## Learning Dynamics In Mechanism Design: An Experimental Comparison Of Public Goods Mechanisms (2003)

Citations: | 12 - 3 self |

### BibTeX

@MISC{Healy03learningdynamics,

author = {Paul J. Healy},

title = {Learning Dynamics In Mechanism Design: An Experimental Comparison Of Public Goods Mechanisms},

year = {2003}

}

### OpenURL

### Abstract

In a repeated-interaction public goods economy, dynamic behavior may affect the efficiency of various mechanisms thought to be efficient in one-shot games. Inspired by results obtained in previous experiments, the current paper proposes a simple best response model in which players' beliefs are functions of previous strategy profiles. The predictions of the model are found to be highly consistent with new experimental data from five mechanisms with various types of equilibria. Interesting properties of a 2parameter Vickrey-Clarke-Groves mechanism help to draw out this result. The simplicity of the model makes it useful in predicting dynamic stability of other mechanisms.

### Citations

2552 |
An Introduction to the Bootstrap
- Efron, Tibshirani
- 1993
(Show Context)
Citation Context ...rate and more precise than the equilibrium model whenever differences between the two are observed. The statistical hypothesis of interest is whether or not the expected equi32 See Efron & Tibshirani =-=[32]-=- for details on the bootstrapping method and related statistical tests. 30Player 5 Player 4 Player 3 Player 2 Player 1 14 7 0 −7 −14 14 7 0 −7 −14 14 7 0 −7 −14 14 7 0 −7 −14 14 7 0 −7 −14 5−period B... |

1210 | Counterspeculation, Auctions, and Competitive Sealed Tenders - Vickrey - 1961 |

1150 | Nonparametric statistics for the behavioral sciences. 2nd ed - Siegel, Jr - 1988 |

706 |
Multipart Pricing of Public Goods
- Clarke
- 1971
(Show Context)
Citation Context ...ontinuous VCG (cVCG) Mechanism The cVCG mechanism represents a particular selection from the class of dominant strategy incentive compatible mechanisms developed independently by Vickrey [28], Clarke =-=[29]-=-, and Groves [30]. In these direct mechanisms (where Mi = Θi), truth-telling weakly dominates all other strategies. However, given m−i, there may exist messages mi ̸= θi such that mi ∈ Bi (m−i; θi). I... |

561 | Public Goods: A Survey of Experimental Research - Ledyard - 1995 |

514 | Rheinboldt, Iterative Solution of Nonlinear Equations - Ortega, C - 1970 |

242 |
Existence and Uniqueness of Equilibrium Points for Concave N-Person
- Rosen
- 1965
(Show Context)
Citation Context ...eriod dynamic must satisfy the adaptive dynamics conditions of Milgrom & Roberts [15]. Proposition 7 then follows from Theorem 8 of that paper. 14 A similar result was previously established by Rosen =-=[18]-=- using the diagonal strict concavity condition on utilities. 15 In this case, positive diagonal dominance occurs when there exists some vector d > 0 such that dj > ∑n i=1 diaij (θ) for each j = 1, . .... |

236 | Characterization of Satisfactory Mechanisms for the Revelation of Preferences for Public Goods - Green, Laffont - 1977 |

168 | Rationalizability, learning, and equilibrium in games with strategic complementarities
- Milgrom, Roberts
- 1990
(Show Context)
Citation Context ...ete sublattice values in these games, so inf B (m, θ) and sup B (m, θ) are both non-decreasing functions of m whose values lie in B (m, θ). Using these properties and the methods of Milgrom & Roberts =-=[15]-=-, global stability of the k-period dynamic is established. Proposition 7 In a supermodular game, if {m t } ∞ 1 is consistent with a kperiod average best response dynamic and E (θ) and E (θ) are the sm... |

149 |
Iterative solution of games by fictitious play
- Brown
- 1951
(Show Context)
Citation Context ...der, for example, the fictitious play dynamic, which is best suited for games with small, finite strategy spaces. As with the k-period model, fictitious play has stable Nash equilibrium points (Brown =-=[23]-=-) and is globally stable in supermodular games (Milgrom & Roberts [24, Theorem 8],) but is also capable of off-equilibrium limit cycles (Shapley [25].) 4 Experimental Design The five public goods mech... |

117 | Econometric Analysis, 4th Edition - Greene - 2000 |

110 |
Optimal allocation of public goods: A solution to the ‘free rider’ problem. Econometrica 45(4): 783–809
- Groves, Ledyard
- 1977
(Show Context)
Citation Context ...all messages. Note that y (m ∗ (θ)) > y P (θ), so the equilibrium level of the public good is inefficiently large. 5.3 Groves-Ledyard Mechanism This efficient mechanism, developed by Groves & Ledyard =-=[26]-=-, requires all agents pay an equal share of the cost plus a penalty term based on deviations from the average of the others’ contributions and on the variance of those contributions. Formally, the out... |

107 |
Equilibrium Points in Nonzero-Sum n-Person Submodular Games
- Topkis
- 1979
(Show Context)
Citation Context ...ative values of k. The above family of examples is chosen to have a particular non-monotonic, non-contracting structure to the best response correspondence. In the class of supermodular games (Topkis =-=[11]-=-) or in games satisfying a dominant diagonal condition (Gabay & Moulin [14],) the best response correspondence is known to have properties sufficient to guarantee global stability of the 1-period best... |

