Search

We consider a model where the same group of players is involved in more than one cooperative (transferable utility) game. A rule determines the payoffs per game, and for each player a utility function evaluates the resulting vector of payoffs. We assume that each player, independently, can make transfers of worth between different games, thereby affecting its payoff vector and, thus, utility. Two transfer systems are considered, resulting in two distinct noncooperative games, and the focus of the paper is on establishing existence and a characterization of Nash equilibria in these games. (C) 2022 The Author(s). Published by Elsevier Inc.

We introduce a discrete-time search game, in which two players compete to find an object first. The object moves according to a time-varying Markov chain on finitely many states. The players know the Markov chain and the initial probability distribution of the object, but do not observe the current state of the object. The players are active in turns. The active player chooses a state, and this choice is observed by the other player. If the object is in the chosen state, this player wins and the game ends. Otherwise, the object moves according to the Markov chain and the game continues at the next period. We show that this game admits a value, and for any error-term $\veps>0$, each player has a pure (subgame-perfect) $\veps$-optimal strategy. Interestingly, a 0-optimal strategy does not always exist. The $\veps$-optimal strategies are robust in the sense that they are $2\veps$-optimal on all finite but sufficiently long horizons, and also $2\veps$-optimal in the discounted version of the game provided that the discount factor is close to 1. We derive results on the analytic and structural properties of the value and the $\veps$-optimal strategies. Moreover, we examine the performance of the finite truncation strategies, which are easy to calculate and to implement. We devote special attention to the important time-homogeneous case, where additional results hold., 34 pages, 2 figures

We present a general existence result for a type of equilibrium in normal-form games, which extends the concept of Nash equilibrium. We consider nonzero-sum normal-form games with an arbitrary number of players and arbitrary action spaces. We impose merely one condition: the payoff function of each player is bounded. We allow players to use finitely additive probability measures as mixed strategies. Since we do not assume any measurability conditions, for a given strategy profile the expected payoff is generally not uniquely defined, and integration theory only provides an upper bound, the upper integral, and a lower bound, the lower integral. A strategy profile is called a legitimate equilibrium if each player evaluates this profile by the upper integral, and each player evaluates all his possible deviations by the lower integral. We show that a legitimate equilibrium always exists. Our equilibrium concept and existence result are motivated by Vasquez (2017), who defines a conceptually related equilibrium notion, and shows its existence under the conditions of finitely many players, separable metric action spaces and bounded Borel measurable payoff functions. Our proof borrows several ideas from (Vasquez (2017)), but is more direct as it does not make use of countably additive representations of finitely additive measures by (Yosida and Hewitt (1952)).

Non-existence of a pure-strategy Nash equilibrium is a persistent problem in oligopoly models where sellers compete in prices and quantities. At the heart of this problem are unserved customers who are still willing to visit a less preferred supplier (i.e., so-called spillover demand). This note develops a consumer behavior model with spillover demand. Within this model there is a class of demand specifications for which a pure-strategy Nash equilibrium exists. This price-quantity equilibrium is shown to coincide with the Bertrand price equilibrium.

We consider decision problems with arbitrary action spaces, deterministic transitions, and infinite time horizon. In the usual setup when probability measures are countably additive, a general version of Kuhn’s theorem implies under fairly general conditions that for every mixed strategy of the decision maker there exists an equivalent behavior strategy. We examine to what extent this remains valid when probability measures are only assumed to be finitely additive. Under the classical approach of Dubins and Savage (2014), we prove the following statements: (1) If the action space is finite, every mixed strategy has an equivalent behavior strategy. (2) Even if the action space is infinite, at least one optimal mixed strategy has an equivalent behavior strategy. The approach by Dubins and Savage turns out to be essentially maximal: these two statements are no longer valid if we take any extension of their approach that considers all singleton plays.

The estate division problem considers the issue of dividing an estate when the sum of entitlements is larger than the estate. This paper studies the estate division problem from a noncooperative perspective. The integer claim game introduced by O'Neill (1982) and extended by Atlamaz et al. (2011) is generalized by specifying a sharing rule to divide every interval among the claimants. We show that for all problems for which the sum of entitlements is at most twice the estate the existence of a Nash equilibrium is guaranteed for a general class of sharing rules. Moreover, the corresponding set of equilibrium payoffs is independent of which sharing rule in the class is used. Well-known division rules that always assign a payoff vector in this set of equilibrium payoffs are the adjusted proportional rule, the random arrival rule and the Talmud rule.

