2021•Applied EconomicsOpen access

Foreign-domestic substitution, import penetration and CGE modelling

Kenneth W. Clements, Marc Jim M. Mariano, George Verikios

Open full text 11 citations

Abstract

Foreign-domestic substitution elasticities (the so-called ‘Armington elasticities’) determine the degree of competitiveness in demand between similar products produced in different countries and are key parameters in a variety of numerical models of international trade. Armington elasticities are part of the explanation of the large increases in market shares of foreign products relative to locally produced ones in Australia, for example. The existing literature provides only limited evidence on these elasticities for Australia with the most disaggregated produced some time ago in 1977 by Alaouze and colleagues.This paper provides up-to-date parametric estimates of Armington elasticities for Australia with a reasonable degree of sectoral disaggregation. We use 22-years of data for 20 types of merchandise commodities, using OLS, panel and restricted-panel approaches. Our estimates range from 0.30 to 2.26, with higher elasticities for Transport and Equipment products and lower ones for Energy and Minerals. We illustrate the use of our elasticities with a trade-policy simulation using a computable generable equilibrium model of the Australian economy. We analyse the sensitivity of the results to the Armington elasticities by also using the previous estimates of Alaouze and colleagues. We find an overestimation of economic effects when using the old Armington values.

About this research paper

What this paper is about

Foreign-domestic substitution elasticities (the so-called ‘Armington elasticities’) determine the degree of competitiveness in demand between similar products produced in different countries and are key parameters in a variety of numerical models of international trade. Armington elasticities are part of the explanation of the large increases in market shares of foreign products relative to locally produced ones in Australia, for example. The existing literature provides only limited evidence on these elasticities for Australia with the most disaggregated produced some time ago in 1977 by Alaouze and colleagues.This paper provides up-to-date parametric estimates of Armington elasticities for Australia with a reasonable degree of sectoral disaggregation. We use 22-years of data for 20 types of merchandise commodities, using OLS, panel and restricted-panel approaches. Our estimates range from 0.30 to 2.26, with higher elasticities for Transport and Equipment products and lower ones for Energy and Minerals. We illustrate the use of our elasticities with a trade-policy simulation using a computable generable equilibrium model of the Australian economy. We analyse the sensitivity of the results to the Armington elasticities by also using the previous estimates of Alaouze and colleagues. We find an overestimation of economic effects when using the old Armington values.

Why it matters

OpenAlex reports 11 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Foreign-domestic substitution elasticities (the so-called ‘Armington elasticities’) determine the degree of competitiveness in demand between similar products produced in different countries and are key parameters in a variety of numerical models of international trade. Armington elasticities are part of the explanation of the large increases in market shares of foreign products relative to locally produced ones in Australia, for example. The existing literature provides only limited evidence on these elasticities for Australia with the most disaggregated produced some time ago in 1977 by Alaouze and colleagues.This paper provides up-to-date parametric estimates of Armington elasticities for Australia with a reasonable degree of sectoral disaggregation. We use 22-years of data for 20 types of merchandise commodities, using OLS, panel and restricted-panel approaches. Our estimates range from 0.30 to 2.26, with higher elasticities for Transport and Equipment products and lower ones for Energy and Minerals. We illustrate the use of our elasticities with a trade-policy simulation using a computable generable equilibrium model of the Australian economy. We analyse the sensitivity of the results to the Armington elasticities by also using the previous estimates of Alaouze and colleagues. We find an overestimation of economic effects when using the old Armington values.

Key concepts: Computable general equilibrium, Economics, Econometrics, Substitution (logic), Elasticity of substitution, Panel data, Parametric statistics, International economics

Related papers

Back to paper searchBrowse research topicsOriginal source
Foreign-domestic substitution, import penetration and CGE modelling — Research Paper | ScholarLens