aggregated preferences or interests of their members in negotiations with other
such units. The implication is that such negotiations must be conducted by
"clubs" or "associations;", as defined in Chapter 3, and perhaps by associations of
associations (or "peak-level" organizations). When that is the case, the theoretical
focus shifts back from the interaction among large organizations to the internal
interactions within associations in which the "logic of influence" in external negotiations is constrained and must be legitimated through the internal "logic of
membership" ( Streeck/ Schmitter 1981). This will be the subject of the next two
The character of the implementation game may, and usually will, differ from the actor constellation that characterizes the substance of the agreement. Hence agreement may
have to deal with a Battle of the Sexes constellation, but implementation could still be an
A three-person game would be represented by a three-dimensional cube figure; for
larger numbers of players, neither form of representation would be useful. However, Appendix 2 contains a tabular representation that is relatively transparent even for larger
A pure zero-sum conflict would be located on the diagonal representing the welfare
Though it is often permissible to identify the NA point with the status quo, this is not
correct: if, for instance, one party could avail itself of an outside option (say, another contract offer) that would also represent an improvement over the status quo, or if one party
can credibly threaten to reduce the other's payoff below the previous status quo if it
should fail to agree to "an offer that you cannot afford to reject.'
The difference between zero-sum and Battle of the Sexes constellations is illustrated
by the history of competition and cooperation among firms in the Japanese auto industry:
Within the closed Japanese market, there was fierce competition for market shares. It was
only after they began to enter foreign markets in the 1970s, where market shares could be
captured from foreign competitors, that cooperation among Japanese firms gained in significance (personal communication from Kjell Hausken and Thomas Pluemper).
The reader should be warned that my use of this literature is nonstandard. In the
standard version the parties will negotiate on a given possibility frontier, and in order to be
acceptable the outcome will only have to be "fair" but not welfare maximizing. Starting
from the Coase Theorem, I assume instead that parties should be negotiating on the welfare frontier (i.e., the highest available welfare isoquant) and that on that welfare level the
distribution should be fair. If side payments or package deals are possible, as must be assumed for the Coase Theorem to hold, there is in fact no reason that rational parties
should negotiate on a given possibility frontier rather than on the welfare frontier.
For individuals that zero point could be defined by the income provided by public
That distribution will not always favor the bigger or more powerful party. When the
constellation is not a Battle of the Sexes (as was assumed in the EU-Hungary example) but
Questia, a part of Gale, Cengage Learning. www.questia.com
Book title: Games Real Actors Play:Actor-Centered Institutionalism in Policy Research.
Contributors: Fritz W. Scharpf - Author.
Publisher: Westview Press.
Place of publication: Boulder, CO.
Publication year: 1997.
Page number: 147.
This material is protected by copyright and, with the exception of fair use, may
not be further copied, distributed or transmitted in any form or by any means.