Rational Choice, Game Theory and Institutional Design. An Analysis of the Nested Game Model

Term Paper, 2008

18 Pages, Grade: A-



1. Introduction

2. Rational Choice, Game Theory and Institutional Design: An Analysis of the Nested Game Model
2.1 The Nested Games Framework: Overview of the Principle Argument

3. Modeling Strategic Interaction
3.1 Rationality as a Subset of Human Behaviour
3.2 Transplanting context into game theory
3.3 Institutions and Institutional Change

4. Universality, scientific value and empirical content

5. Conclusion

6. Appendix

7. Bibliography

1. Introduction

The theory of rational choice is central to modern political science and is used in many other disciplines such as sociology. Starting with Anthony Down’sAn Economic Theory of Democracy (1957),and Mancur Olson’sThe Logic of Collective Action (1965),in which authors used rational choice assumptions to explain the dilemmas of collective action and the behaviour of voters, political parties and interest groups in democracies, the theory of rational choice became one of the central theoretical concepts with which to explain political and social phenomena. George Tsebelis’Nested Games: Rational Choice in Comparative Politicsmakes theoretical and methodological contributions to the literature dealing with applications of rational choice assumptions to various subfields of comparative politics. Substantively, the book provides a systematic way of explaining the impact of contextual factors on human behaviour. Methodologically, it offers a formal approximation of strategic interactions under specific conditions.

Building on a fundamental assumption that human activity is goaloriented and selfmaximizing, Tsebelis develops a model of nested games defined as ‘the whole network of games in which an actor is involved’ to explain the puzzle of seemingly suboptimal or irrational choices (p.5). Game theory is the essential part of Tsebelis’ methodology. However, the major focus is on political context and institutions.

This paper will argue that Tsebelis’ nested games framework has extended the boundaries of traditional rational choice analysis by bringing together assumptions of rational choice and institutionalism. It would be reasonable to state that Tsebelis is rational choice institutionalist who builds his theory on two propositions: that the behaviour of individuals is statistically rational, and that it is also rulegoverned. This approach makes the ‘Nested Games’ volume particularly valuable for the body of literature attempting to depart from the pattern of simplistic explanations of human behaviour.

The paper proceeds as follows. In the next section, I briefly outline the nested games framework developed by Tsebelis, including principles upon which it is built. I proceed with the discussion of the complexity of the rational choice theory and the logic of institutional design. Finally, the paper concludes with the assessment of universality, scientific value and empirical content of Tsebelis’s model in comparative perspective.

2. Rational Choice, Game Theory and Institutional Design: An Analysis of the Nested Game Model

2.1 The Nested Games Framework: Overview of the Principle Argument

Tsebelis’ nested games model was developed to explain the cases of apparently suboptimal choices taken by the political actors. More precisely, the book analyses situations in which an actor confronted with a series of choices does not pick up the alternative that appears to be the best. These choices seem to be puzzling, since rational actions are the only possible way of human behaviour in the political arena.

Choices that do not appear to be the best an actor can do are puzzling because most observes assume (at least implicitly) that people try to behave in ways that maximize the achievement of their presumed goals, that is, they make optimal choices. The goal of this book is to provide a systematic, empirically accurate, and theoretically coherent account of apparently suboptimal choices (i).

The model developed by Tsebelis rests upon two fundamental assumptions, (1) that of rationality of individual actors and (2) importance of institutions (p.46). The author disputes the irrationality of political actors and argues that ‘if, with adequate information, an actor’s choices appear to be suboptimal, it is because the observer’s perspective is incomplete’ (p.7). The observer focuses attention on only one game, and fails to grasp the logic of the whole network of nested games. Thus, the action seems suboptimal in the case when observer is focused on the implications for the principle arenas. If implications for other arenas (games in multiple arenas) are considered, the actor’s choice is optimal. A whole network of games that the actor is involved in is called by Tsebelisnested games,and the arena that attracts the observer’s attention is calledthe principle arena.

Tsebelis distinguishes between two kinds of games that appear to be crucial for the political actor and the observer: games in the principal arena (that is obvious for the observer) and institutional design (the game about the rules of the game). In the case, the observer does not see that the actor is involved not only in one game in the principle arena, but also in a game about the rules of the game, the situation is coined asinstitutional design. Both kinds of games leave space for suboptimal choices. However, the case of institutional design is more complicated, as in this case, the game in the principle arena is nested inside a bigger game where the rules of the game are variable, and the set of options is larger.

