nep-gth New Economics Papers
on Game Theory
Issue of 2009‒11‒27
fourteen papers chosen by
Laszlo A. Koczy
Budapest Tech and Maastricht University

  1. The Strategy of Manipulating Conflict By Sandeep Baliga; Tomas Sjostrom
  2. Leadership in a Weak-Link Game By Joris Gillet; Edward Cartwright; Mark Van Vugt
  3. The Lifeboat Problem By Konrad, Kai A.; Kovenock, Dan
  4. Pre-electoral Coalitions and Post-election Bargaining By Siddhartha Bandyopadhyay; Kalyan Chatterjee; Tomas Sjostrom
  5. Strategies in Social Network Formation By Anna Conte; Daniela Di Cagno; Emanuela Sciubba
  6. A Dynamic Model of Network Formation with Strategic Interactions By König, Michael; Tessone, Claudio J.; Zenou, Yves
  7. Group Membership, Team Preferences, and Expectations By Francesco Guala; Luigi Mittone; Matteo Ploner
  8. A New Payment Rule for Core-Selecting Package Auctions By Erdil, Aytek; Klemperer, Paul
  9. Measuring attention and strategic behavior in games with private information By Brocas, Isabelle; Camerer, Colin; Carrillo, Juan D; Wang, Stephanie W.
  10. Strategic Vagueness and Appropriate Contexts By Kris De Jaegher; Robert van Rooij
  11. Communication, Renegotiation, and the Scope for Collusion By Cooper, David J.; Kühn, Kai-Uwe
  12. Testing Unilateral and Bilateral Link Formation By Comola, Margherita; Fafchamps, Marcel
  13. Existence of an interim and ex-ante minimax point for an asymmetric information game By Marialaura Pesce; Nicholas C Yannelis
  14. The Incentive Effects of Second Prizes in a Dynamic Contest Model By Martin Grossmann; Helmut Dietl; Markus Lang

  1. By: Sandeep Baliga (Northwestern); Tomas Sjostrom (Rutgers)
    Abstract: Two decision-makers choose hawkish or dovish actions in a conflict game with incomplete information. The decision-making can be manipulated by "extremists" who send publicly observed cheap-talk messages. The power of extremists depends on the nature of the underlying conflict game. If actions are strategic complements, a "hawkish extremist" can increase the likelihood of conflict by sending messages which trigger a "fear-spiral" of hawkish actions. This reduces the welfare of both decision-makers. If actions are strategic substitutes, a "dovish extremist" (pacifist) can send messages which cause one decision-maker to back down and become more dovish. This reduces his welfare but benefits the other decision-maker. The hawkish extremist is unable to manipulate the decision-makers if actions are strategic substitutes, and the pacifist is equally powerless if actions are strategic complements.
    Keywords: global strategy
    JEL: C7
    Date: 2009–08–28
  2. By: Joris Gillet; Edward Cartwright; Mark Van Vugt
    Abstract: We investigate, experimentally, the effects of leadership in a four player weak-link game. A weak-link game is a coordination game with multiple Pareto-ranked Nash equilibria. Because the more efficient equilibria involve a degree of strategic uncertainty groups typically find it difficult to coordinate on more efficient equilibria. Previous studies have shown that leadership by example - in the form of one player acting publicly before the rest of the group - can lead to increased cooperation in collective action problems and we are interested in finding out whether this result extends to weak-link games. Our results suggest that leadership has no effect on initial behavior; the first time that they play the game participants behave the same with leadership as without. We also observe, however, that leadership can allow groups to raise efficiency over time and therefore overcome inefficiency. There doesn't appear to be a difference between voluntary leaders and leaders that are (randomly) appointed.
    Keywords: Leadership; coordination game; weak-link game; minimum-effort game
    JEL: C72 D01 H41
    Date: 2009–10
  3. By: Konrad, Kai A.; Kovenock, Dan
    Abstract: We study an all-pay contest with multiple identical prizes ("lifeboat seats"). Prizes are partitioned into subsets of prizes ("lifeboats"). Players play a two-stage game. First, each player chooses an element of the partition ("a lifeboat"). Then each player competes for a prize in the subset chosen ("a seat"). We characterize and compare the subgame perfect equilibria in which all players employ pure strategies or all players play identical mixed strategies in the first stage. We find that the partitioning of prizes allows for coordination failure among players when they play nondegenerate mixed strategies and this can shelter rents and reduce rent dissipation compared to some of the less efficient pure strategy equilibria.
    Keywords: all-pay contest; lifeboat; multiple prizes; rent dissipation
    JEL: D72 D74
    Date: 2009–08
  4. By: Siddhartha Bandyopadhyay (Birmingham); Kalyan Chatterjee (Penn State); Tomas Sjostrom (Rutgers)
    Abstract: Pre-electoral coalitions occur frequently in parliamentary democracies. They influence post election coalition formation and surplus division. We study a game theoretic model where political parties can form coalitions both before (ex ante) and after (ex post) the elections. Ex ante coalitions can commit to a seat-sharing arrangement, but neither to a policy nor to a division of rents from office; coalition members are even free to break up and join other coalitions after the election. Equilibrium ex ante coalitions are not necessarily made up of the most ideologically similar parties, and they form under (national list) proportional representation as well as plurality rule. They do not form just to avoid "splitting the vote", but also because seat-sharing arrangements will influence the ex post bargaining and coalition formation. The ex post bargaining protocol matters greatly: there is more scope for coalition formation, both ex ante and ex post, under an Austen-Smith and Banks protocol than under "random recognition".
