In my attempt to be both brief and trenchant, I seem to have confused Gil with respect to my use of power as the determinant of executive incomes and the uselessness of the neoclassical framework to try to justify CEO's pay and perks. I will try to be more clear in the following elaboration. The concept of marginal productivity involves the addition of a single unit of the variable factor (which must be homogenous with previous units of the factor or it is impossible to sort out the productivity of what). Now, if we add a CEO to an existing firm, is his mp the total value of the firms output (on the assumption that the firm can not operate without a CEO)? Or is it the change in TP when a second CEO is added? (an obvious contradiction>, [D? Or is it the change inTP when one CEO is replaced by another? This then would indicate that all CEO renumeration (subtracting opportunity wages) is a form of rent. (i.e. the rent to a natural or developed talent e.g. the return to Wayne Gretsky's hockey skills.) However, as Ricardo pointed out, rent is a result of price, not a cause of price. Since CEO's are in a position to influence price through market power, they are also in a position to some extent to determine their rents. However, this is rather tortuous analysis and the concept of marginal productivity is so unreal (we have gone through all this before) that neoclassical theory in this regard "has no clothes". In any case, all rents in the long run are a return to power, either in the form of ownership rights that include the right to restrict output, monopoly market power, power of the office to allocate rent, etc. To quote Marc Lavoie's comment on the importance of power: "...power is the ultimate objective of the firm: power over its environment, whether it be economic, social or political. 'Power is the ability of an individual or a group to impose its purpose on others'. (Galbraith, 1975, p. 108) The firm wants power over its suppliers, over its customers, over the government, over the kind of technology to be put in use.... THE NOTION OF POWER, EXCEPT WHEN RELATED TO THE CASE OF THE PURE MONOPOLY, HAS BEEN SYSTEMETICALLY IGNORED IN ECONOMICS, WITH THE EXCEPTION OF INSTITUTIONALISTS AND MARXISTS." (Lavoie, Foundations of Post-Keynesian Economic Analysis, pp 99-100) In short to deal with the issue of power in income distribution we have to leave the certai, equilibrium world of neoclassical economics and utilize the models of surplus (post-classical or heterodox) economics. It is here that the fundamental issue of power is joined. It Becomes the question of who has the power to distribute surplus. Why do American CEO's receive much greater incomes than do Japanese or European CEO's? Why do CEO's of private utilities receive greater remuneration than _the same_ CEO's received prior to privatization despite no change in productivity? Why do CEO's of profit losing firms get commensurate remuneration with those of profit making firms? (etc. etc.) none of which can be explained by mp theory or even with any reasonable application of neoclassical rent theory. However, they can all be explained within surplus models by modelling the sources and distribution of power (although not necessarily in an econometrically operational sense.) Many Marxists, for instance, talk about working class bargaining power over distributive shares in terms of the size or proportion of the reserve army of unemployed. This was the context in which I used the example of the mugger (which has been used on this list in the past in more or less this context.) The mugger does not produce any marginal product, but his power over the use of force allows him to redirect, to himself, part of the surplus in the form of the above subsistence wages of the muggee. I trust this makes sence of my earlier elliptical post. Paul Phillips, Economics, University of Manitoba