WikiSummary, the Social Science Summary Database

Milgrom and Roberts: Bargaining costs, influence activities, and the organization of economic activity

From WikiSummary, the Free Social Science Summary Database


Milgrom and Roberts. 1990. Bargaining costs, influence activities, and the organization of economic activity. In Perspectives on Political Economy, eds. James Alt and Kenneth Shepsle, 57-89. Cambridge: Cambridge University Press.

In Brief

The authors hope to add two elements to the theory of the firm as we currently know it:

  1. A focus on bargaining costs (as opposed to existing emphases on asset specificity, uncertainty, and iteration), and
  2. Influence costs: losses caused by individuals in the firm attempting to influence company decisions for their own private benefit, and the costs the firm incurs trying to prevent this. (See also Kiewiet and McCubbins on "agency loss."

Variables and Hypotheses

The variables, then, are as follows:

  • Y = Creation and size of an organization (i.e. firm)
  • X = Bargaining costs and influence costs (discretionary authority costs)

Basically, the independent variables are the costs that come from delegating authority to individual members of the enterprise. Bargaining costs includes monitoring/measurement costs and asset specificity. Influence costs are costs incurred by people trying to influence company decisions in ways that favor themselves at the company's expense (e.g. corruption), and costs the company incurs to prevent this. This mechanism (corruption) led to most inefficiencies in command economies.

Main hypothesis: Bargaining costs lead to firm creation; influence costs put a cap on firm expansion. Firms expand until bargaining costs no longer exceed influence costs.

Place in the Literature

Milgrom and Roberts engage in a long-running debate about why (and when) it makes economic sense to form a firm--and how large the firm should be. Their main departure from Coase, who started this debate, is to emphasize the influence costs that set an upper limit on firm size; Coase emphasized bureaucracy costs as creating the upper limit.


Role of Bargaining Costs

If there were no bargaining costs, there would be no firms. Bargaining costs are the reason you create a firm; influence costs determine the maximum size of the firm.

Bargaining costs include contracting costs. If you could always write perfect contracts, asset specificity wouldn't create hold up threats. Although other authors have emphasized the importance of asset specificity to the theory of the firm (see e.g. Klein), asset specificity matters only because it increases bargaining costs. And higher bargaining costs create an incentive to integrate.

Comment: But are "bargaining costs" any different from the pre-existing (and widely used) concept of "transactions costs" that made Coase famous? See page 74 (and on) for what goes into bargaining costs--it sounds much like transactions costs. Bargaining costs include the following:

  • Coordination failures: Think battle of the sexes: we spend time battling over (Pareto equivalent) outcomes. Once we make an agreement, we get a Pareto improvement. But if we spend more bargaining than we will gain by finally agreeing, then we are worse off than if we hadn't tried bargaining in the first place.
  • Measurement and information costs. This includes hidden type (adverse selection), hidden action (moral hazard), and also more general types of information costs.

Role of Influence Costs

Firms centralize authority, and this authority can be influenced. So greater centralization represents increasing chances for corrupt influence. Compare this with Coase, who talked about declining returns to scale due to rising bureaucracy costs. Influence costs are presented as a somewhat broader category of cost, that possibly includes bureaucracy costs; in addition to bureaucracy costs, everybody is trying to kiss the manager's rear end, which creates costs.

Take Home Message

"Our general proposition is that any centralization of authority, whether in the public or private sector, creates the potential for intervention and so gives rise to costly influence activities and to excessive intervention by the central authority. These costs need to be weighed against the benefits of centralization to determine the efficient extent and locus of authority."