Title: Strategic bargaining theory
1- Strategic bargaining theory
- and negotiations on global public goods
Alejandro CAPARRÓS Spanish National Research
Council (CSIC) and University Carlos III
2 The issue
- International negotiations on
- Climate change
- Ozone layer hole
- Biodiversity conservation
- Other (global) environmental issues
- Main common feature
- Prisoners dilemma
- The tragedy of the commons (Hardin, 1968)
3Main approaches
- The small stable coalitions thesis
non-cooperative game theory and cartel literature
results (external and internal stability concept) - The grand coalition thesis cooperative game
theory (core concept) - The negotiations approach bargaining à la
Rubinstein
4Motivation for the negotiations approach I
- F.O. Hampson and M. Hart (1995). Multilateral
Negotiations Lessons from Arms Control, Trade
and the Environment. JHU Press. - multilateral negotiation is essentially a
bargaining process that occurs between different
coalitions - Coalitions are defined as the unification of
the power of resources (or both) of two or more
parties so that they stand a better chance to
obtain a desired outcome or of controlling others
not included in the coalitions. Most
negotiations when N is large have a tendency to
break down into negotiations between different
bargaining sets or coalitions
5Motivation II
- Characteristics of these international
negotiations - Heterogeneous countries are negotiating
- These countries are regrouped in heterogeneous
coalitions - G77 and China
- This coalition has been relatively stable in
climate change negotiations, although it has
split up in several key moments - On other issues, as trade negotiations, it
recently broke down G20, LDC. - Umbrella Group
- This coalition broke-down during climate change
negotiations - European Union
6Motivation III
- Countries forming a coalition have internal
meetings, which are closed when important issues
are treated. - They set up a common strategy and individual
negotiators representing the coalition have
bilateral or multilateral meetings. - Northern countries are in general (in any case in
climate change), responsible for the problem and
ready to move first (climate change), give
development aid (at least to some extent), accept
reductions in their current privileges (trade
negotiations) - On climate change negotiations, northern
countries (except USA) have accepted to move
first, accepting quantitative restrictions to
emissions (Kyoto Protocol)
7Motivation IV
- Negotiations take time, several negotiation
rounds - United Nations Framework Convention on Climate
Change, signed in 1992 (ratified by all
countries) - Kyoto Protocol, signed in 1997, ratified in
2001-2004 - Marrakech Accords, signed in 2001, ratified with
the Kyoto Protocol - Post-Kyoto climate change agreement, currently
negotiated - Information is not complete
- Close meetings of the coalitions
8Motivation V
- The Presidents of the different COPs have had
different strategies, e.g., favoring bilateral
meetings or favoring multilateral meetings. - Negotiations sometimes fail or have success
depending on crunch issues e.g. carbon sinks
in The Hague Bonn. - Negotiations frequently deal with different
subjects, which are not necessarily related, e.g.
climate change negotiations and WTO negotiations.
9Strategic Bargaining Theory I
- Some papers on international negotiations (on
global public goods) and strategic bargaining
theory - Caparrós, Péreau and Tazdaït, 2004. North-South
Climate Change Negotiations a Sequential Game
with Asymmetric Information. Public choice, 121
(3-4) 455-480. - Negotiations between coalitions
- Two-three coalitions negotiating
- ? We are now extending the framework to n
coalitions. - Infinite time horizon
- Asymmetric information
10Strategic Bargaining Theory II
- Caparrós, Péreau and Tazdaït, 2005. Bargaining
with Non-monolithic Players. FEEM Working paper
95.2005. - Coalition haves the possibility to split-up
strategically during the negotiations. - Camiña and Porteiro, 2004. The Role of Mediation
in Peacemaking and Peacekeeping Negotiations.
CORE Discussion Paper 2004/5. - Mediation, not arbitration, is the role of a
President of a COP.