    Keywords: NA
    Date: 2009–09–15
  5. By: Anna Conte (Max-Planck-Institut für Ökonomik); Daniela Di Cagno (LUISS Guido Carli); Emanuela Sciubba (Birkbeck College)
    Abstract: We run a computerised experiment of network formation where all connections are beneficial and only direct links are costly. Players simultaneously submit link proposals; a connection is made only when both players involved agree. We use both simulated and experimentally generated data to test the determinants of individual behaviour in network formation. We find that approximately 40% of the network formation strategies adopted by the experimental subjects can be accounted for as best responses. We test whether subjects follow alternative patterns of behaviour and in particular if they: propose links to those from whom they have received link proposals in the previous round; propose links to those who have the largest number of direct connections. We find that together with best response behaviour, these strategies explain approximately 75% of the observed choices. We estimate individual propensities to adopt each of these strategies, controlling for group effects. Finally we estimate a mixture model to highlight the proportion of each type of decision maker in the population.
    Keywords: network formation, experiments, mixture models
    Date: 2009–11–16
  6. By: König, Michael; Tessone, Claudio J.; Zenou, Yves
    Abstract: In order to understand the different characteristics observed in real-world networks, one needs to analyze how and why networks form, the impact of network structure on agents' outcomes, and the evolution of networks over time. For this purpose, we combine a network game introduced by Ballester et al. (2006), where the Nash equilibrium action of each agent is proportional to her Bonacich centrality, with an endogenous network formation process. Links are formed on the basis of agents' centrality while the network is exposed to a volatile environment introducing interruptions in the connections between agents. A remarkable feature of our dynamic network formation process is that, at each period of time, the network is a nested split graph. This graph has very nice mathematical properties and are relatively easy to characterize. We show that there exists a unique stationary network (which is a nested split graph) whose topological properties completely match features exhibited by real-world networks. We also find that there exists a sharp transition in efficiency and network density from highly centralized to decentralized networks.
    Keywords: Bonacich centrality; nested split graphs; network formation; social interactions
    JEL: A14 C63 D85
    Date: 2009–10
  7. By: Francesco Guala; Luigi Mittone; Matteo Ploner
    Abstract: Group membership increases cooperation in social dilemma games, altruistic donation in dictator games, and fair offers in ultimatum games. While the empirical study of group action has grown rapidly over the years, there is little agreement at the theoretical level on exactly why and how group membership changes individual behaviour. According to most theorists, the effect of group framing is channelled primarily via the beliefs of group members, while a dissenting minority identifies changes in preference as the key explanatory mechanism. We report an experiment using the minimal group paradigm and a prisoner’s dilemma with multiple actions, in which we manipulate players’ beliefs and show that mutual knowledge of group affiliation is not necessary for group action. Our results question previous empirical findings, refute theories of social norms based on mutual expectations, and support a specific theory of team preferences based on “circumspect reasoning”
    Keywords: group identity, team preferences, social dilemmas, experimental economics.
    JEL: C72 C91 H41
    Date: 2009
  8. By: Erdil, Aytek; Klemperer, Paul
    Abstract: We propose a new, easy-to-implement, class of payment rules, "Reference Rules," to make core-selecting package auctions more robust. Small, almost riskless, profitable deviations from "truthful bidding" are often easy for bidders to find under currently-used payment rules. Reference Rules perform better than existing rules on our marginal-incentive-to-deviate criterion, and are as robust as existing rules to large deviations. Other considerations, including fairness and comprehensibility, also support the use of Reference Rules.
    Keywords: combinatorial auction; core; core-selecting auction; multi-object auction; package auction; robust design; simultaneous ascending auction; Vickrey; Vickrey auction
    JEL: C71 D44
    Date: 2009–10
  9. By: Brocas, Isabelle; Camerer, Colin; Carrillo, Juan D; Wang, Stephanie W.
    Abstract: In experiments, people do not always appear to think very strategically or to infer the information of others from their choices. To understand this thinking process further, we use "Mousetracking" to record which game payoffs subjects look at, for how long, in games of private information with three information states, which vary in strategic complexity. Subjects often deviate from Nash equilibrium choices, converge only modestly toward equilibrium across 40 trials, and often fail to look at payoffs which they need to in order to compute an equilibrium response. Theories such as QRE and cursed equilibrium, which can explain non-equilibrium choices, are not well supported by the combination of both choices and lookups. When cluster analysis is used to group subjects according to lookup patterns and choices, the clusters appear to correspond approximately to level-3, level-2 and level-1 thinking in level-k cognitive hierarchy models. The connection between looking and choices is strong enough that the time durations of looking at key payoffs can predict choices, to some extent, at the individual level and at the trial-by-trial level.
