Practical general equilibrium estimation of resource pulls under trade liberalization

This paper discusses the application of a model similar to the one developed in Leif Johansen's Multi-sectoral study of economic growth (1964) to the problem of determining general equilibrium responses of the economy to changes in commercial policy. This method amounts essentially to specifying a log-linear approximation to the general equilibrium solution for the economy, and solving the resulting linear equations for changes in endogenous variables as functions of exogenous variable changes. For a 35-sector model of the Chilean economy with labor as the only variable factor of production (to avoid the problem of overdetermination of many commodity price and output shifts when only two factors are considered in constant returns production functions), it is found that (i) the specification of the way in which intermediate inputs enter the production function is numerically important in determining output responses to tariff changes, detracting from the credibility of fixed coefficient effective rate of protection calculations if variable intermediate input coefficients are the rule (as appears likely empirically); (ii) exchange rate elasticities with respect to individual tariff changes are fairly large, so that the usual partial equilibrium assumption of exchange rate insensitivity to ‘small’ tariff revisions is not valid; (iii) employment effects of different tariff revisions are highly variable and in some cases substantial.