The model developed by Tsebelis highlights an interesting interplay between the rational actors and institutional structures (rules of the game). In the game theory, the players face a series of options and choose one strategy that is most suitable at the moment. According to Tsebelis, the rules of the game (i.e. institutional structures) determine strategies. Each player has to make a decision/a choice that would be optimal for everybody and that would maximize each other’s payoffs. Given the fact that the rules and payoffs are fixed, it would be reasonable to suggest that the behaviour of political actors is predetermined. However, according to the author, the situation gets more and more complicated as long as the actor becomes involved in a set of interrelated games. Thus the actor innovates increasing the number of available options, and the ultimate choice will depend on both tactical and strategic considerations.

Institutional design case is described by the author as a process of consciousness planning that provides a systematic way of thinking about political institutions. Thus, similar to other scholars (North, Berman, others), Tsebelis believes that institutions are not only iterated constraints, but also objects of human activity (p.9 and North, pp.73 and 79):

The core of the theory consists of some very simple ideas: seemingly suboptimal choices indicate the presence of nested games (either games in multiple arenas or institutional design); in games in multiple arenas, events or strategies in one arena influence the way the game is played in another arena; institutional design refers to the choice of strategies inside existing rules.

It is important to note that the analysis of apparently suboptimal choices is conducted within the institutional framework of a particular society (p.40). Again, Tsebelis follows North insofar as he assumes that institutional context matters in predictable ways and that individual action is an optimal adaptation to an institutional environment (p.40). In sum, the theoretical foundation of the nested games model is a rational choice institutionalism.

The author argues that rational choice explanations of suboptimal choices are more powerful when institutional factors are taken into consideration. In this scheme, human behaviour is always a response to the variable set of institutional options. This means that the rationality depends on institutions. Therefore, it seems necessary to undertake analysis of the contextual and institutional framework within which actors operate.

3. Modeling Strategic Interaction

3.1 Rationality as a Subset of Human Behaviour

The agenda of the ‘Nested Games’ volume is defined by the rationality assumption. Tsebelis started his argument with the proposition that people are rational, that is, that they are goal oriented and choose the optimal means to achieve their goals. On the basic level, following economists, most rational choice scholars consider human behaviour as being purposeful and goaloriented in the sense that all individuals aim at maximizing their utility. Tsebelis’s book, however, departed from the bulk of comparative politics research by making the rationality assumption explicit, deriving its consequences and drawing upon it when formulating explanations (p.235). Tsebelis shows that such processes as learning, natural selection, heterogeneity of individuals, and statistical averaging can lead to the same outcomes as rationality. Whenever such conditions hold, rationality offers a good approximation of reality (p.237).

Given the fact that human behaviour cannot be always rational, most rational choice scholars use an `as if` thesis. According to Tsebelis, however, the `rationalityas-model` argument is not satisfactory for the reason that the assumptions of the theory are equated with the conclusions of the theory. Tsebelis proposes a different way of thinking. Instead of the concept of rationality as a model of human behaviour, he suggests the concept ofrationality as a subset of human behaviour(p.32):

The change of the perspective is important: I do not claim that rational choice can explain every phenomenon and that there is no room for other explanations, but I do claim that rational choice is a better approach to situations in which the actor`s identity and goals are established and the rules of interaction are precise and known to the interacting agents. As the actor`s goals become fuzzy, or as the rules of the interaction become more fluid and imprecise, rational choice explanations will become less applicable (p.32).

Tsebelis provides several arguments explaining why rationality is important in the game theoretic framework: (1) people prefer to conform with the behaviour prescribed by the theory, (2) people transform subjective probabilities into objective ones through a sequence of trial and error (learning process); (3) equilibrium does not change if one supposes that some people are less sophisticated than the others: the social outcome of choices made by a small portion of sophisticated individuals approximates the equilibrium that would prevail if everybody was sophisticated; (4) finally, statistical analysis proves that people are rational on average (though it may not hold for a particular individual). Unlike other scholars Tsebelis assumes that rational choice theory is a legitimate approximation of real processes, though the latter may have no positivist appeal.

Following this assumptions, Tsebelis operationalizes the notion of rationality distinguishing between two different sets of it: weak requirements of rationality that assure the internal coherence of preferences and beliefs and strong requirements of rationality that presuppose external validity, i.e. the correspondence of beliefs with reality (p.24). The weak requirements imply 1) the impossibility of contradictory beliefs and preferences, 2) the impossibility of intransitive preferences, 3) conformity to the axioms of probability calsulus, meaning that an actor has not only to have internally cohesive beliefs, but also be able to realistically evaluate the likelihood of fulfillment of his/her profitmaximizing goals (p.24).

The strong requirements of rationality require that the individual is realistic and responsive to the demands of the external world, that is 1) his strategies are optimal and conform to the prescriptions of the game theory; 2) probabilities approximate objective frequencies in equilibrium, and 3) beliefs approximate reality in equilibrium (p.28). Conformity with the prescriptions of the game theory means that an actor is operates within an equilibrium situation from which no one wants to deviate.