    Keywords: asymmetric information; attention; laboratory experiment; mousetracking
    JEL: C92 D82
    Date: 2009–11
  10. By: Kris De Jaegher; Robert van Rooij
    Abstract: This paper brings together several approaches to vagueness, and ends by suggesting a new approach. The common thread in these approaches is the crucial role played by context. Using a single example where there is a conflict of interest between speaker and listener, we start by treating game-theoretic rationales for vagueness, and for the related concepts of generality and ambiguity. We argue that the most plausible application of these models to vagueness in natural language is one where the listener only imperfectly observes the context in which the speaker makes her utterances. We next look at a rationale for vagueness when there is no conflict between speaker and listener, and which is an application of Horn's rule. Further, we tackle the Sorites paradox. This paradox apparently violates standard axioms of rational behaviour. Yet, once it is taken into account that vague language is used in an appropriate context, these axioms are no longer violated. We end with a behavioural approach to vagueness, where context directly enters agents. preferences. In an application of prospect theory, agents think in terms of gains and losses with respect to a reference point. Vague predicates now allow agents to express their subjective valuations, without necessarily specifying the context.
    Keywords: Vagueness, signalling games, decision theory, prospect theory
    JEL: D82 D83
    Date: 2009–11
  11. By: Cooper, David J.; Kühn, Kai-Uwe
    Abstract: We use experiments to analyze what type of communication is most effective in achieving cooperation in a simple collusion game. Consistent with the existing literature on communication and collusion, even minimal communication leads to a short run increase in collusion. However, in a limited message-space treatment where subjects cannot communicate contingent strategies, this initial burst of collusion rapidly collapses. When unlimited pre-game communication is allowed via a chat window, an initial decline in collusion is reversed over time. Content analysis is used to identify multiple channels by which communication improves collusion in this setting. Explicit threats to punish cheating prove to be by far the most important factor to successfully establish collusion, consistent with the existing theory of collusion. However, collusion is even more likely when we allow for renegotiation, contrary to standard theories of renegotiation. What appears critical for the success of collusion with renegotiation is that cheaters are often admonished in strong terms. Allowing renegotiation therefore appears to increase collusion by allowing for an inexpensive and highly effective form of punishment.
    Keywords: collusion; communication; experiments; guilt aversion; renegotiation; trust
    JEL: C72 C73 C92 D43 L13 L41
    Date: 2009–11
  12. By: Comola, Margherita; Fafchamps, Marcel
    Abstract: The literature has shown that network architecture depends crucially on whether links are formed unilaterally or bilaterally, that is, on whether the consent of both nodes is required for a link to be formed. We propose a test of whether network data is best seen as an actual link or willingness to link and, in the latter case, whether this link is generated by an unilateral or bilateral link formation process. We illustrate this test using survey answers to a risk-sharing question in Tanzania. We find that the bilateral link formation model fits the data better than the unilateral model, but the data are best interpreted as willingness to link rather than an actual link. We then expand the model to include self-censoring and find that models with self-censoring fit the data best.
    Keywords: network architecture; pairwise stability; risk sharing
    JEL: C12 C52 D85
    Date: 2009–08
  13. By: Marialaura Pesce; Nicholas C Yannelis
    Date: 2009
  14. By: Martin Grossmann (Institute for Strategy and Business Economics, University of Zurich); Helmut Dietl (Institute for Strategy and Business Economics, University of Zurich); Markus Lang (Institute for Strategy and Business Economics, University of Zurich)
    Abstract: This paper models an infinitely repeated Tullock contest in which two contestants contribute efforts to accumulate individual asset stocks. We analyze the dynamic effects of second prizes on this accumulation process. Our analysis shows that optimal asset stocks and its speed of convergence to the steady state depend on the elasticity of marginal effort costs. With linear (strictly convex) costs, asset stocks immediately (smoothly) converge to the steady state. If costs are linear, a higher spread between the first and second prize increases incentives to accumulate asset stocks in each period, but does affect competitive balance neither in the short nor in the long run. In case of strictly convex costs, a higher prize spread increases individual and aggregate asset stocks, but does not alter competitive balance in the long run. During the transition, a higher prize spread increases asset stocks as well as the competitive balance in each period.
    Keywords: Dynamic contest,logit contest, multiple prizes, second prizes, rent-seeking
    JEL: C73 D72 L13
    Date: 2009–11

This nep-gth issue is ©2009 by Laszlo A. Koczy. It is provided as is without any express or implied warranty. It may be freely redistributed in whole or in part for any purpose. If distributed in part, please include this notice.
General information on the NEP project can be found at For comments please write to the director of NEP, Marco Novarese at <>. Put “NEP” in the subject, otherwise your mail may be rejected.
NEP’s infrastructure is sponsored by the School of Economics and Finance of Massey University in New Zealand.