The theoretical model of labour supply used so far has provided a convenient laboratory within which to explore different approaches to the formation of public policy. For this purpose, the simplicity of the model has been an advantage. However, there can be little doubt that, in order to compare the realworld impact of the basic income scheme with that of the existing alternative of social insurance, a richer theoretical framework is necessary. A key element in the comparison is that the existing social security provisions are tied to specific contingencies such as unemployment or sickness, and we need therefore a model in which contingencies of this kind can arise. Unemployment may appear in the optimum taxation frame- work outlined earlier, in the sense of people choosing to work zero hours, but an adequate treatment needs to take account of such factors as efficiency wages, segmentation of the labour market, and involuntary unemployment.
The introduction of these considerations is particularly relevant to the incidence of the policy reform. In the simple optimum taxation analysis it was assumed that the factor prices (and, implicitly, the product prices) are unchanged by the introduction of the Basic Income/Flat Tax. There is assumed to be an infinitely elastic demand for labour of each quality at the specified wage rate. In contrast, the models of general equilibrium tax incidence of the type developed by Harberger ( 1962) tend to make simpler assumptions about the distribution of income, but to allow for changes in factor and product prices. They too are of the Arrow - Debreu type, with unemployment only appearing if labour is in excess supply at a zero wage rate, and what is needed is to extend these models to bring in other explanations of unemployment.