Role of government spending in a dual economy model: A theoretical analysis

This paper extends a two-sector, general-equilibrium model of rural-urban migration to include government spending. Provision of public goods acts as a productivity-enhancing input in private production that results in external economies of scale. This approach is generalized by introducing an unbalanced allocation of public expenditure in rural and urban sectors due to political economy considerations, differential sector output elasticities with respect to government input, and distortionary taxation. The paper studies the effects of an increase in public spending and taxation on sectoral outputs, factor prices, urban unemployment, and welfare. Of particular concern here is to study the effect of an unbalanced allocation of government spending between rural and urban areas.

This paper starts with a full-employment model in which the rural-urban wage differential is exogenously given. The paper then presents the model in terms of rates of change, relying heavily on Jones (1965), and lays out the basic equations for comparative static analysis. This is followed by an analysis of the stability condition of the system, following the classical contribution of Mayer (1974) and Yabuuchi (1992). The paper then shows, through comparative static results, the effects of a change in the share of the rural sector in the government budget and a change in the tax rate. This section also analyzes the welfare impact of changes in the allocation and tax policy variables. The paper then introduces urban unemployment and the Harris-Todaro equilibrium into the two-sector model with government spending. The impact of fiscal policy that changes the share of government expenditure in rural areas is analyzed. The conclusion to the paper includes suggestions for future research. We have attempted to make one calibration using Cobb-Douglas production functions (like Temple, 2005) and we found that with these results, it is clear that government expenditure in rural areas is more effective in reducing rural-to-urban migration in the Harris-Todaro model than in the full-employment neoclassical model. Sizable urban unemployment gives a stronger incentive for rural labor to stay in agriculture, and also attracts some unemployed labor from the other sector, when public expenditure raises the marginal productivity of labor in rural production. Of course the results are dependent on specific parameter values.