical results are available to policy makers. Consequently, this study must be regarded as exploratory. The author is of the opinion that a model of this nature is only of real use if the component parts can be assembled quickly. For this reason inter-industry analysis should no longer be a one-man occupation, but be carried out by a team. This would tend to speed up the data assembly operation, and a very important fringe benefit would be a tendency towards greater accuracy and detail.