NEO-POPULISM AND MARXISM 13 from others). It may be noted that this violates the assumption of competitive markets on which the neoclassical theory is based: with a market in labour-hiring, under the neoclassical perfect competition assumptions, an equilibrium cannot exist of capitalist farms coexisting with differing returns to hired labour and to family workers; no equilibrium can exist, either, of the two types of holdings coexisting with differing output per acre, if there is a competitive market in land renting. Thus, Chayanov's theoretical framework has this odd feature that it adopts a large part of the neoclassical theory (utility maximization) combined with implicit non-adoption of the rest (perfectly competitive markets, or indeed any markets at all in labour-hiring and land leasing insofar as these theoretical exerciser are concerned). Elsewhere, however, Chayanov does refer to labour-hiring: inadequate land or fall in product price obliges the peasant to seek employment on wages. This suggests that he was far from consistent theoretically, and did not see the implications of his-own numerical example comparing the family farm and capitalist farm. The modern models of peasant equilibrium, such as A K Sen's mentioned earlier, do operate with the explicit assumption of competitive markets. But they reconcile the existence of equilibrium in which "family farms" and "capitalist farms" coexist with identical production functions but differing returns to labour and differing output per unit area, by saying that "in reality" markets are imperfect. This incidentally is another example of the use of tautology. As long as the subjective equilibrium conditions in these models do not violate the assumption of perfect competition, we hear nothing about ''in reality" imperfect markets; as soon as the subjective equilibrium conditions (which, as we have seen, merely translate the given observation into subjective terms) do turn out to be inconsistent with the perfect competition assumption, the "in reality" imperfection of markets is invoked. Either way, the utility model definitionally holds good. (The only thing sacrificed is reality; which, chameleon-like, is now perfect and now imperfect depending on requirements!) Leaving this tangle of theoretical inconsistencies and tautological reasoning, let us approach the whole question from the alternative perspective of conceptualizing on the basis of historically-given reality: the procedure of classical political economy in general and Marxism in particular. What does this have to say on the determinacy of "consumption" for family-based producers and on the wage-rate? In reality, it is not the case that in history family farms have been able to confine themselves to producing only for their own—subjectively determined—consumption needs. On the contrary, typically, they have been under the objective necessity of producing a surplus for payment to the feudal overlord or to the state as rent or tax, as a condition for producing their subsistence. "Consumption" therefore is nothing but the historically-determined "necessary labour", the share