We examine the guarantee levels of the players in a type of zero sum games. We show how these levels depend on the sigma algebras that are being employed on the players' action spaces. We further argue that guarantee levels may therefore also depend on set theoretic considerations. Additionally, we calculate the guarantee levels for finitely additive strategies. The solutions of catch games essentially differ among these setups. We find optimal strategies for almost all cases.

In second-price auctions with interdependent values, bidders do not necessarily have dominant strategies. Moreover, such auctions may have many equilibria. In order to rule out the less intuitive equilibria, we define the notion of distributional strictly perfect equilibrium (DSPE) for Bayesian games with infinite type and action spaces. This equilibrium is robust against arbitrary small perturbations of strategies. We apply DSPE to a class of symmetric second-price auctions with interdependent values. We show that the efficient equilibrium defined by Milgrom (1981) is a DSPE, while a class of less intuitive, inefficient, equilibria introduced by Birulin (2003) is not.

We consider a discrete-time dynamic search game in which a number of players compete to find an invisible object that is moving according to a time-varying Markov chain. We examine the subgame perfect equilibria of these games. The main result of the paper is that the set of subgame perfect equilibria is exactly the set of greedy strategy profiles, i.e. those strategy profiles in which the players always choose an action that maximizes their probability of immediately finding the object. We discuss various variations and extensions of the model., Comment: 14 pages, 0 figures

We study a variation of hotelling’s location model in which consumers choose between firms based on travel distances as well as the number of consumers visiting each firm. The model in which the network externality is the same for all firms was proposed by kohlberg (econ lett 11:211–216, 1983), who claims that no equilibrium exists for more than two firms. We assume the network effects to be linear and, in contrast to the claim in kohlberg (econ lett 11:211–216, 1983), derive a condition under which a subgame perfect nash equilibrium exists for four and six firms. Moreover, we show that for more than two firms the equilibrium locations of the firms are different from the equilibrium locations in hotelling’s location model. Our results suggest that a subgame perfect nash equilibrium exists if and only if the number of firms is even. We also provide examples of subgame perfect equilibria in which the network externality is different for some of the firms.

Using a network approach we provide a characterization of a separating equilibrium for standard signaling games where the sender's payoff function is quasi-linear. Given a strategy of the sender, we construct a network where the node set and the length between two nodes are the set of the sender's type and the difference of signaling costs, respectively. Construction of a separating equilibrium is then equivalent to constructing the length between two nodes in the network under the condition that the response of the receiver is a node potential. When the set of the sender's type is a real interval, shortest path lengths are antisymmetric and a node potential is unique up to a constant. A strategy of the sender in a separating equilibrium is characterized by some differential equation with a unique solution. Our results can be readily applied to a broad range of economic situations, such as for example the standard job market signaling model of Spence, a model not captured by earlier papers.

We consider two-player zero-sum games with infinite action spaces and bounded payoff functions. The players' strategies are finitely additive probability measures, called charges. Since a strategy profile does not always induce a unique expected payoff, we distinguish two extreme attitudes of players. A player is viewed as pessimistic if he always evaluates the range of possible expected payoffs by the worst one, and a player is viewed as optimistic if he always evaluates it by the best one. This approach results in a definition of a pessimistic and an optimistic guarantee level for each player. We provide an extensive analysis of the relation between these guarantee levels, and connect them to the classical guarantee levels, and to other known techniques to define expected payoffs, based on computation of double integrals. In addition, we also examine existence of optimal strategies with respect to these guarantee levels.

We consider a network pricing game on a parallel network with congestion effects in which link owners set tolls for travel so as to maximize profit. A central authority is able to regulate this competition by means of a (uniform) price cap. The first question we want to answer is how such a cap should be designed in order to minimize the total congestion. We provide an algorithm that finds an optimal price cap for networks with affine latency functions and a full support Wardrop equilibrium. Second, we consider the induced network performance at an optimal price cap. We show that for two link networks with affine latency functions, the congestion costs at the optimal price cap are at most $8/7$ times the optimal congestion costs. For more general latency functions, this bound goes up to 2 under the assumption that an uncapped Nash equilibrium exists. However, in general such an equilibrium need not exist and this can be used to show that optimal price caps can induce arbitrarily inefficient flows., 25 pages, 8 figures

We identify the maximal voting correspondence which is Condorcet Consistent and satisfies two participation conditions, namely the Top Property and the Bottom Property — thereby extending a result in Perez (2001). The former participation condition says that if an alternative is in the chosen set at a profile of rankings and a ranking is added with that alternative on top, then it remains to be a member of the chosen set. The latter says that if an alternative is not in the chosen set at a profile of rankings and a ranking is added with that alternative at bottom, then the alternative is again not in the chosen set. In particular, voting functions (single-valued voting correspondences) with these three properties select from this maximal correspondence, and we demonstrate several ways in which this can or cannot be done.

