conflict. The two-stage threat game appropriately depicts the constellation of
CEECs as ‘members with a derogation’ coming up to full EMU-membership.
Instead of simply adhering to the Maastricht provisions, CEECs are better
off by utilizing the perverted incentive structure within the framework of
ERM II. We have discussed the assumptions of the brinkmanship game
as well as anticipated potential objections regarding the game structure.
The remainder of the game-theoretic analysis focuses on the Rubinstein
bargaining solution.
The particular share c
i
of costs of convergence, which CEEC
i
can load on
to the EU-15, is a bargaining outcome. In this context, the Rubinstein-
Bargaining Solution (RBS) creates a very appropriate framework of infinite
time horizons with alternating offers to our game Γ(ς
i
, d). This is due to
the fact that its non-cooperative characteristic brings the RBS close to our
understanding of the two-stage threat game. In regard to the brinkmanship
game, an advantage of the RBS is that the subsequent haggling about
the distribution of the costs of convergence gains more prominence. This
is taking into account that the duration of negotiating affects markets
calculus—i.e. the mechanism of punishing in this game. The latter is quite
realistic: The longer the negotiations take, the more it becomes obvious
from the perspective of markets that both parties are unable or unwilling
to agree on a CEEC
i
’s appropriate policy mix in ERM II—i.e. basically
fiscal policy as monetary policy is brought to a standstill in the context of
high capital mobility. Both players are well-aware that time is not on their
side. Contrary to the general association of the RBS as a ‘shrinking pie’
(i.e. utility) over time, here the ‘pie of costs’ inflates over time and will
leave its mark on both parties upon collapsing—i.e. both are incurred with
costs of default. For that reason, the RBS makes use of players’ ‘patience’,
so that despite an infinite horizon the impeding risk of a breakdown makes
both parties agree on the distribution of the costs of convergence in finite
time. Accordingly, the settlement of the bargaining is especially dependent
on each players’ negotiating skills in terms of ‘patience’.
13
The following figure 2 depicts the RBS for our bargaining on costs of
convergence. The players are well-informed, i.e. the assumptions of complete
information and common knowledge apply to the RBS. The negotiations
proceed as follows: At the outset of their brinkmanship—i.e. CEEC
i
pursue
a ‘threaten-thy-neighbour’-policy congruent with choosing ‘overvalue’—the
CEEC
i
makes a particular offer in t = 0. This offer consists of demanding
13
As is the case with poker, the player who can longer convincingly conceal his fear of
being faced with the costs of a negotiations breakdown is better off.
18
Convention