Publications

 

CTN Newsletter n.13 - March 2010

Index

 

Editorial

by Frédéric Deroïan, Groupement de Recherche en Economie Quantitative d'Aix-Marseille (GREQAM) - Université de Marseille

Dear All,

This year, the CTN annual workshop will be held in Marseille, on June 17-18. This unusual date has been chosen by the committee in order to open the workshop to new colleagues, potentially interested in the themes of the workshop, and notably our trans-Atlantic colleagues. We will have three guest speakers: Adam Szeidl (Berkeley University), Kaïvan Munshi (Brown University), and Parag Pathak (MIT). For more information about the workshop, please visit the website http://www.vcharite.univ-mrs.fr/~ctn2010/index.

We would like to remind you that this workshop will be followed by the Journées Louis-André Gérard-Varet, a two days conference in public economics, to be held on June 21-22 in Marseille (http://www.idep-fr.org/spip.php?article243). The guest speakers will be Dennis Epple (Carnegie Mellon University), Roger H. Gordon (University of California) and Sir Partha Dasgupta (Cambridge University). Attending both conferences, and eventually submitting papers about networks, coalitions and matching to the LAGV conference, is a good way to confront our works with other economists.

Last, we would like to remind you that the PET conference will be held on June 25-26-27 (http://www.accessecon.com/pubs/PET10/). Keynote speakers will be Thomas Palfrey (California Institute of Technology), Pierre Pestieau (University of Liege) and Philip J. Reny (University of Chicago). PET10 welcomes papers in all aspects of public economics and related areas.

We cordially invite you to visit the CTN webpage for updates. We are looking forward to meet you in these events,

Frédéric Deroïan

 

CTN Announcements

15th Coalition Theory Network Workshop - Call for papers - Submission deadline postponed to March, 20
Marseilles, France, June 17-18, 2010
http://www.feem-web.it/ctn/events/10_Marseilles/ctn15i.htm

The Groupement de Recherche en Economie Quantitative d’Aix-Marseille (GREQAM) is organizing the 15th Coalition Theory Network Workshop in Marseille, France, on June 17-18, 2010. We cordially invite you to attend.

The Workshop will present the state of the art of theoretical and empirical research in:

• Coalition formation
• Network formation
• Matching.

During the Workshop there will be three invited sessions. The keynote speakers will be:

Kaivan Munshi (Brown University, USA)
Parag Pathak (Massachusetts Institute of Technology, USA)
Adam Szeidl (Berkeley University, USA)

The Workshop solicits papers in economics, in the field of coalition theory, networks and matching. Completed articles are preferred but extended abstracts are also welcome. Papers can only be submitted at ctn2010@ehess.univmed.fr . Submissions of contributed papers will be due by March 20th, 2010. The submitters will be notified about the status of their submissions by April 1st, 2010.
Papers presented in the Workshop are eligible for publication in the FEEM CTN Working Paper series.

The scientific coordinators of the event will be Sebastian Bervoets, Francis Bloch and Frédéric Deroïan. The scientific committee will be composed by representatives of the Coalition Theory Network: Centre d'Économie de la Sorbonne (CES) - Université Paris 1, Center for the study of the Organisations and Decisions in Economics (CODE) - Universitat Autònoma de Barcelona, Center for Operation Research and Econometrics (CORE) - Université Catholique de Louvain, Fondazione Eni Enrico Mattei (FEEM), Groupement de Recherche en Economie Quantitative d'Aix-Marseille (GREQAM) - Université de Marseille, Department of Economics - Universiteit Maastricht, Department of Economics - University of Warwick, Vanderbilt University.

More information about the Workshop at www.vcharite.univ-mrs.fr/~ctn2010.
For further inquiries contact the organizers at ctn2010@ehess.univmed.fr

Note that the Journées Louis-André Gerard Varet will be held in the same week in Marseilles.


2nd Valencia-Louvain Workshop on Game Theory and Economic Behavior
Louvain-la-Neuve, Belgium, May 28, 2010
http://www.uclouvain.be/en-309135.html

CORE (Université catholique de Louvain) and CEREC (Facultés universitaires Saint-Louis), in cooperation with ERI-CES (Universitat Valencia), will be organizing the 2d Valencia-Louvain workshop on Game Theory and Economic Behavior in Louvain-la-Neuve (Belgium) on May 28, 2010. Coordinators: Vincent Vannetelbosch (CORE) and Jose Sempere-Monerris (ERI-CES). The programme will be available soon.