We study bargaining problems between one buyer and one seller for a single object when each set of valuations is finite. We derive a necessary and sufficient condition for interim-implementable trading rules, analogous to the result in Myerson and Satterthwaite (1983), and a necessary condition for mechanisms that are free from ex post regret on and off equilibrium path. Based on these two theorems, our contribution is three-fold. First, we illustrate the results by means of three examples. Second, in contrast to the continuous model, there always exist strictly positive probability mass functions such that ex post efficiency and interim-implementability is feasible. Third, we show the difference between interim- and ex post-implementation. In the class of threshold trading rules, interim-implementation only precludes inefficient trade, while ex post-implementation is equivalent to posted price implementation. (C) 2016 Elsevier B.V. All rights reserved.

A minimal diversity game is an n player strategic form game in which each player has m pure strategies at his disposal. The payoff to each player is always 1, unless all players select the same pure strategy, in which case, all players receive zero payoff. Such a game has a unique isolated completely mixed Nash equilibrium in which each player plays each strategy with equal probability, and a connected component of Nash equilibria consisting of those strategy profiles in which each player receives payoff 1. The Pareto superior component is shown to be asymptotically stable under a wide class of evolutionary dynamics, while the isolated equilibrium is not. In contrast, the isolated equilibrium is strategically stable, while the strategic stability of the Pareto-efficient component depends on the dimension of the component, and hence on the number of players, and the number of pure strategies.

We consider effectivity functions for finitely many players and alternatives. We assume that players have incomplete information—with private values—about the preferences of the other players. Our main result is the characterization of effectivity functions which have an ex post Nash consistent representation, i.e., there is a game form such that (i) the distribution of power among coalitions of players is the same as in the effectivity function and (ii) there is an ex post Nash equilibrium (in pure strategies) for any preference profile.

We show that Nash equilibrium components are universal for the collection of connected polyhedral sets. More precisely for every polyhedral set we construct a so-called binary game—a game where all players have two pure strategies and a common utility function with values either zero or one—whose success set (the set of strategy profiles where the maximal payoff of one is indeed achieved) is homeomorphic to the given polyhedral set. Since compact semi-algebraic sets can be triangulated, a similar result follows for the collection of connected compact semi-algebraic sets. We discuss implications of our results for the strategic stability of success sets, and use the results to construct a Nash component with index k for any fixed integer k .

For the allocation of heterogeneous items, it is known that the buyers-are-substitutes condition is necessary and sufficient to ensure that a pricing equilibrium can yield the same allocation and payments as the vcg mechanism. Furthermore, concavity of the corresponding transferable utility tu-game guarantees that this vcg outcome can also be achieved by an ascending price auction. We show that concavity, and hence the buyers-are-substitutes condition, holds for the tu-game of the assignment problem with general capacities. Therefore, the vcg mechanism is supported by a pricing equilibrium which can also be achieved by an ascending auction. We also show that the buyers-are-substitutes condition, and hence concavity, does not hold anymore for very natural and straightforward extensions of this problem. This shows that the necessity of the substitutes property is a considerable restriction on the applicability of the vcg mechanism.

In this paper, we analyze equilibria in competitive environments under constraints across players’ strategies. This means that the action taken by one player limits the possible choices of the other players. In this context, the usual approach to show existence of equilibrium, Kakutani’s fixed point theorem, cannot be applied directly. In particular, best replies against a given strategy profile may not be feasible. We devise a new fixed point correspondence to deal with the feasibility issue. Our main motivation to study this problem of co-dependency comes from the field of supply chain planning. A set of buyers is faced with external demand over a planning horizon, and to satisfy this demand they request inputs from a set of suppliers. Both suppliers and buyers face production capacities and they plan their own production in a decentralized manner. A well-known coordination scheme for this setting is the upstream approach where the plan of the buyers is used to decide the request to the suppliers. We show the existence of equilibria for a (shared) inventory cost minimization version of this coordination scheme in which a distribution center manages the inventory of the inputs. However, we illustrate with an example that the centralized solution is not, in general, an equilibrium, suggesting that regulation may be needed.