102 | Independent Private Value Auctions: Bidder Behaviour in First-, Second-, and ThirdPrice Auctions with Varying Numbers of Bidders - Kagel, Levin - 1993 |

88 |
Some topics in two-person games
- Shapley
- 1964
(Show Context)
Citation Context ...ous play has stable Nash equilibrium points (Brown [23]) and is globally stable in supermodular games (Milgrom & Roberts [24, Theorem 8],) but is also capable of off-equilibrium limit cycles (Shapley =-=[25]-=-.) 4 Experimental Design The five public goods mechanisms under consideration were tested in a laboratory environment using human subjects. All experiments were run at the California Institute of Tech... |

85 | On informationally decentralized systems - Hurwicz - 1972 |

76 |
Public goods provision in an experimental environment
- Isaac, McCue, et al.
- 1985
(Show Context)
Citation Context ...y provide contributions halfway between the Pareto-efficient level and the free riding level,” and that “contributions decline with repetition.” For example, in an early paper by Isaac, McCue & Plott =-=[3]-=-, payoffs dropped from 50% of the maximum in the first period to 9% by the fifth period. Strategies quickly converge toward the free-riding dominant strategy through repetition. In the decades since t... |

74 | Evolutionary stability in asymmetric games - Samuelson, Zhang - 1992 |

73 |
Mean value methods in iteration
- Mann
- 1953
(Show Context)
Citation Context ...ording to the k-period average best response dynamic and B is upper hemi-continuous and non-empty, 9 For a proof of this and other useful facts about weighted averages of infinite sequences, see Mann =-=[9]-=- and Dotson [10]. t=1 10then q is a Nash equilibrium strategy profile. Asymptotic stability requires that the dynamic path from all initial points in some neighborhood of q converge to q. By Proposit... |

73 | Adaptive and sophisticated learning in normal form games - Milgrom, Roberts - 1991 |

54 | Multidimensional incentive compatibility and mechanism design - McAfee, McMillan - 1988 |

48 |
Axiomatic characterization of the quadratic scoring rule
- Selten
- 1998
(Show Context)
Citation Context ...e average score generally decreases in tmin, indicating that the models are less accurate in early periods than in later 25 A ‘scoring rule’ such as the quadratic scoring rule characterized by Selten =-=[31]-=- would be more appropriate if the behavioral models generated probabilistic predictions of play. With deterministic behavioral models and a continuum strategy space, the scoring rule simply counts the... |

45 |
Incentives in teams, Econometrica 41
- Groves
- 1973
(Show Context)
Citation Context ...CG) Mechanism The cVCG mechanism represents a particular selection from the class of dominant strategy incentive compatible mechanisms developed independently by Vickrey [28], Clarke [29], and Groves =-=[30]-=-. In these direct mechanisms (where Mi = Θi), truth-telling weakly dominates all other strategies. However, given m−i, there may exist messages mi ̸= θi such that mi ∈ Bi (m−i; θi). In the cVCG mechan... |

39 | Outcome Functions Yielding Walrasian and Lindahl Allocations at Nash Equilibrium Points - Hurwicz - 1979 |

33 | The Groves-Ledyard mechanism: An experimental study of institutional design
- Chen, Plott
- 1996
(Show Context)
Citation Context ...blem, experimental tests have focused primarily on Nash mechanisms. In particular, the Groves-Ledyard mechanism has been studied by several authors in repeated one-shot laboratory tests. Chen & Plott =-=[4]-=- study the effect of the punishment parameter and find that strategies converge rapidly to equilibrium for large parameter values; an observation consistent with known convergence results for best res... |

27 |
A Simple Incentive Compatible Scheme for Attaining Lindahl Allocations
- Walker
- 1981
(Show Context)
Citation Context ...rst to Nash implement optimal public goods allocations. However, the mechanism has a few shortcomings, including the fact that its outcomes may not satisfy individual rationality. In response, Walker =-=[27]-=- developed the following ‘paired difference’ mechanism that implements the Lindahl allocation in Nash equilibrium. The outcome functions are defined as y (m) = ∑ i∈I mi and respectively, where τi (y (... |

22 | Fictitious Play: A Statistical Study of Multiple Economic Experiments - BOYLAN, EL-GAMAL - 1993 |

22 | Incentive Compatible Experimental Processes for the Provision of Public Goods - Smith - 1979 |

19 | Experiments with the Pivot Process for Providing Public Goods
- Attiyeh, Franciosi, et al.
- 2000
(Show Context)
Citation Context ...best response dynamic models. The first well-controlled public goods test of the dominant strategy Pivot mechanism (the VCG mechanism with a binary public choice) is run by Attiyeh, Franciosi & Isaac =-=[6]-=-. Subjects are given positive or negative values for a proposed project and must submit a message indicating their demand. For example, large negative messages represent strong opposition to the proje... |