3rd Maastricht Behavioral and Experimental Economics Symposium: Theory and Experiments
Maastricht, The Netherlands, June 11, 2010
http://www.fdewb.unimaas.nl/meteor-seminar-et/M-BEES-2010/

The Economics Department (AE1) of Maastricht University hosts the 3rd Maastricht Behavioral and Experimental Economics Symposium (M-BEES) on June 11. The broad topic of the Symposium is Theory and Experiments and centers around the question if and how economic experiments can inform economic theory and vice versa. The Symposium takes place on a single day and closes with a social dinner. Invited keynote speakers:

B. Douglas Bernheim ((Stanford University)
Yan Chen (University of Michigan)


PET 2010 - Call for papers
Istanbul, Turkey, June 25-27, 2010
http://www.apet.org/

The Association of Public Economic Theorists (APET) is very pleased to announce that it will hold its eleventh annual conference (PET10) at Bogaziçi University, Istanbul, Turkey, between June 25-27, 2010, beginning with a welcoming reception on the evening of June 24th.

Keynote Speakers:
Thomas Palfrey, California Institute of Technology
Pierre Pestieau, University of Liege

As with previous PET conferences, PET10 welcomes papers in all aspects of public economics and related areas.

APET is pleased to announce that PET 10 will host the concluding conference of the ESF Network, “Public Goods, Public Projects and Externalities.” Also, several sessions will be organized in collaboration with the Coalition Theory Network.

Deadline for submissions: March 15, 2010.



Article: A new approach for study coalition issues in climate change

Zili Yang, State University of New York at Binghamton

International cooperation in dealing with climate change is one of the most significant testing grounds and applications for coalition theory. Potential climate change also generates accelerated expansion of literature in applied coalition theory and game theoretical applications in environmental modeling. In November 2008, the MIT Press published my monograph, titled "Strategic Bargaining and Cooperation in GHG Mitigations - an Integrated Assessment Modeling Approach." In this book, cooperative and non-cooperative game theoretic solution concepts are introduced in the RICE model (Nordhaus and Yang, 1996) and coalition issues are examined in the RICE model through numerical simulations.

Overview of the methodologies
Most coalition studies are set up in highly stylized models. Numerical methods, if used, are likely for illustration purposes. In the monograph, cooperative, non-cooperative games and coalitional analysis are directly constructed in generic model of stock externality provision and then applied numerically to the RICE model, which is a stock externality provision model with details of regional economic growths, greenhouse gas (GHG) emissions, mitigation costs, and climate damages.

General problem of efficient provision of stock externality can be expressed as an optimal control problem of a social planner's problem. The social planner maximizes the weighted sum of individual agents' intertemporal utility functions. External effects are fully internalized in this optimal control problem. The most significant inefficient outcome of stock externality provision problem is the open-loop Nash equilibrium, which can be expressed as the solution of a corresponding system of differential equations.

The social welfare weights in the social planner's problem, {φi}, normalized onto simplex S = φi │∑ φi, play crucial role in the analytical approach here. Social welfare weights are parameters in the optimal control problem. Any point on S, maps to a solution of the optimal control problem, thus an efficient outcome of stock externality provision problem. From well-established general equilibrium theory, we know that S maps onto the entire efficiency frontier of stock externality provision problem. Under each set of {φi}, the solution of the optimal control problem is represented by a unique set of profiles of control and state variables. Now, the interesting research problem is follows: which subset of S (namely, conditional on guarantee of efficiency) is of interests from game theoretical and coalitional viewpoint?

Intuitively, {φi} on S is a "zero-sum" choice for n agents. If one agent's social welfare weight increases, at least one or more other agents' weights have to decrease. A larger φi, ceteris paribus, implies less mitigation efforts exerted by agent i and more benefits from other n-1 agents' efforts in providing stock externality, and vice versa. In other words, higher φi ensures higher payoffs for agent i in an efficient scheme of stock externality provision, and vice versa. The cooperative game of stock externality provision is built on the above intuition.

The outline of the cooperative game is as follows. In an economy with stock externality, n agents bargain and negotiate {φi} on S, having complete information on the social planner's optimal control problem as well as the outcome of non-cooperative Nash equilibrium. Once and if such {φi} is agreed upon, all agents adopt respective control variables as determined by the solution of the optimal control problem under this specific set of {φi}. Because stock externality is fully internalized in the solution, the outcome of this cooperative game achieves the efficiency. If one or more agents free ride, other agents penalize such defection by playing the non-cooperative Nash game thereafter.

Evidently, not all efficient solutions of the social planner's problem can be an outcome of the cooperative game defined above. Only if the solution is in the core (in the context of general equilibrium theory) of efficiency frontier, it is necessarily a possible outcome of the cooperative game. In this approach, the issue of finding a cooperative game strategy profile is transformed into identifying the subset of S that corresponds to the core allocations.

By definition, an efficient allocation is in the core if no sub-coalition can block the grand coalition for this allocation. Particularly, all agents in this grand coalition are better off than their respective non-cooperative Nash position (the individual rationality (IR) requirements). In a well-defined system of stock externality provision, the core is not empty and the Lindahl equilibrium (the counter part of the Walrasian equilibrium in public good economy) is inside the core, according to the general equilibrium theory (see Foley, 1970 and Starrett, 1976).

