solutions that are Pareto superior to the status quo. In the figure, these are locations in the northeast quadrant -- which includes only half of the policy space
above and to the right of the diagonal. If the number of independent veto positions is increased, Negative Coordination implies a rapidly shrinking policy space
and an increasing immobilism, as the cumulation of vetoes will rule out more
and more attempts to depart from the status quo.
It should perhaps also be pointed out that discussion in the following chapters is explicitly about interactions between interdependent actors, not about the internal interactions within a collective actor, which were discussed in Chapter 3. However, given the formal parallelism of between-actor and within-actor interactions in our multilevel analyses,
it can be assumed that much of what is being said here will, with appropriate adjustments,
also help to explain interactions within corporate or collective actors. In actual case studies, we will therefore employ these same analytical concepts at several levels. An association, for instance, is considered a collective actor whose internal interactions (when they
need to be empirically considered) may take place within an arena structure. In its external
interactions, this association may then be part of a coalition (a higher-level collective actor) that was formed within a preexisting network structure in order to win a majority
vote within the parliamentary arena. In this fashion, a relatively limited set of relatively
simple conceptual tools can be employed to reconstruct and to explain quite complex and
convoluted empirical interactions.
Unlike in neoclassical economics, rational-choice theories are of course much concerned with the problem of how the existence of institutions may be theoretically derived
from the rational choices of purely self-interested individuals. For empirical policy research, however, this is not really much of a problem -- we only need to recognize institutions where they happen to exist.
3. Thus in Figure 5.2 both Row and Column could use a randomizing device to determine their choices between Up and Down or Left and Right. By selecting the probability
distribution that maximizes their expected payoff, given the choice of the other party, they
will arrive at a Nash equilibrium in mixed strategies. In Figure 5.2 that would imply that
both of the players assign a probability of 50 percent to each of their pure strategies, which
would give each of them an expected payoff of 2.5. 4.
It is nevertheless rare that analyses in the media are developed in explicitly gametheoretic terms, as was done, in a very enlightening column in the Economist, for the options that Prime Minister John Major and opposition leader Tony Blair might consider
with regard to a referendum on the European Union ("Bagehot" 1994).
Game theorists have developed a series of concepts that will reduce the number of
equilibria that will have to be considered. Thus it is plausible that Pareto-inferior equilibria. will be ignored, and in sequential games it also seems reasonable to require that only
"subgame-perfect" equilibria should be considered ( Osborne/ Rubinstein 1994). In a much
more ambitious effort, John Harsanyi and Reinhard Selten ( 1988) have attempted to develop a general theory that, through a hierarchy of criteria, should lead to the selection of
a unique equilibrium in each game. The prescriptive plausibility of some of these criteria
is still in dispute among game theorists, and Harsanyi ( 1995) has in the meantime pre
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: 114.
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.