This paper studies the sales of a single indivisible object where bidders have continuous valuations. In Grigorieva et al. [14] it was shown that, in this setting, query auctions necessarily allocate inefficiently in equilibrium. In this paper we propose a new sequential auction, called the c-fraction auction. We show the existence of an ex-post equilibrium, called bluff equilibrium, in which bidders behave truthfully except for particular constellations of observed bids at which it is optimal to pretend a slightly higher valuation. We show c-fraction auctions guarantee approximate efficiency at any desired level of accuracy, independent of the number of bidders, when bidders choose to play the bluff equilibrium. We discuss the running time and the efficiency in the bluff equilibrium. We show that by changing the parameter c of the auction we can trade off efficiency against running time.

In an estate division problem an estate has to be divided among several players whose total entitlement to the estate exceeds its size. This paper extends the non-cooperative approach through a claim game, as initiated by o'neill (1982), by allowing players to put multiple claims on the same part of the estate, and by considering the case where individual entitlements may exceed the estate. A full characterization of the set of nash equilibria of the claim game is obtained both for restricted estate division problems, where individual entitlements do not exceed the estate, and for the general case. Variations on the claim game are considered, which result in proportional division in equilibrium.

In this note we give a survey of bisection auctions. Bisection auctions have been introduced in order to reduce the number of rounds and increase privacy of information in iterative implementations of Vickrey auctions. First, we present the case of discrete valuations. we discuss the strategic properties of this auction and recent results which show that--for 2 bidders--the auction dominates in a particular sense any other auction with respect to the number of bits revealed. For the case of continuous valuations we contrast its properties with the result that no practical query auction can achieve full efficiency in ex-post equilibrium.

A core concept is a solution concept on the class of balanced games that exclusively selects core allocations. We show that every continuous core concept that satisfies both the equal treatment property and a new property called independence of irrelevant core allocations (iic) necessarily selects egalitarian allocations. Iic requires that, if the core concept selects a certain core allocation for a given game, and this allocation is still a core allocation for a new game with a core that is contained in the core of the first game, then the core concept also chooses this allocation as the solution to the new game. When we replace the continuity requirement by a weak version of additivity we obtain an axiomatization of the egalitarian solution concept that assigns to each balanced game the core allocation minimizing the euclidean distance to the equal share allocation.

In this paper we present a new iterative auction, the bisection auction, that can be used for the sale of a single indivisible object. The bisection auction has fewer rounds than the classical english auction and causes less information to be revealed than the vickrey auction. Still, it preserves all characteristics the english auction shares with the vickrey auction: there exists an equilibrium in weakly dominant strategies in which everyone behaves truthfully, the object is allocated in accordance with efficiency requirements to the buyer who has the highest valuation, and the price paid by the winner of the object equals the second-highest valuation.

In this paper, an alternative definition of stable sets, defined by Mertens [Mertens, 1989. Stable equilibria – a reformulation. Part I. Definitions and basic properties. Mathematics of Operations Research 14, 575–625], is given where perturbations are interpreted as restrictions on the strategy space instead of perturbations of the payoffs. This alternative interpretation is then used to compute a special type of stable sets –called standard stable sets – in the context of bimatrix games, exclusively using linear optimization techniques and finite enumerations.

Hillas (1990) introduced a definition of strategic stability based on perturbations of the best reply correspondence that satisfies all of the requirements given by Kohlberg and Mertens (1986). Hillas et al. (2001) point out though that the proofs of the iterated dominance and forward induction properties were not correct. They also provide a proof of the IIS property, a stronger version of both iterated dominance and forward induction, using the results of that paper. In this note we provide a direct proof of the IIS property.

In this paper we study the geometrical properties of the set of Lorenz allocations of a transferable utility game. We provide procedures to compute a single Lorenz allocation, and even the entire set of Lorenz allocations, that rely solely on linear optimization techniques. These procedures only require a finite number of elementary operations and are therefore easy to implement.