The tasks here are to identify the subset of S that corresponds to the core allocations and other refined game theoretical solution concepts in the optimal control problem of stock externality provisions. For these tasks, we need to define additional concepts, develop operational algorithms, and implement them numerically in the RICE model. The tasks are accomplished in the monograph.

In integrated assessment modeling of climate change, the above cooperative game is equivalent to international cooperation on the burden sharing of initial GHG mitigation quota. Such international cooperation is all-inclusive grand coalition. Regions impacted by climate change should all take actions or commit to certain GHG mitigation burdens.

Major results

The cooperative game solution concepts are sought numerically in the RICE model. Series results and conclusions are obtained. The follow are some highlights:

" The social welfare weights that correspond to the core allocations are identified;
" The refined cooperative game solution concepts, namely, the Lindahl equilibrium and the Shapley value, are identified;
" The Lindahl equilibrium solution has the strongest incentive properties for all regions. The grand coalition under the Lindahl social welfare weights is near "self-enforcing;"
" Some popular efficient solution concepts, such as the utilitarian and the Negishi solutions, are not in the core. Therefore, they cannot reflect a coalitional or cooperative game outcome;
" Cooperative game solutions are on the more "conservative" area in the efficiency frontier in terms of GHG mitigation intensity;
" Achieving the core allocations and the Lindahl equilibrium does not require endowment transfers. The conclusion is consistent with the first fundamental welfare theorem for the economy with public good;
" The core allocations are "renegotiation-proof" with many types of unexpected shocks to the system in closed-loop strategy space, but not so stable in other unexpected shocks.

The monograph contains detailed analysis on the issues outlined in this article. The approach pioneered in this research can be extended in several directions. The author continues his research on these issues.


Reference
Foley, D., 1970: "Lindahl's solution and the core of an economy with public goods,"
Econometrica, Vol. 38, pp. 66-72.
Nordhaus, W. and Z. Yang, 1996: A regional dynamic general equilibrium model of alternative climate change strategies. American Economic Review, Vol. 86: pp.741-765.
Starrett, D., 1973: "Externalities and the core," Econometrica, Vol. 41, No. 1, pp. 179-183
Yang, Z., 2008: Strategic bargaining and cooperation in greenhouse gas mitigations - An integrated assessment modeling approach. The MIT Press, Cambridge, MA.

 

Report: International Workshop on Fairness and the Commons:
Socio-Economic Strategies and Resource Dynamics

Alessandro Tavoni, FEEM and Ca' Foscari University of Venice

The International Workshop on Fairness and the Commons: Socio-economic Strategies and Resource Dynamics, organised by the International Center for Climate Governance (ICCG) and by the Centro Euro Mediterraneo per i Cambiamenti Climatici (CMCC) in cooperation with Princeton University, was held in Venice last 19-20 October.

The main themes that have emerged in the workshop, with respect to the integration between behavioral and environmental economics, can be summarized as follows:
1. The role of the discount factor in shaping individual and collective decision: how do we trade off the interests of the current generations with those of future generations?
2. The issue of scale and context in the achievement of cooperation among agents interacting in complex adaptive systems (CAS) with dynamically changing landscapes.
3. The importance of the often neglected diversity of behavior: can we advance in our modeling efforts by resorting to a combination of bounded rationality and optimization?

For what concerns the first point, much discussion has been directed to the problem of aggregation of individual discount factors and whether the resulting measure can be thought of as an instrument in the hands of policy makers. Empirical, neuroscientific and theoretical evidence seems to suggest that we discount the future differently at different points in time, and that the integration of heterogeneous opinions leads to hyperbolic discounting at the society level. However, it is not trivial to assess what happens at the individual level; it may be the case that a hyperbolic discount rate results from the aggregation of exponential ones at the individual level, or that the hyperbolic discounting also takes place at the disaggregate level. Another open question concerns how to incorporate the discount factor in a dynamic view of cooperation in time and space, something overlooked by most related literature. One suggested possibility is to use simulations to look at how different rates result in population-wide (average) behavior.

Regarding the evolution of cooperation in the commons, possible mechanisms that have been discussed to explain it range from genetic relatedness to reciprocal altruism, viscosity leading to clusters of cooperation and social norms arising within groups. Important factors ultimately determining the successfulness of the above are played by modularity, scale (local vs. global), as well as topology and network structure. In a further level of investigation, one may ask how networks arise. Moreover, one emergent conclusion from the discussion concerns the importance of variation and its maintenance in order to guarantee the robustness of CAS and avoid their collapse. Another commonality in the body of work presented in the workshop concerns the need of incorporating norms as endogenous to the models. Such norms, and the agents who are confronted with them, are embedded in specific contexts: the way they are perceived may ultimately affect their effectiveness in promoting cooperation, as magnetic resonance (fmri) experiments on human and animal brain suggests. This is why thinking about norms means more than averaging individual behavior: they have a life of their own which is deeply interconnected with collective behavior. In order to tackle such complexity, the participants propose to allow the modeling efforts to be informed and feedback into experimental and empirical work.

