The limits of the efficiency wage theory[Les limites de la théorie du salaire d'efficience]

Efficiency wage theory, evolved by the "new keynesians", is part of a research programme concerned with explaining "involuntary" unemployment as part of the "general equilibrium". So, these works are a synthesis attempt of neo-classical and keynesian analyses in the employment area. Indeed, the authors place their analysis in a walrasian framework, but they integrate a keynesian concept to complete their models. In order to achieve their purpose, the efficiency wage theorists formulate two propositions on which their argument is based: 1° the relation supposed rising between the rate of wage and the productivity of wage earners, 2° the determination of the rate of wage outside the market. We are going to show that these propositions are not fitting with the general equilibrium. On the one hand, it's not the determination of the rate of wage which determines the marginal productivity of producing services. On the other hand, the rate of wage being known only at the equilibrium in the walrasian model, contractors couldn't change this rate. But we can't at one and the same time place oneself in a theorical setting without respecting these foundations. In other words, the efficiency wage theory is based on an incoherent argument, it means that this theory does not provide anything to the economic analysis. And more generally, it appears that we shouldn't unify neo-classical and keynesian theories whose foundations are profoundly distinct. Whereas the general equilibrium is based on an exchange economy, the keynesian system is based on a production monetary economy where the product is measured from the production by the sum of the monetary remunerations of producing services. So, any synthesis attempt of neo-classical and keynesian theories in the employment area is doomed to failure.