In this paper the structure of the set of equilibria for two person multicriteria games is analysed. It turns out that the classical result for the set of equilibria for bimatrix games, that it is a finite union of polytopes, is only valid for multicriteria games if one of the players only has two pure strategies. A full polyhedral description of these polytopes can be derived when the player with an arbitrary number of pure strategies has one criterion.

In this paper we derive characterizations of egalitarian solutions on two subclasses of the class of balanced games. Firstly we show that the Dutta–Ray solution is the only solution that satisfies symmetry, independence of irrelevant core allocations, and continuity on the class of convex games. Secondly, together with the other two requirements, a strengthening of continuity to monotonicity in the value of the grand coalition turns out to be sufficient for the characterization of the lexicographically maximal solution on the class of large core games.

A central unanswered question in economic theory is that of price formation in disequilibrium. This paper lays the groundwork for a model that has been suggested as an answer to this question in, particularly, Arrow [Toward a theory of price adjustment, in: M. Abramovitz, et al. (Ed.), The Allocation of Economic Resources, Stanford University Press, Stanford, 1959], Fisher [Disequilibrium Foundations of Equilibrium Economics, Cambridge University Press, Cambridge, 1983] and Hahn [Information dynamics and equilibrium, in: F. Hahn (Ed.), The Economics of Missing Markets, Information, and Games, Clarendon Press, Oxford, 1989]. We consider sellers that monopolistically compete in prices but have incomplete information about the structure of the market they face. They each entertain a simple demand conjecture in which sales are perceived to depend on the own price only, and set prices to maximize expected profits. Prior beliefs on the parameters of conjectured demand are updated into posterior beliefs upon each observation of sales at proposed prices, using Bayes’ rule. The rational learning process, thus, constructed drives the price dynamics of the model. Its properties are analysed. Moreover, a sufficient condition is provided, relating objectively possible events and subjective beliefs, under which the price process is globally stable on a conjectural equilibrium for almost all objectively possible developments of history.

The classical bankruptcy problem is extended by assuming that there are multiple estates. In the finite estate case, the agents have homogeneous preferences per estate, which may differ across estates. In the more general infinite estate problem, players have arbitrary preferences over an interval of real numbers each of which is regarded as an estate. A strategic estate game is formulated in which each agent distributes his legal entitlement over the estates, resulting in individual claims per estate: each estate is then divided according to some allocation rule. The paper focuses on existence and on computational aspects of Nash equilibria in finite and infinite estate games, with some focus on the proportional allocation rule. For this rule, it also studies Pareto optimality and envy-freeness of equilibrium allocations. (C) 2014 Elsevier Inc. All rights reserved.

Kohlberg and Mertens (1986) argued that a solution concept to a game should be invariant under the addition or deletion of an equivalent strategy and not require the use of weakly dominated strategies. In this paper we study which of these requirements are satisfied by Kalai and Samet’s (1984) concepts of persistent equilibria and persistent retracts. While none of these concepts has all the invariance properties, we show that a slight rephrasing of the notion of a persistent retract leads to a notion satisfying them all.

By a player splitting we mean a mechanism that distributes the information sets of a player among so-called agents. A player splitting is called independent if each path in the game tree contains at most one agent of every player. Following Mertens (1989), a solution is said to have the player splitting property if, roughly speaking, the solution of an extensive form game does not change by applying independent player splittings. We show that Nash equilibria, perfect equilibria, Kohlberg-Mertens stable sets and Mertens stable sets have the player splitting property. An example is given to show that the proper equilibrium concept does not satisfy the player splitting property. Next, we give a definition of invariance under (general) player splittings which is an extension of the player splitting property to the situation where we also allow for dependent player splittings. We come to the conclusion that, for any given dependent player splitting, each of the above solutions is not invariant under this player splitting. The results are used to give several characterizations of the class of independent player splittings and the class of single appearance structures by means of invariance of solution concepts under player splittings.

The goal of this paper is twofold. Firstly a short proof of the unicity of the reduced form of a normal form game is provided, using a technique to reduce a game originally introduced by Mertens. Secondly a direct combinatorial-geometric interpretation of the reduced form is described. This description is then used to derive an algorithm for the calculation of the reduced form of a game.