Such considerations pave the way for the third pointed highlighted above, that of bounded rationality; moreover, even the considerations on the discount factor are pertinent, as the behavioral foundation of hyperbolic discounting lies in the preference for immediate rewards (and aversion to delayed gratification). More generally, behavioral evidence suggests that in many domains humans follow heuristics rather than optimization in solving their "games". This since in reality individuals are confronted with metagames with a high degree of uncertainty: that is one doesn.t know what game he or she will play, and therefore resorts to heuristics that work well on average. One such example is that of imitation of successful behavior, which has been much employed in the models presented here in the form of replicator dynamics. Heuristics can arise as conventions to solve coordination problems, like those inherent in the management of the commons. Their emergence is strictly linked to the social pressure needed to enforce them: the convention of driving on one side of the street is maintained on the grounds of the sanctioning of violators in the response to the harm caused by the deviant action. The question arises about the cause of sudden changes in norms: one key actor is the external regulator, who has the power to influence individual behavior, for example in the management of the commons. One way to incorporate this idea in models is to allow for a deliberative process of reasoning to lay down the rule (for example by recurring to optimization techniques assumed to be in the hands of the environmental regulator), and then to have the enforcement stage to assess whether the rule becomes a norm which successfully fixates in a boundedly rational population (characterized for example by replicator behavior. In today.s large communities it may be the case that the formation of norms takes this top-down route in place of the canonical bottom-up pattern characteristic of smaller communities. Two dimensions have been recognized an important role in the above discussion: time and space. For what regards the former, it should not be forgotten that the life-time of individuals and institutions is different, therefore it may not only
be a matter of different levels of rationality, but of horizon, uncertainty and ultimately discount factor. Regarding the spatial aspects, the network structure impacts the enforcement of a norm: for example those deviating from traffic rules may have similar characteristics, and their relative abundance and topology may determine whether they are subject to social pressure or not. Lastly, it should be noted that in the absence of a legitimate institution solving the coordination problem in the commons, a decentralized process may lead to the emergence of a sub-optimal norm. Such process, however, may not be an option if the scale of the coordination problem is too large: the climate change problem (which due to its nature of threshold public good game can be seen as a coordination game), for example, requires a regulator since actors are institutions and enforcement is expensive.

Abstracts of presentations

Intergenerational equity and discounting
Simon Levin
A problem common to biology and economics is the transfer of resources from parents to children. We consider the issue under the assumption that the number of offspring is unknown and can be represented as a random variable. There are 3 basic assumptions. The first assumption is that a given body of resources can be divided into consumption (yielding satisfaction) and transfer to children. The second assumption is that the parents. welfare includes a concern for the welfare of their children; this is recursive in the sense that the children.s welfares include concern for their children and so forth. However, the welfare of a child from a given consumption is counted somewhat differently (generally less) than that of the parent (the welfare of a child is ..discounted..). The third assumption is that resources transferred may grow (or decline). In economic language, investment, including that in education or nutrition, is productive. Under suitable restrictions, precise formulas for the resulting allocation of resources are found, demonstrating that, depending on the shape of the utility curve, uncertainty regarding the number of offspring may or may not favor increased consumption. The results imply that wealth (stock of resources) will ultimately have a log-normal distribution.

The survival of the conformist: equity-driven ostracism & renewable resource management
Alessandro Tavoni, Maja Schluter
This paper examines the role of pro-social behavior as a mechanism for the establishment and maintenance of cooperation in resource use under variable social and environmental conditions. By coupling resource stock dynamics with social dynamics concerning compliance to a social norm prescribing non-excessive resource extraction in a common pool resource (CPR), we show that when reputational considerations matter and a sufficient level of social stigma affects the violators of a norm, sustainable outcomes are achieved. We find large parameter regions where norm-observing and norm-violating types coexist, and analyze to what extent such coexistence depends on the environment.


Evolution and the Framing of Norms of Behavior
Brian Skyrms
Sender-Receiver games are simple, tractable models of information transmission. They provide a basic setting for the study the evolution of meaning. It is possible to investigate not only the equilibrium structure of these games, but also the dynamics of evolution and learning . with sometimes surprising results. Generalizations of the usual binary game to interactions with multiple senders, multiple receivers, or both, provide the elements of signaling networks. These can be seen as the loci of information processing, of group decisions, and of teamwork.
A specific kind of interaction of the kind that game theorists put under the microscope . for instance, ultimatum bargaining . may be a member of various classes of social interactions. Each of these classes may carry its own norm, quite appropriate to the class, and the norms may
conflict. We suggest that framing of a decision problem should be interpreted as a signal about the relevant class of social interactions, and that evolutionary analysis should be redirected to systems of classes of social interactions. In this way, many ¡°anomalous¡± findings of experimental game theory may lose the air of mystery.