11Strategic Bargaining Theory III
- Manzini and Mariotti, 2005. Alliances and
negotiations. Journal of Economic Theory. - Coalitions are negotiating and have internal
voting rules unanimity or majority - Issue linkage
- Bac and Raff (1996)
- Fehrstman (1990, 2000)
- Bush and Horstmann (1997, 2002)
12- North-South Climate Change Negotiations
- a Sequential Game with Asymmetric Information
Caparrós, Péreau and Tazdaït Public choice
(2004), 121 (3-4) 455-480
13The Model General Characteristics I
- Heterogeneous coalitions are negotiating
- Two in the bilateral version (North vs South),
three in the multilateral version (South splits
up) - Countries inside the coalition are identical
- Simultaneous alternative offers
- The Northern countries coalition (The North)
starts proposing an amount of transfers to the
Southern countries coalition (The South) - Afterwards, alternative sequential offers follow
(if no agreement is reached) - Infinite time horizon (or till an agreement is
reached)
14The Model II
- The objective function of the coalitions is
general, depending on the transfers proposed by
the North to the South to reduce emissions - Transfers (the sole strategic variable of the
model) affect - negatively the welfare of the North
- positively the welfare of the South
- The welfare of the coalition is the sum of the
welfare of the countries forming it - The minimum amount of transfers asked by the
South is decided in an internal vote
15The Model III
- That is
- Where ? Utility function of the North, U
Utility function of the South, B abatement
benefit, C number of countries of the Southern
coalition, d discount rat ,t transfer, b
reservation transfer
16The Model IV
- Asymmetric information
- the North does not know the true type of the
South, there are two possibilities (for a given
level of effort) - Strong requirements (b)
- Weak requirements (b-)
- the South does know the type of the North
- In the multilateral framework we analyze the
outcome with - Complementary and,
- Substitutable Southern coalitions.
17The Model V
- Sequential Equilibrium (Kreps and Wilson, 1982)
- The strategies of both players are optimal in all
the points of the game, given the believes of the
North - For any information set reached the believes of
the North are given by Bayes law - The beliefs must be the limit of, at least, a
sequence of reasonable beliefs (we assume that
the North has pessimistic believes)
18Complete Information (Multilateral)
- In general
- the South is interested in negotiating together
(only one coalition), while - the North prefers to negotiate with two different
coalitions - Taking into account the nature of substitutable
or complementary of the Southern coalitions - if they are complementary they are interested in
acting separately - if they are substitutable, they are interested in
acting together
19Incomplete Information (Bilateral)
- We distinguish Southern coalitions with
- weak (b?) and
- with strong demands (b)
- Using a series of Lemmas, we show that a Southern
coalition b? has interest in acting as if it were
a b - A Southern coalition b? will always suggest a
counter-offer identical to the one proposed by b
- However, for some levels of transfers, b? will
accept in the first negotiation round
20Incomplete Information (Bilateral)
- If the offer proposed in the first period
satisfies - and the a priori beliefs of the North lower than
a level g (pltg) - Separating equilibrium
- b? will accept
- b will reject
- if pgtg, both coalitions obtain the level
corresponding to b - Pooling equilibrium
- That is The solution of the game is strongly
influenced by the a priori believes of the North.
21Incomplete Information (Multilateral)
- We only analyze situations where a delay appears.
- This delay can have two different
interpretations - For the North to get information about the
characteristics of the South - For the South to dissociate and to show its real
requirements
22Incomplete Information (Multilateral)
- Fist case two Southern coalitions with different
requirements (bj? and bk) - sub-case 1 acting together they ask b?. They are
interested in separating themselves if bj? lt ?1
- sub-case 2 acting together they ask b. They are
interested in separating themselves if bj? lt ?2
- since ?1 lt ?2, for bj? lt ?1 whatever the claim of
the union, the Southern groups favor disunity,
but - for sub-case 1, bk is confronted with a delay
while the union would lead to an agreement from
the first stage, - for sub-case 2, the negotiation delay is less
costly for bk than in the previous case, since
the delay would also appear acting together.
23Incomplete Information (Multilateral)
- Second case two Southern coalitions with the
same requirements (both high bj, bk) - the negotiation will end in the second period
with or without union - it shows that each group is interested in acting
separately when their respective demands,
although high, are not enough to take a relevant
part of the benefit (bk lt ?3) - Third case two Southern coalitions with the same
requirements (both low bj?, bk?) - the delay can take place when the groups act
jointly and require b (the delay can be
interpreted as the means to affirm high demands) - The paper discusses also other configurations.
