This study presents the development of a nonlinear regional input-output model. This model is developed to address a number of the limitations of the standard input-output model. Some of these limitations include the assumption of fixed or average relationships between the productive sectors of the economy and constant labour productivity. This allows the model to move away from the strictly linear behavioural relationships to incorporate greater behavioural detail through the application of marginal relationships in key components of the standard model including primary inputs.
The input-output table construction techniques are applied using sophisticated hybrid table construction techniques. Specific techniques include the application of the augmented fleggs location quotient and insertion of a wide range of superior data. This provides a deeper understanding of economic connectivity within industries in regional economies and these relationships are explored in Australia’s major mining economies: Western Australia and Queensland. Net multipliers are also compared with gross output multipliers to address the issue of overestimation of sectoral contribution within an input-output framework.
It is argued that the nonlinear input-output modelling developed for this study provides a better platform for regional economic impact analysis than the standard input-output model. In particular, this model is geared towards providing more realistic estimates of income and employment impacts which is driven by marginal relationships versus average relationships with the latter a principal feature, and limitation, of the standard input-output model. This is demonstrated through the application of an economic impact analysis study applying both the standard model and the partial nonlinear input-output model within scenarios of mining development in regional economies in Western Australia.
With policymakers always interested in being able to predict the employment and income impacts of policy changes and major projects, it is argued that this model can provide a useful vehicle to understanding such impacts. This model is particularly suited to simulating short-term impacts at the regional level and when time and data constraints prevent the use of more detailed interindustry modelling techniques including computable general equilibrium modelling.