Inequality and Rule Performance in the Governance of Water Resources
Carmen Marchiori
This paper focuses on collective action problems in the governance of water resources.
Within this context, it investigates the interaction between inequality in landholding and water allocation rules through the joint incentives to cooperate that those generate. The proposed model involves two types of farmers who differ in terms of their initial endowment of land and can voluntarily contribute to the construction/maintenance of an irrigation network. The total amount of irrigation water is distributed according to some allocation rule and used by each farmer as an input of production in combination with land. The analysis identifies two key forces: an .efficiency based. and an .incentive based. force, which affect the distribution of water in opposite directions. The tradeoff between those two forces determines the nature of the relationship between inequality and optimal allocation rules. This tradeoff, in turn, critically depends on the degree of complementarity between agents' efforts. The predictions of the model offer a theoretical explanation for the apparently contradicting results of the theoretical and empirical literature in this field.


Coordinating towards a common good
Jorge M. Pacheco
In the animal world, collective action to shelter, protect, and nourish requires the cooperation of group members. Among humans, many situations require the cooperation of more than two individuals simultaneously. Most of the relevant literature has focused on an extreme case, the N person prisoner.s dilemma. Here we introduce a model in which a threshold less than the total group is required to produce benefits, with increasing participation leading to increasing productivity. This model constitutes a generalization of the 2-person stag-hunt game to an N-person game. Both finite and infinite population models are studied. In infinite populations this leads to a rich dynamics which admits multiple equilibria. Scenarios of defector dominance, pure coordination or coexistence may arise simultaneously. On the other hand, whenever one takes into account that populations are finite and when their size is of the same order of magnitude as the group size, the evolutionary dynamics is profoundly affected: it may ultimately invert the direction of natural selection, compared to the infinite population limit.

The full program of the workshop as well as the list of papers and presentations can be found at: http://www.iccgov.org/events-1_CONF_2009-03.htm.



Report: Workshop on "Political Economy and the Environment

Stéphane Zuber, CORE, Université catholique de Louvain

The workshop "Political Economy and the Environment" was held on the 23 and 24 October 2009 in Louvain-la-Neuve (Belgium). The workshop was co-organized by CORE, Chair Lhoist Berghmans in Economics and Management (Université catholique de Louvain) and EQUIPPE (Université Lille 1). The event was attended by an active audience, which had the opportunity to learn about some of the latest advances in the field of political economy and the environment.

The aim of the workshop was to illustrate how methodological tools and concepts developed in political economy can be used to analyze environmental issues. In particular, the organizers wanted to highlight what happens when decisions are not taken by a benevolent social planner, but result from bargaining among different parties, voting, predatory behaviors of the government, cultural norms, etc. The contributions were mainly theoretical while simulation and empirical studies were also presented.

The following report summarizes the different contributions presented at the workshop. It is organized by themes with a concluding section.

International environmental agreements
A major theme tackled during the workshop is one of particular interest for CTN: International Environmental Agreements (IEAs). It gave rise to four stimulating presentations focusing on different aspects of IEAs.
Altamirano-Cabrera, Weikard and Haffoudhi examine the effects of political pressure groups (lobbies) on the size and stability of IEAs. They consider two types of lobbies: industry and environmentalists. They show the potential effects of lobbying using the STAbility of COalitions (STACO) model. They find that lobby contributions may help to establish an IEA that would not emerge in the absence of lobbying. However, the stable agreement provides little additional environmental benefits. Secondly, they find that regions participating in an IEA will not always be influenced by an environmentalist lobby. They show that, contrary to intuition, a member of a stable agreement may well collect industry contributions.

Wagner estimates strategic complementarities in the case of the Montréal Protocol on substances that deplete the ozone layer. He develops a strategic model of the timing of treaty ratification which is amenable to structural estimation. The model predicts that strategic complementarities accelerate ratification, to the extent that ratification by one country may trigger ratification by another. He exploits this prediction to identify strategic complementarities in a structural econometric model of ratification of the Montreal Protocol. He estimates that strategic complementarities reduced the average time to ratification by 35 weeks. The main sources of strategic complementarities are concerns about reputation and the preference for equity. There is no evidence that bilateral trade flows affected the strength of strategic complementarities.

Weikard, Wangler and Freytag study the impact of minimum participation rules (MPRs) on IEAs. They consider a cartel game with open membership and heterogeneous countries to study the endogenous choice of MPRs and their role for the success of international agreements. In particular they extend previous studies to the case of heterogeneous countries. In their model, where the threat is full break-up and not one-by-one defections, the outcome is almost full cooperation.

Z. Yang studies IEAs that can embody the ideals of fairness and historical justice. He uses a Negishi-type result that connects a specific bargaining equilibrium with the solution of a social optimum problem. He investigates whether taking into account past emissions can modify the burden borne by specific countries and the overall emission reduction in an IEA on climate change. He finds that few differences would be obtained compare to a Lindhal equilibrium.

