WSP05559 - Laserfiche WebLink

Home Browse Search

(2) The projections and measures provided by the model, however, should be interpreted in the light of the assumptions and limitations which are embodied in this particu lar \IlOde ling methodo logy. Some of the major assumptions and limitations have already been described above. These and other potential shortcomings of the approach will be discussed in more detail in what follows. Fixed Coefficients The fixed coefficients which characterize sector interactions per unit of output presume a linear, homogeneous production function over all levels of output. While the coefficient estimation employed here attempts to anticipate technological change there will be some changes J (especially those involving the interaction of technological choice and relative price) which, no doubt, will not be captured. Moreover, the linear nat)lre of the production relationships cannot be modified in this framework; at best, one can only make linear approximations of any nonlinearities 'n / production. The fixed character of input-output coefficients is especially limiting in the case of the coefficients which represent interregional trade. Trade coefficients have been observed to be more volatile than process coefficients. Mitigating this concern to a certain extent is the fact that the array of interregional trade coefficients are sparse and have very small entries. Fixed coefficients do not allow for other dynamic features of economic growth observed by economists: agglomeration economies and economies of scale. As a region grows, the basic sectors of the region tend to attract additional, similar types of activity to the region (agglomeration) and as regions grow, the larger markets for goods previously imported to the region now attract producers of those goods to the region. Both of these forces would tend to accelerate or sustain growth beyond the "development" scenarios