A digital mechanism is defined as an iterative procedure in which bidders select an action, from a finite set, in each iteration. When bidders have continuous valuations and make strategic reports, we show that any ex post implementation of the Vickrey choice rule via such a mechanism needs infinitely many iterations for almost all realizations of the bidders’ valuations. Thus, when valuations are drawn from a continuous probability distribution, the Vickrey choice rule can only be used at the expense of a running time that is infinite with probability one. This infeasibility result even holds in the case of two bidders and the Vickrey choice rule only being required to be established with probability one. Establishing the efficient allocation when the n bidders’ report truthfully contrasts starkly to the previous setting: a bisection procedure has a finite running time almost always, and an expected number of reports are equal to 2 n . Using a Groves payment scheme rather than Vickrey’s second price payment scheme somewhat mitigates the problem. We provide an example mechanism with a Groves payment scheme, in which the running time of the mechanism in equilibrium is finite with probability 1 2 .

We investigate the algorithmic performance of Vickrey-Clarke-Groves mechanisms in the single item case. We provide a formal definition of a Vickrey algorithm for this framework, and give a number of examples of Vickrey algorithms. We consider three performance criteria, one corresponding to a Pareto criterion, one to worst-case analysis, and one related to first-order stochastic dominance. We show that Pareto best Vickrey algorithms do not exist and that worst-case analysis is of no use in discriminating between Vickrey algorithms. For the case of two bidders, we show that the bisection auction stochastically dominates all Vickrey algorithms. We extend our analysis to the study of weak Vickrey algorithms and winner determination algorithms. For the case of two bidders, we show that the One-Search algorithm stochastically dominates all column monotonic weak Vickrey algorithms and that a suitably adjusted version of the bisection algorithm, the WD bisection algorithm, stochastically dominates all winner determination algorithms. The WD bisection algorithm Pareto dominates all column monotonic winner determination algorithms in the n bidder case.

The model we use is a stochastic game, with transition probabilities between the states, depending on the actions performed in each state. The model's parameters can be chosen in such a way that a variety of economic situations can be analyzed. We derive results for the model with three states and five states. We are able to fully derive analytical results for the three states situation. If firms are very much focused on the short term, the only equilibrium is one in which both play the passive strategy (collusion). If the firms instead have high discount factors, the only remaining equilibrium is one in which they play aggressively. For average discount factors, a variety of (a)symmetric pure and mixed equilibria are found, in which monopolisation occurs with probability 1 in the long run. For the five states situation, we present numerical results using different parameter settings. We are able to provide equilibrium conditions for any kind of symmetric stationary strategy in these five states, including a collusive one in which firms always allow the market situation to go back to a symmetric state in which both firms play passively. As the model's parameters can be seen as indicators of market characteristics, predictions on the market circumstances under which firms will behave either collusively or competitively are made on the basis of our analytical and numerical results. Furthermore, we prove that if we extend the number of states to infinity, only the passive strategy equilibrium remains.

We show that, when bidders have continuous valuations, any ex post equilibrium in an ex post individually rational query auction can only be ex post efficient when the running time of the auction is infinite for almost all realizations of valuations of the bidders. We also show that this result applies to the general class of bisection auctions. In contrast we show that, when we allow for inefficient allocations with arbitrarily small probability, there is a query auction (to be more specific, a bisection auction) that attains this level of approximate efficiency in equilibrium, while additionally the running time of the auction in equilibrium is finite for all realizations of valuations.

We define an algebro-topological concept of essential map and we use it to prove several results in the theory of general equilibrium and nash equilibrium refinement.

Incomplete preferences over lotteries on a finite set of alternatives satisfying, besides independence and continuity, a property called bad outcome aversion are considered. These preferences are characterized in terms of their specific multi-expected utility representations (cf. Dubra et al., J Econ Theory, 115:118-133, 2004), and can be seen as generalized stochastic dominance preferences.

Two-person noncooperative games with finitely many pure strategies are considered, in which the players have linear orderings over sure outcomes but incomplete preferences over probability distributions resulting from mixed strategies. These probability distributions are evaluated according to t-degree stochastic dominance. A t-best reply is a strategy that induces a t-degree stochastically undominated distribution, and a t-equilibrium is a pair of t-best replies. The paper provides a characterization and an existence proof of t-equilibria in terms of representing utility functions, and shows that for large t behavior converges to a form of max–min play. Specifically, increased aversion to bad outcomes makes each player put all weight on a strategy that maximizes the worst outcome for the opponent, within the supports of the strategies in the limiting sequence of t-equilibria.