Discussion on Part III

Abstract

The invited discussant of the paper, Professor Fernando de Holanda Barbosa, averred that the rate of off-farm migration was not a function of wage differentials alone. Other variables such as educational facilities, housing, etc., played an important role in the landless labourer’s decision to migrate to urban areas. The government’s role in investment allocation should have been represented in the relevant equation. The analysis ought to have been in terms of amounts, net of tax, as there were large differences in the tax treatment of the agricultural and non-agricultural sectors. The risk variable needed to be included in the immigration equation. The analysis should also have included both the price of land and the ratio between the price of land and the lagged price of land as explanatory variables. The estimation should have taken into account both expected and unexpected components and problems of bias, and inconsistency should have been considered. For long period studies like the present one, unstable coefficients were to be expected as coefficients changed over time.