19 | Incentive Compatibility since 1972 - Groves, Ledyard - 1987 |

19 | Characterizations of the Pivotal Mechanism - Moulin - 1986 |

19 |
On the uniqueness and stability of Nash-equilibria in noncooperative games
- Gabay, Moulin
- 1980
(Show Context)
Citation Context ...lar non-monotonic, non-contracting structure to the best response correspondence. In the class of supermodular games (Topkis [11]) or in games satisfying a dominant diagonal condition (Gabay & Moulin =-=[14]-=-,) the best response correspondence is known to have properties sufficient to guarantee global stability of the 1-period best response dynamic. The following propositions demonstrate that these global... |

18 | Are Groves–Ledyard equilibria attainable - Muench, Walker - 1984 |

18 | When Does Learning in Games Generate Convergence to Nash Equilibria? The Role of Supermodularity in an Experimental Setting
- Chen, Gazzale
- 2005
(Show Context)
Citation Context ...m′, θ) and ˆm ∧ ˆm ′ ∈ B (m, θ). 12The proof of Proposition 7 appears in the appendix 13 . These results are of particular significance given the recent claims of Chen [16, 2004b] and Chen & Gazzale =-=[17]-=- that supermodularity is sufficient for convergence in a variety of environments tested in the laboratory. Since the k-period dynamic is predicted to converge under supermodularity, outside evidence a... |

17 |
Learning and incentive-compatible mechanisms for public goods provision: An experimantal
- Chen, Tang
- 1998
(Show Context)
Citation Context ...nd that strategies converge rapidly to equilibrium for large parameter values; an observation consistent with known convergence results for best response dynamics in su3permodular games. Chen & Tang =-=[5]-=- compare the Groves-Ledyard mechanism to the Walker mechanism. The results of these studies imply that the GrovesLedyard mechanism is significantly more efficient than the Walker mechanism, apparently... |

17 | Selection Dynamics, Asymptotic Stability, and Adaptive Behavior - Huyck, Cook, et al. - 1994 |

17 | The Set of Nash Equilibria of a Supermodular Game Is a Complete Lattice - Zhou - 1994 |

16 | Implementation, elimination of weakly dominated strategies and evolutionary dynamics. Rev. Econ. Dynam. 3, 247–282. personal copy 374 - Cabrales, Ponti - 2000 |

15 | A Differential Approach to Dominant Strategy Mechanisms - Laffont, Maskin - 1980 |

14 | Adaptation and convergence of behavior in repeated experimental Cournot games - Rassenti, Reynolds, et al. - 2000 |

14 | On the Nonexistence of a Dominant Strategy Mechanism for Making Optimal Public Decisions - Walker - 1980 |

12 | Secure implementation experiments: Do strategy-proof mechanisms really work
- Cason, Saijo, et al.
- 2003
(Show Context)
Citation Context ...es in this experiment are not significantly different from that expected by completely randomized strategy choice, possibly indicating confusion among subjects. Most recently, Cason, Saijo & Sjostrom =-=[8]-=- compare the Pivot and cVCG mechanisms and show that deviations from truthful revelation, when they occur, tend to result in Nash equilibria in weakly dominated strategies. These weakly dominated equi... |

12 | Groves’scheme on restricted domains. Econometrica - Holmström - 1979 |

11 |
On theMann iterative process
- Dotson
(Show Context)
Citation Context ...-period average best response dynamic and B is upper hemi-continuous and non-empty, 9 For a proof of this and other useful facts about weighted averages of infinite sequences, see Mann [9] and Dotson =-=[10]-=-. t=1 10then q is a Nash equilibrium strategy profile. Asymptotic stability requires that the dynamic path from all initial points in some neighborhood of q converge to q. By Proposition 4, this is c... |

10 | On the Generic Non-optimality of Dominant Strategy Allocation Mechanisms - Hurwicz, Walker - 1990 |

10 |
Implementation of Lindahl equilibrium: an integration of the static and dynamic approaches
- Vega-Redondo
- 1989
(Show Context)
Citation Context ...nt. However, restricting attention to quasilinear environments, a dynamically stable game form is introduced that implements the Lindahl correspondence. Similarly, de Tranqualye [34] and Vega-Redondo =-=[35]-=- introduce mechanisms that converge to the Lindahl allocation under Cournot best response behavior. Some mechanisms such as that studied by Smith [36] use convergence of tatonnemont-like dynamics as a... |

9 |
Can the pivotal mechanism induce truth-telling? An experimental study
- Kawagoe, Mori
(Show Context)
Citation Context ...ing one’s demand is a (weak) dominant strategy, only ten percent of observations are consistent with this prediction, with thirteen of twenty subjects never revealing their true value. Kawagoe & Mori =-=[7]-=- extend the Attiyeh et al. result by comparing the above treatment to one in which subjects are given a payoff table. The effect of having players choose from the table is significant, as demand revel... |

9 | Pricing Public Goods: An Experiment with Two Proposed Pricing Systems," Public Choice - Scherr, Babb - 1975 |

8 | Mathematics for stability and optimization of economic systems - Murata - 1977 |

6 | Dynamic stability of Nash-efficient public goods mechanisms: Reconciling theory and experiments - Chen - 2004 |