Search results

Why are firms sometimes more efficient than markets at organizing transactions? Why are most transactions arrayed neither at the pure "market" nor at the pure "hierarchy" end of the continuum, but rather in the "swollen middle," incorporating features of both "market" and "hierarchy"? Why don't firms make greater use of price incentives? This paper addresses these three questions by developing a model of the choice of institution. One key building block is the distinction between organizing methods (hierarchy and the price system) and institutions (firms and markets). Hierarchy and the price system are two distinct methods for organizing transactions, each with particular costs and benefits. Markets and firms are institutions which use one or both of these methods. Although markets predominantly use prices and firms rely principally on hierarchy, there is not a one-to-one correspondence between prices and markets or between hierarchy and firms. Indeed, the paper argues that it is generally more efficient to use a mix of both methods than to specialize in either. The paper focuses on the enforcement properties of prices and hierarchy. Hierarchy controls individuals directly by constraining their behavior (by imposing behavior constraints) while prices do it indirectly by measuring their outputs (through price constraints). Under hierarchy, individuals receive a salary to do as told, while self-employed individuals governed by the price system are rewarded on the basis of their output. Each system has its own biases: using prices maximizes effort (minimizes shirking) but incites individuals to inflate the price and/or reduce the quality of their output. (It encourages cheating.) Relying on hierarchy results in the opposite bias: under hierarchy individuals are not paid in function of their output, but instead are rewarded for following directives. They have, thus, strong incentives to minimize effort (to shirk) unless properly supervised, but, being paid a fixed sum to follow orders, they have few incentives to cheat. Hence the price system experiences low shirking, but potentially high cheating costs, while hierarchy faces low cheating but high shirking costs. Organizing costs are the sum of shirking and cheating costs. Any given transaction will be organized by the mix of price and hierarchy (i.e., by the mix of price and behavior constraints) that minimizes organizing costs. A transaction will be organized within a firm if the reduction in cheating costs achieved by replacing price constraints by behavior constraints exceeds the resulting increase in shirking costs and by the market in the opposite case. The paper shows that cheating and shirking costs increase more than proportionately as one concentrates in either pure price or behavior constraints. Hence using a mix of both methods generally minimizes the sum of cheating and shirking costs. This explains why most transactions exhibit features of both markets and hierarchy. The paper shows clearly the tradeoff involved between price and behavior constraints. It explains the costs and benefits of using two types of price incentives in firms, piecework and profit centers, and predicts when they will be used.