This paper covers some essential ideas of a much larger study by James Dean and myself of the way in which the behavior of economic agents, when aggregated, contributes to the process of inflation. Space limitations dictate a compromise between brevity and minimal completeness. Therefore many of the basic ideas will be stated baldly. Qualifying remarks are omitted. The conventional macro approach can be subsumed under the slogan of too much money chasing a given amount of goods. It does not explain how specific absolute prices are set. Further, a theory of money demand creation from a micro viewpoint is missing. I shall emphasize the world where real live flesh-and-blood humans set prices, and where their activities influence the demand for money. The essential scheme is simplicity itself. Inflation is determined by incentives that lead sellers, using decision rules based mostly on conventional behavior, to raise absolute prices, which in turn creates an increase in the demand for credit which, in its turn, the banking system accommodates to some degree. Can we start with a set of reasonable postulates about firms and markets consistent with this type of firm behavior? I believe we can. The postulates I focus on (taken from X-efficiency theory) can be summarized by the following phrases: 1) inert areas, 2) mostly nonmaximizing behavior, 3) incomplete employment contracts, 4) imperfect markets (including some bargaining power), and 5) a take-it-or-leaveit bargaining style.

Louis De Alessi's paper (1983) begins in an ecumenical spirit and suggests that rights and the X-efficiency approach have not drawn on each other's contributions. I share that spirit. There is much to learn from a variety of approaches. But, as De Alessi proceeds, the ecumenical spirit withers. last three sections argue to the effect that neoclassical (i.e., rights theory)' obtains the same results as X-efficiency theory, but has a superior methodology. De Alessi's paper puts me in the position of a reluctant dualist. His approach resembles an intellectual takeover bid since it argues 1) that neoclassical is flawed, 2) that the rights approach has corrected the flaws, and 3) by implication that it supercedes other micro approaches.2 concepts of rights, transaction costs, and adjustment costs, and their motivational implications, are certainly useful. I see no reason to argue against them, or against rights theory. But I believe that there is more to the explanation of inefficiency. Thus I will stress the special nature of the X-efficiency approach, examine our differences, and attempt to correct some of De Alessi's misinterpretations. Nevertheless, I believe that De Alessi's article renders an important service since it gives a succinct summary of rights and explains its significance. main differences between us lie in the basic presupposition behind the two approaches. In answering the motivation question, rights seems to say that rights are the only motivating force, and that people act to fully maximize wealth given the limits of their rights. When they do so, their activities are efficient. X-efficiency approach raises a different type of question. Namely, it asks (within the firm and outside of the exchange relationship) what are the motivating forces that determine effort and productivity, including the motivating forces leading to maximizing or less-than-maximizing behavior. Property rights are, of course, a motivating force. However, it is one of many, not the only motivating force, nor in many instances is it the most important one determining effort. It is important to highlight these differences because different modes of analysis determine not only results, but what the analyst looks for and possibly finds. value of a depends in part on the research it generates. Another difference lies in the interest in inefficiency. It is not clear whether De Alessi cares about inefficiency. postulate of maximization of utility by all individuals and the explicit assertion by De Alessi that The equilibrium solution associated with a given set of constraints is efficient (p. 73) gives the impression that inefficiency is assumed away. Clearly, the X-efficiency approach sees inefficiency as a major problem. Only if one employs an approach that at least recognizes within it the possibility of inefficiency can one have empirical results which indicate an approximation to efficiency. Inefficiency must be a variable if we are to be able to find meaningful empirical instances which are efficient. Compared to the neoclassical approach, the rights approach seems to dissect the firm into component individuals and their implicit contracts. After dissection, the rights approach assumes that the ordinary utility-maximization rules still apply. *Professor of Economics and Population, Department of Economics, Harvard University, 204 Littauer Center, Cambridge, MA 02138. I am indebted to H. Barkai, J. Dean, R. Frantz, E. Helpman, J. Medoff, and J. Rothenberg for valuable comments. They are not responsible for errors or viewpoints. 'The labels property rights theory and general neoclassical theory will be used interchangeably. 2To stake his claim, De Alessi might have considered at least some other alternative approaches, such as the work of Oliver Williamson (1970) and that of Richard Nelson and Sidney Winters (1982).

This paper attempts to show that it is useful to view productivity as a prisoners' dilemma problem, that conventions are alternative solutions to the prisoners' dilemma, that effort conventions are usually nonoptimal, and that a shock is necessary in order to shift from one nonoptimal solution to another. Within this framework the invisible hand does not produce a Pareto optimal result. It is consistent with any number of nonoptimal effort conventions.