David A. Lax

Robust Estimators of Scale: Finite-Sample Performance in Long-Tailed Symmetric Distributions

Abstract This article presents the results of a Monte Carlo study of the robustness of scale estimates in the presence of long-tailed, symmetric distributions. The article examines the performance of several families of estimates in samples of size 20 from several distributions. The family of A-estimators, finite sample approximations to the asymptotic variance of M-estimators of location, appears to be more robust than the sample standard deviation, the median absolute deviation from the median (MAD), trimmed standard deviations, and M-estimators of scale. The most successful A estimator uses the biweight weighting function, which is also the basis for high-performance robust location estimates.

Negotiating Through an Agent

Agents often bargain on behalf of their principals. In many common negotiating situations, especially where ex post ratification of the agents agreement is required (e.g., union contracts, treaties), an agent faces inherent uncertainty about the terms that are minimally acceptable to the principal (the principals “reservation price”). In fact, the agents entire payoff function may be uncertain. We study bargaining behavior in these circumstances and show that the agents minimum demands unambiguously increase with increases in uncertainty about the principals reservation price, with increases in uncertainty about the payoff function, and with increases in the agents degree of risk aversion. We then fashion these results about an individual agents behavior into conclusions about the difficulty of reaching agreement in the overall negotiations. Using Axelrods measure of the “conflict of interest” in a game, optimal insistence prices in a one-shot bargaining situation, and two equilibrium concepts in a common commitment game, we show that the inherent uncertainty of agency bargaining can frequently make disagreement more likely.

Optimal Search in Negotiation Analysis

A negotiators reservation price or “bottom line” depends directly on the value of the no-agreement alternative to a proposed negotiated agreement. Often, ones no-agreement alternatives are uncertain and finding them requires a costly search, as in the case of a seller who must expend effort, time, and money in finding potential buyers. The value of the search should determine the sellers bottom-line or reservation price in dealings with a prospective buyer. Optimal search and stopping theory suggest useful procedures and heuristics for evaluating ones reservation price in negotiation and for searching among alternatives.

Three ethical issues in negotiation

ConclusionThe overall choice of how to negotiate, whether to emphasize moves that create value or claim it, has implications beyond single encounters. The dynamic that leads individual bargainers to poor agreements, impasses, and conflict spirals also has a larger social counterpart. Without choices that keep creative actions from being driven out, this larger social game tends toward an equilibrium in which everyone claims, engages constantly in behavior that distorts information, and worse.Most people are willing to sacrifice something to avoid such outcomes, and to improve the way people relate to each other in negotiation and beyond. The wider echos of ethical choices made in negotiation can be forces for positive change. Each person must decide if individual risks are worth general improvement, even if such improvement seems small, uncertain, and not likely to be visible. Yet a widespread choice to disregard ethics in negotiation would mark a long step down the road to a more cynical, Hobbesian world.