Further, Tsebelis makes several inferences from the rationality requirements. First, if an actor holds contradictory beliefs/preferences, anything can follow from a false antecedent (i.e. he can choose any option). Second, an actor who holds intransitive preferences finds himself in a less advantageous position (i.e. he may become victim of other rational actors’ choices). Thirdly, the maximization of expected utility is achieved through the multiplication of the utility derived from an event by the probability that this event will occur (p.26). All these inferences depend on equilibrium analysis. However, the latter presupposes another negative inference described as a failure of the rational choice logic to account for the paths each actor chooses to arrive at the prescribed equilibrium.

In addition to successfully operationalizing the concept of rationality, Tsebelis defines the conditions in which the application of the rationalchoice approach is justified on the grounds of its ability to predict the behaviour of political actors more accurately than other approaches to politics. Specifically, Tsebelis claims that the power of predictions is dependent on the presence of some external factors, including the existence of defined institutions and wellstructured situations, the ability of actors to learn from their previous experience, heterogeneity of individuals, the ability of the observer to solve the problem of natural selection, and the awareness of the observer that he might predict the behaviour not of a specific individual, but on an average individual (pp.33). Previously, Tsebelis stated that the power of predictions declines with the increase of fuzziness of individual beliefs/preferences. Thus, the predictive power of Tsebelis’ theory appears to be dependent on the presence of very well defined beliefs and very well defined institutions.

Rational choice cannot claim to explain all human behaviour. Only behaviour in situations covered by my five arguments can be the domain of reasonable rationalchoice applications (p.38).

In sum, Tsebelis’ concept of rationality is somewhat innovative. First, in his model, an abstract axiom of rationality as a model of human behavior is transformed into a situational contextdependent concept of rationality as a subset of human behavior. Second, not all individuals are assumed to be rational. Tsebelis asserts that only a small number of individuals appear to make sophisticated choices. However, that is enough to maintain equilibrium within a given institutional context.

I submit that political games (or most of them) structure the situation as well and that the study of political actors under the assumption of rationality is a legitimate approximation of realistic situations, motives, calculations and behaviour (p.33).

To summarize, [...] contrary to the dominant justification among rationalchoice sympathizers, which claims that the validity of the rational choice approach stems from good predictions, I claim that it is a legitimate approximation of real processes. People will approximate the rationalchoice prescriptions when the issues are important, and the degree of approximation will vary with information (p.38).

Tsebelis does not leave space for mistakes and irrationality in the political arena. Given the fact that people are statistically rational, mistakes may be attributed only to the observers rather than rational political actors.

There are two cases in which the actor does choose suboptimally: if he cannot choose rationally, or if he makes a mistake (p.7)... I do not think the first case is important in the study of political phenomena. The second case cannot occur often because if the actor recognizes that he was mistaken, he will presumably correct his behaviour (p.7).

Compared to North’s view of rationality, Tsebelis’ argument stays within the framework of the classical rational choice approach. Contrary to North who developed the concept of procedural rationality as opposed to instrumental rationality, arguing that procedural rationality has the ability to account for the incomplete and imperfect markets but also leading the researcher to the key issues of just what it is that makes markets imperfect (North, p.108), Tsebelis does not incorporate into his model any historical and cultural explanations asserting that the latter appear to be useless when applied to social reality: ‘My position deviates from an important body of literature that tries to explain essential aspects of human activity in terms of evolutionary principles’ (p.101).

Thus, in contrast with North’s concept of rationality that is historical and dynamic, Tsebelis’ rationality is situationally dynamic, yet historically static. Two types of dynamism discussed here are situational dynamism and evolutionary dynamism. North’s model is more responsive to the complexity of human reality, pathdependency and psychological processes. Tsebelis’s model is more traditional for the rational choice literature in a sense that it lacks dynamism, being based on normative rather than positive assumptions.


Excerpt out of 18 pages


Rational Choice, Game Theory and Institutional Design. An Analysis of the Nested Game Model
University of Toronto
Catalog Number
ISBN (eBook)
ISBN (Book)
File size
539 KB
rational, choice, game, theory, institutional, design, analysis, nested, model
Quote paper
Svetlana Inkina (Author), 2008, Rational Choice, Game Theory and Institutional Design. An Analysis of the Nested Game Model, Munich, GRIN Verlag, https://www.grin.com/document/356305


  • No comments yet.
Read the ebook
Title: Rational Choice, Game Theory and Institutional Design. An Analysis of the Nested Game Model

Upload papers

Your term paper / thesis:

- Publication as eBook and book
- High royalties for the sales
- Completely free - with ISBN
- It only takes five minutes
- Every paper finds readers

Publish now - it's free