problems of input-output models and some to the manner in which repercussions on
the environment are modelled. The problems include:
(i) Utilising the input-output model as a production function in an economic model
means that strong assumptions are made on the linearity of production processes: that is, input coefficients are treated as fixed.
(ii) As a result, substitution between inputs resulting from pollution abatement is
easily neglected.
(iii) The effects of international competition between national industries are not
taken into account.
(iv) Little attention is given to other means of reducing the emission of pollutants.
For instance, when the importance of environmental conservation is acknowledged, new capital goods may have a quite different technical profile from
existing capital stock. This may include the more efficient use of other raw
materials and the recycling of waste products.
(v) Even though pollution problems are conceptually dealt within distinct sectors,
the fact is that pollution is often treated by use of end-of-line equipment
installed by the polluting company.
These limitations on the use of input-output analysis have led to important
adaptations. Methods have been developed to adapt I-O coefficients to changing
technological circumstances for example, Linear Programming and its extended
methods, and quadratic/Least Squares methods.
Another major drawback is the amount of time and effort involved in collecting
the basic data. It is unlikely that many developing countries possess national inputoutput tables with data on environmental quality. To construct even a modest inputoutput model requires several years of work and a great deal of co-operation from
government and industry. There is a risk that by the time the data have been
compiled, the model for making accurate policy appraisals will already be outdated.
Finally, even though the “best” strategy can, in principle, be found by iterative
procedures, the process can become cumbersome and frustrating for decisionmakers: especially when large numbers of variable and diverse opinions are
involved.
8.9.6 Linear Programming Models
There has been some emphasis on the significance of accounting for environmental
quality, both locally and countrywide through a general equilibrium model.
However, complex situations involving a high number of variables and several
choices, are probably better suited to mathematical programming models. Linear
programming and its associated models have been widely used to solve environmental quality problems and optimise the results.
Linear programming deals mainly with the allocation of scarce resources. The
model aims to optimise a predetermined goal, or set of objectives, given a set of
8.9 Macroeconomic Models
243
Précédent

- 271/700

Suivant