Political economy approaches to environmental policies
Several papers have investigated different political economy models that can explain the design of environmental policies within countries.
Chiroleu-Assouline and Fodha endorse a policy reform approach to investigate which changes in the tax schedule would be unanimously acceptable. They characterize the necessary conditions for the obtaining of a Pareto improving shift when the revenue of the pollution tax is recycled by a change in the labor tax rate or by a change in the distributive properties of the labor tax. They show that, depending on the production function elasticities and on the heterogeneity characteristics of labor supply, an appropriate policy mix could be designed in order to leave each workers' class unharmed by the environmental tax reform. It will consist in an increase of the progressivity of the labor tax together with a decrease of the minimal wage tax rate.

Borissov, Bréchet and Lambrecht study a model of voting on a tax aimed at environmental quality maintenance. They consider two groups of consumers: patient and impatient. The stock of environmental quality is deteriorated through the polluting emissions of firms. They show that the voting equilibrium maintenance is the one of the median voter. They also show a) that an increase in total factor productivity may produce effects described by the Environmental Kuznets Curve, b) an increase in the patience of impatient households may foster environmental quality if the median voter is impatient and maintenance positive, and c) a decrease in inequality may lead to an increase in environmental quality.

De Jaeger, Eyckmans, Van Parijs and Verbeke study the issue of tax-mimicking in the context of waste reduction policies. They estimate a model that shows that waste tourism plays no role in tax-mimicking. The result suggests that tax competition to attract citizens is a driving force.

Morbee develops a political economy model of resource taxation, in order to explain the differences in the share of resource revenues that governments of resource-rich countries claim from the producing companies. The model predicts that that government's take increases when the country's government is dictatorial, but decreases when the government is benevolent (as opposed to predatory). The model is confirmed empirically for the case of petroleum. Government's take also increases when a country is a petroleum exporter or has large discovered petroleum reserves, but it decreases with the size of undiscovered petroleum reserves.

Other topics
Durante examines the impact of the environment on social norms that may affect the economy. He specifically studies the relationship between historical climatic variability and contemporary social trust. He argues that norms of generalized trust developed because they facilitated collective action and mutual insurance among farmers exposed to weather-related risk. The norms have persisted even after weather patterns had become unimportant. He tests the hypothesis in the context of Europe and finds results consistent with his theory. Finally, he finds that climate variability also appears to have a negative effect on the strength of family ties, which suggests that trust within and outside the family may be cultural substitutes.

S. Yang analyses a model where a pollution externality is present. He investigates how to regulate a monopoly firm under asymmetric information on marginal benefit of pollution abatement. He finds that asymmetric information may not distort the first-best effcient level of production and pollution abatement. He also finds that a Pigouvian tax may improve social welfare under perfect information.

Summary
The two-days workshop has shown the diversity of political economy approaches to environmental economics. It has also suggested that many research avenues are open, and that much work can be done on the topic. Here are some lessons from the workshop and some interesting research directions:
- IEAs is very active field of research and it seems to be the more advanced area among the political economy approach of environmental economics. The topic is of great interest both from the theoretical point of view and for its policy implications. There is no denying that many research directions can still be investigated in this domain. In particular, the dynamics of coalition formation, with the possibilities of strategic complementarities and renegotiation, is a promising research avenue.
- Other political approaches seem surprisingly underdeveloped, although the contributions presented during the workshop suggest that interesting results could be obtained in that direction. In particular models of voting on environmental policies may bring interesting new insights. The environment is a domain where public opinion has played a key role in recent years. Studying of public opinion can shape policy would be of great interest. Relatedly, lobbies are known to have a crucial role in the making of environmental policies. Research on the topic should be encouraged.

The full program of the workshop as well as the list of papers and presentations can be found at: http://www.uclouvain.be/en-269709.html.



Report: Workshop MINT 1 - Models of Influence and Network Theory

Agnieszka Rusinowska, GATE, CNRS - University of Lyon

The First International Workshop "Models of Influence and Network Theory" (MINT 1) was held at February 15 and 16 in Lyon, Ecully. The workshop was organized by the CNRS laboratory GATE Lyon St Etienne (Groupe d'Analyse et de Théorie Economique, University of Lyon) and sponsored by the ANR (National Agency for Research - Agence Nationale de la Recherche). It was the first scientific meeting of the project MINT - Models of Influence and Network Theory (Programme Blanc, ANR-09-BLAN-0321-01) which is an official collaboration between researchers of two CNRS laboratories (GATE and CES - Centre d'Economie de la Sorbonne - University of Paris 1) and the Cooperation Institute on Social Choice Theory (SKT - Sociale Keuze Theorie) in the Netherlands.

The workshop was attended by about 20 participants working in France (University of Lyon, University of Saint Etienne, University of Paris 1, Ecole Polytechnique, Ecole Normale Supérieure LSH), in the Netherlands (Tilburg University, Maastricht University, University of Nijmegen), in Belgium (Facultés Universitaires Saint-Louis in Brussels, University of Gent), in Hungary (Óbuda University in Budapest), in Russia (Saint Petersburg), and in Germany (University of Leipzig).