24Incomplete Information (Multilateral)
- When the Southern groups form a union (when they
are substitutable) the delay favors the North. - That delay corresponds to the time necessary for
it to acquire the relevant information concerning
the demands of the groups. - When the Southern groups act separately
(complementary coalitions), the delay has several
interpretations - E.g. if one of the two Southern groups is of type
b, while the other is of type b?, the delay
corresponds to the time necessary for b to
dissociate from b?.
25Incomplete Information (Multilateral)
- In general, in the multilateral framework, the
information asymmetry reinforces - the negotiation power of the complementary
coalitions, which will generally act separately - reduces that of the substitutable coalitions,
which will generally act together. - When both groups are of type bi, the conditions
of complementarity-substitutability with
asymmetric information are identical to those
obtained with complete information. - On the contrary, when at least one of the groups
is of type bi? , the conditions of
complementarity (substitutability) are wider
(more restrictive).
26Conclusions I
- Objective determine the different forms that
international negotiations to enlarge the Kyoto
Protocol to include the developing countries
might take. - The framework developed is wide enough to
represent most of the international negotiations
between the North and the South, as long as - the North is willing to transfer money (or other
economic advantages) in exchange of some
commitment by the Southern countries - data about the North are more accurate than data
about the South.
27Conclusions II
- In a bilateral framework, the asymmetry of
information is not always beneficial for the
South - the North may delay the moment of an agreement to
obtain information about the characteristics of
the Southern coalition, - compared to complete information, the South will
pay a negotiation cost due to the delay. - In a multilateral framework, the information
asymmetry reinforces - the negotiation power of the complementary
coalitions, which will generally act separately, - reduces that of the substitutable coalitions,
which will act together.
28Applied Conclusions
- Developing countries are interested in becoming
complementary by setting at an early stage
(pre-game) of the negotiations a high minimum
level of ratification for the agreement to come
into force (as for the Kyoto Protocol). - If the southern coalitions become complementary
- they will act separately,
- the agreement will not be reached in the first
negotiation period, - the delay can be interpreted as the time needed
by the coalition with strong demands to
dissociate from the coalition with weak demands.
29- Generalization to C southern countries/coalitions
Work in progress
30Generalization The Model I
- The North (N) negotiates with C southern
countries, which can form coalitions. - All C southern countries are identical. We denote
ci the number of countries in coalition i - t is the transfer proposed by the North (N) and
ei is the effort provided by coalition i - To simplify, we
- normalize the payoffs setting the minimum
requirements (bi) equal to zero - assume that the North always starts the
negotiations - assume perfect information.
31Generalization The Model II
- Two stages
- Stage 1 the C countries form n coalitions.
- Provisional criterion maximization of aggregated
welfare of the Southern countries - Simultaneous choice of coalition structure
- Stage 2 The n coalitions negotiate with the
North - Simultaneous negotiation
- Sequential negotiation
- In both cases, time between offers tends to zero
(e.g. we are in negotiations within a COP)
32Generalization The Model III
- The utility function of the Southern coalitions
can be written as - And that of the North
33Generalization The Model IV
- We consider that the Southern coalitions can be
- Perfectly complementary, when the North needs to
reach an agreement with all of them. - Perfectly substitutable, when an agreement with a
subset of the coalition structure suffices to
have benefits for the North. - Partially substitutable, when there exist a
minimum number of signatories (S) needed to have
positive benefits for the North.
34Generalization Preliminary Results
- With simultaneous as well as with sequential
negotiations the three following statements hold - When the efforts of the countries are perfectly
complementary, Southern countries will act
separately. - When the efforts of the countries are perfectly
substitutable, Southern countries will act
together, regrouped in a single coalition. - If there is a minimum number of participants
1ltSltC, the countries will regroup in the maximum
number of coalitions that ensures that all
coalitions are needed to reach S.
35Generalization Example I
- Total number of Southern countries needed (or
percentage of emissions) C 100. - S 55
- S 67
- S 30
36Thank you for your attentionE-mail
acaparros_at_ieg.csic.es