Notes, Psc 760, 9-11

1. Recall the three divisions: strategy, tactics, language to implement tactics

2. You're given a 'problem.'  You must decide whether or not it is wise to try to negotiate a solution.  You first apply some simple game theoretic analysis as a heuristic device to answer the question: just what is the mix of cooperative incentives with conflictual incentives?

    a. You will never encounter a zero-sum game.  If outcomes to the Parties are measured with respect to their utilities, there can exist no zero-sum games, because interpersonal comparisons of utility is meaningless.

    b. All negotiated solutions are "win-win," to some degree.  Recall that we insist that negotiation is a voluntary process among 2 or more autonomous and reasonably intelligent Parties.  But note also that "win" and "lose" need to be measured with respect to some reference point of condition.  Let's follow that tradition by calling the reference the "best alternative to negotiated  agreement" or BATNA.

    c.  Negotiating processes are said to be of two types: distributive or integrative.  The metaphor for distributive negotiation is 'dividing the pie' (a very American food!).  That is misleading: let's use dividing the pizza with two toppings.

    d. Negotiation is a process which anticipates a possibility of exchange.  Consider the simplest situation: two Parties (who are also Principals) anticipating the possible exchange of a bicycle for money.  Concepts: buyer's surplus, seller's surplus, negotiating set, reservation price (Shell calls this -- Ugh! -- "bottom line"), negotiating set. 

        i. We first do a one-dimensional schematization, using 'price' as the medium for comparison.

        ii. We next do a two-dimensional schematization using the parties' cardinal Utilities as dimensions.  (Aside1 on Utility: its measurement yields a ration scale.  Aside2 on utility: we may try to measure utility on either an ordinal or cardinal scale.  In the political arena, ordinal utility is practicable and all the we can aspire to.  It is sufficient for the negotiator's purposes.)

        iii. This can easily be generalized to 3+ dimensions, when more parties are added.  This leads to the application off N-person game theory, which is really the analysis of possible coalitions.

        iv. Finally, think about a 'policy  space,' in which points are (perhaps) combinations of negotiators' offers (or positions).  In constructing this we appeal to the von Neumann-Morgenstern utility function, which asserts that single number, as a measure of Utility (or 'all-inclusive goodness') can be assigned to any state of affairs however complex.  Who does such an assignment?  Some individual: perhaps the negotiator herself if she is the Principal; else the authority who gives the negotiating agent her instructions.

3.  Now we're ready for the Man-Woman Game of Anatol Rapoport, which brilliantly illustrates all of the important concepts of 2-person game theory.  (Handout)

__________________________________________________________________________________________________

Designing the negotiating exercise

Of what is this an instance?  (Recall the Big 3?)

Parties, Issues, occasion and location, logistics, ...

What should we try to achieve?  What constraints must we respect?

The negotiator's instructions: guidance and (political or bureaucratic) protection

Emphasis upon writing and editing negotiating proposals.  What do they look like? 

    a. Do we seek and agreement in Principle, in Detail, or of some other sort?

    b. Do we seek to reach agreement by working on issues in sequence, in parallel, by 'baskets', or some other way?

    c. What degree of trust exists among negotiators -- (among agent? among Principals?)