The main aims of the workshop were: discussing selected MINT research topics (e.g., graph games, coalition formation, coalitional network games, power and influence indices, ingratiation, assignment rules), searching for the common research interests, and starting (or continuing) collaboration between the workshop participants on the topics related to the MINT project. During this two-days workshop, 12 talks were presented and discussed. Although most of the papers presented a theoretical approach to the topics in question, two contributions concerned experiments, and some papers illustrated theoretical results by using data on coalitions formed in reality.

Several papers presented during the first day of the workshop were devoted to graph games. Jean-Jacques Herings presented his joint work with Gerard van der Laan, Dolf Talman and Zaifu Yang on the average tree (AT) solution for cooperative games with communication structure. First, the AT solution on the class of cycle-free graph games is proposed, and next the AT solution is extended to the class of all games for communication structures. It is shown that the AT solution coincides with the Shapley value for games with a complete communication graph. The authors introduce the notion of link-convexity, under which the game is shown to have a non-empty core. The AT solution belongs to the core whenever the game is link-convex. In general, link convexity is weaker than convexity. For games with a cycle-free communication structure, link-convexity is even weaker than super-additivity.

The paper by Sylvain Béal, Eric Rémila and Philippe Solal (presented by Sylvain during the workshop) concerns compensations in the Shapley value and the compensation solution for graph games. The authors introduce an interpretation of the Shapley value that is based on a system of compensations. Next, this idea is generalized to cooperative games with a communication graph. They consider forest graphs and extend the compensation vector by considering all rooted spanning trees of the graph, and characterize the associated allocation rule, called the compensation solution. Next, arbitrary graphs are considered and rooted spanning trees are constructed by using the algorithms DFS (for Depth-First Search) and BFS (for Breadth-First Search). It is shown that for a complete graph, the compensation solutions associated with DFS and BFS coincide with the Shapley value and the equal surplus division, respectively.

Anna Khmelnitskaya presented her work with Dolf Talman on values for cycle-free directed graph games. The authors consider a class of cycle-free directed graph games in which all players are partially ordered and a possible communication via bilateral agreements between participants is presented by a directed graph without directed cycles. They introduce values for cycle-free digraph games axiomatically, provide the explicit formula representation, and study their stability. Furthermore, it is shown that the problem of sharing a river with a delta and with multiple sources among different agents located at different levels along the river bed can be embedded into the framework of a cycle-free digraph game.

The paper presented by Amandine Ghintran is her joint work with Sylvain Béal, Eric Rémila and Philippe Solal and concerns egalitarian tree solutions for graph games. The associated payoff vector can be characterized by component efficiency and component fairness for subordinates. For each nonempty set of rooted spanning trees, the authors define the egalitarian tree solutions as the average over this set of the associated payoff vectors. For the class of forest graph games, they provide two axiomatic characterizations of the egalitarian tree solution. A further extension to arbitrary graph games is studied in which rooted spanning trees are constructed by classical algorithm DFS and BFS. It is shown that when the graph is complete, then the associated egalitarian tree solutions coincide with the consensus value and the equal surplus division, respectively. As such, the egalitarian tree solution can be seen as a generalization of the consensus value for graph games.

The work by Aymeric Lardon concerns the γ-core in Cournot oligopoly TU-games with capacity constraints. It is assumed that firms react to a deviating coalition by choosing individual best reply strategies. The author deals with the problem of the non-emptiness of the induced core, the γ-core, by applying two different approaches. The first one establishes that the associated Cournot oligopoly TU-games are balanced if the inverse demand function is differentiable and every individual profit function is continuous and concave on the set of strategy profiles. The second approach, restricted to the class of Cournot oligopoly TU-games with linear cost functions, provides a single-valued allocation rule in the γ-core called Nash Pro rata value. The author provides an axiomatic characterization of this solution by means of four properties: efficiency, null player, monotonicity and weighted fairness.

The second day of the workshop started with a talk given by Jean-François Caulier who presented his joint work with Ana Mauleon and Vincent Vannetelbosch on coalitional network games. Coalitional network games are real-valued functions defined on a set of players organized into a network and a coalition structure. Such games model situations where the total productive value of a network among players depends on the players' group membership, and can thus capture the public good aspect of bilateral cooperation, i.e., network games with externalities. After studying the specific structure of coalitional networks, the authors propose an allocation rule under the perspective that players can alter the coalitional network structure. This means that the value of all potential alternative coalitional networks can and should influence the allocation of value among players in any given coalitional network structure.

Several presentations of the workshop concerned power and influence indices. The first one on power to influence and influence to power was delivered by Laszlo Koczy. He introduces a model where players can choose their partners in cooperation. The so called strategic power indices are power indices for rational power-maximising players. For normalised indices strategic quarrelling may increase power. It is shown that strategic power is well defined if power is measured by an index that takes only minimal winning coalitions into account. The author presents an example based on the Council of Budapest.

