Modeling input-output impacts with substitutions in the household sector: A numerical example

Although there have been many elaborations of the basic input-output approach, including multi-regional models, dynamic models, models with variable coefficients, supply-side models, etc., these approaches all have the same limitation. The fixed-coefficients production function assumptions ignore substitutions in response to price changes that can be expected to accompany most shocks- skipping over the heart and soul of market economics. This research note suggests a simple approach to estimating new technical coefficients matrices after a shock so that the consequences of short-term substitution effects can be studied. Given a reduction in income (as reflected in the value added row), households are likely to make substitutions, reducing their final demand by less than the application of base-year I-O coefficients would indicate. But if ex post changed income and consumption can be observed, the application of RAS procedures can generate an appropriately modified A matrix. The resulting set of interdependent substitutions that occurred can be identified. Due to some well known limits in applying the traditional RAS approach, we reformatted it and suggest a new economic model that can link coefficient adjustments to degrees of a priori substitutability and complementarity. Based on this resolution, we look forward to detailed studies of specific coefficients and how they evolve over the short term.