Harrie de Swart presented his joint paper with Rudolf Berghammer and Agnieszka Rusinowska on a relation-algebraic approach to simple games. The authors present relation-algebraic models of simple games and develop relational specifications for solving some basic problems of them. In particular, they test certain fundamental properties of simple games and compute specific players and coalitions, like e.g. minimal winning coalitions, swingers, dominant and central players. They also apply relation algebra to determine power indices. This leads to relation-algebraic specifications, which can be evaluated with the help of the BDD-based tool RelView after a simple translation into the tool's programming language. In order to demonstrate the visualization facilities of RelView, the authors consider an example of the city council of Tilburg.

The paper presented by Michel Grabisch that is joint research with Agnieszka Rusinowska concerns a model of influence with a continuum of actions. A two-action (yes-no) model of influence is generalized to a framework in which every player has a continuum of actions and he has to choose one of them. Each player has an inclination to choose one of the actions, but due to influence among players, the final decision of a player, i.e., his choice of one action, may be different from his original inclination. The authors introduce and study a measure of positive influence of a coalition on a player, as well as a measure of negative influence. Several unanimous influence functions and linear influence functions are studied. The authors also investigate other tools for analysing influence, like e.g. the concept of a follower of a coalition. The set of fixed points under a given influence function is analysed.

The first talk in the afternoon of the second workshop day was delivered by Francis Bloch who presented his paper with Nicolas Houy on optimal assignment of durable objects to successive agents. The authors analyse the assignment of durable objects to successive generations of agents who live for two periods. The optimal assignment rule is stationary, favours old agents and is determined by a selectivity function which satisfies an iterative functional differential equation. More patient social planners are more selective, as are social planners facing distributions of types with higher probabilities for higher types. The authors also characterize optimal assignment rules when monetary transfers are allowed and agents face a recovery cost, when agents' types are private information and when agents can invest to improve their type.

As mentioned before, also experimental economics was represented during the workshop, although not in majority. Two papers on experiments have been presented: one dealing with an experiment on coalition formation and one with an experiment on influence and ingratiation. The workshop started with an "experimental talk" delivered by Jana Vyrastekova. She presented her paper with Yuki Funaki and Ai Takeuchi on their experiment on coalition formation in a common pool resource game. Until now, behaviour in social dilemmas has been studied experimentally mainly in the context of individual decision-making, but little is known about the role of coalition formation. The authors address this issue by an experiment where subjects are not forced to form singleton coalitions. The only restriction on a larger coalition to be formed is that all its members unanimously agree on an action profile for the members of the coalition. In their experiments, the authors observe behaviour that is strikingly different from the behaviour in the individualistic environments studied by now. Efficient and symmetric solution is nearly exclusively observed in the laboratory.

The workshop finished in the experimental laboratory of GATE. All workshop participants were invited to participate in a pilot experiment on influence and ingratiation which is a project conducted by the MINT participants working at GATE: Stéphane Robin, Agnieszka Rusinowska and Marie-Claire Villeval. The experiment is related to the model of influence presented by Michel Grabisch. While in the traditional framework of leadership, usually research on influence that a leader exercises on the follower(s) is conducted, in the model of ingratiation tested in this experiment, the authors focus on "influencing" the decider's opinion by the agents who are evaluated by the decider. They study the concept of ingratiation, that is, the willingness of people to change their preliminarily declared opinions (inclinations) to improve their expected monetary payoffs by influencing the opinion of a decider.

We can fairly conclude that the main aims of the workshop have been realized successfully. The MINT participants plan to meet regularly and to invite for collaboration other researchers working on similar topics. During the official realization period of the MINT project, which will last till September 2013, at least three more MINT workshops and some joint MINT seminars are planned. Moreover, we intend to organize special MINT sessions during some scientific meetings, in particular, during some annual conferences on game theory. We would like to keep the variety of our approaches applied to the main MINT topics, that is, to conduct joint research by using theoretical, experimental, and empirical approaches to, e.g., networks, coalitions, and influence.

The full program of the workshop with the attached papers and the list of all participants can be found at: http://www.gate.cnrs.fr/mint1



Visit the CTN Members' seminars web pages

CODE: http://idea.uab.es/code/workshops.htm
CORE: http://www.uclouvain.be/en-43617.html
CES: http://ces.univ-paris1.fr/seminaires_ces.htm
FEEM: http://www.feem.it/Feem/Pub/Seminars/default.htm
Greqam: http://greqam.univ-mrs.fr/seminars.php
Maastricht: http://www.fdewb.unimaas.nl/meteor-seminar-et
Vanderbilt: http://www.vanderbilt.edu/econ/seminars-research.htm
Warwick: http://www2.warwick.ac.uk/fac/soc/economics/forums/

 

About the CTN Newsletter

The CTN Newsletter is prepared with the contribution of all the CTN Partner Institutions. Please send comments and questions to: silvia.bertolin@feem.it.
The next issue will be published in September 2010

Subscribe to the CTN Newsletter