COLLECTIVE ACTION AND EXCHANGE
A Game-Theoretic Approach to Contemporary Political Economy
By WILLIAM D. FERGUSON
Stanford University Press Copyright © 2013 Board of Trustees of the Leland Stanford Junior University
All rights reserved.
COLLECTIVE-ACTION PROBLEMS AND INNOVATIVE THEORY
1.1. COLLECTIVE-ACTION PROBLEMS, POLITICAL ECONOMY, AND EXCHANGE
Collective-action problems (CAPs) arise from any deviation between unfettered pursuit of individual goals (typically, self-interest) and the perceived well-being of at least a portion of some group&mash;be it a nation, region, community, firm, or sports club. Because CAPs represent the archetypal dilemma of strategic interaction among purposeful agents, and because exchange underlies economic and social development, the relationships between collective action and exchange serve as the unifying principle of this book. Collective-action problems reveal the logic behind multiple market failures, notably those associated with public goods, externalities, common-pool resources, and problems of coordination and enforcement. Furthermore, market success depends on resolving CAPs, especially those related to coordination and enforcement. Thus CAPs and the potential for their resolution lie at the foundations of political economy.
There are two basic types of collective-action problems: first- and second-order CAPs. First-order CAPs signify free-rider problems that are associated with providing public goods, reducing the production of negative externalities, increasing the production of positive externalities, and limiting the use of common-pool resources to sustainable levels. We define all of these terms broadly. For example, an institution is a public good, as is establishing a sense of trust that facilitates exchange within a community. Second-order CAPs are problems of orchestrating the coordination and /or enforcement needed to render agreements for resolving first-order CAPs credible. Indeed, coordination and enforcement are themselves types of public goods&mash;namely, public goods that lend credibility to agreements.
Collective-action problems matter for political economy not only because they represent key linkages between politics and market exchange, but also because they indicate the core rationale for economic and social policy. Most policies seek to resolve or ameliorate CAPs (with varying degrees of success or failure). More fundamentally, the existence of first-order CAPs signifies market or group failure, and successful exchange requires some prior resolution of second-order CAPs. In fact, resolution of second-order CAPs precedes reliable definition and enforcement of the property rights that underlie market exchange. Furthermore, such resolution facilitates the formation of institutional environments, as well as the associated mutual understandings and trust that allow multiple forms of economic, political, and social exchange to occur at all. Resolution of collective-action problems thus underlies all substantive economic, political, and social development.
1.2. ANALYTICAL APPROACHES TO COLLECTIVE ACTION AND EXCHANGE
Much of the motivation for this book arises from relatively recent developments in game theory, economics, and policy theory. These developments both expand and refocus political-economic inquiry. They permit the formalization and modeling of principles that previously had been too complex or considered too indeterminate&mash;and were thus relegated to the sidelines of analysis until recently. Our summary of these developments serves three purposes: it offers intellectual context, points to general methodology, and outlines important concepts that will appear in the remainder of this text.
Our discussion now proceeds to thumbnail sketches of eight critical realms of newly emerging theory: game theory, social network analysis, information economics, social preference theory, rationality theory, institutional and governance theory, policy theory, and spatial-location theory. This section concludes by merging these ideas into a unified conception of game-theoretic political economy that summarizes the book's core assertions.
If CAPs represent the archetypal strategic dilemma for political economy, then game theory offers its methodological bedrock. Formally, game theory facilitates modeling strategic interaction among two or more agents. Strategic interaction occurs whenever agents share strategic interdependence&mash;meaning that their actions affect outcomes for others and that agents typically understand such interdependence. In an oligopoly, for instance, one firm's production can affect the quantities produced and prices set by other firms. Similarly, a decision by a contractor to hide mistakes can affect its client's profits. In political campaigns, one candidate's decision to advertise can alter prospects for opponents. Likewise, a decision by one student to invite another to a high-school prom may affect the happiness of would-be partners. These types of strategic interaction permeate this book's game-theoretic discussion of collective action and exchange.
By contrast, the constrained-maximization approach of traditional economics and rational choice theory often fails to address strategic interaction. Herein lies a core distinction between recent political-economy approaches to exchange and more conventional approaches based on pure competition and similar models. In pure competition, an agent's decisions focus only on self and environment. Firms and consumers individually choose their quantities produced or purchased&mash;and respond to such external variables as market prices&mash;without reference to actions of other participants. Others' actions merely become indistinguishable elements of an environment (a field) that constrains individual activity and affects outcomes. Although the sum of others' actions can (as part of the field) influence prices, no individual consumer or firm can have such impact. Given this lack of influence, agents can safely disregard others' strategies in their decision calculus.
Strategic interaction is more complicated and intricate than constrained maximization because other parties matter. For example, an oligopolistic firm considers the expected behavior of its (few) competitors before deciding on price or output&mash;while understanding that its competitors share a similar strategic perspective. Such strategic interaction constitutes a game, not just an individual decision. Indeed, the constrained-maximization perspective of the traditional approach applies only to exceptional cases. Furthermore, even in competitive markets, game theory can represent transactions that involve any type of commitment, such as a contract, or any private information&mash;such as knowing far more about one's qualities, motivation, or behavior than others do (Dixit, Skeath, and Reiley 2009, 19). Conditions of this sort are ubiquitous in complex economies. Even though market competition can undermine many possible strategies and force players to reveal information, economic choices remain profoundly strategic because outcomes depend on numerous interactions among participants.
This text draws upon four broad types of game theory: classical, evolutionary, behavioral, and epistemic. In classical game theory, agents share some level of mutual awareness of their interdependence. Chosen actions reflect best-response strategies&mash;that is, action plans that specify responses to every conceivable contingency&mash;most notably, envisioned combinations of others' actions. Game outcomes reflect the conditions (or states of the world) that follow from a game's enacted combinations of strategies. Examples of game outcomes include the price of an item, the distribution of resources, and the election of a candidate. Each possible outcome generates payoffs to each involved player, where payoffs capture the net utility gain (or subjective valuation) from that outcome. Utility returns may be either material (as in money) or social (as in status). Classical game theory thus mimics traditional economic reasoning by utilizing best-response maximization; yet it embeds such responses in strategic domains. Applications of these principles, moreover, extend beyond economics to any strategic interaction.
Evolutionary game theory offers an alternative perspective&mash;one in which strategies consist of programmed or inherited "phenotypes." In social scientific applications, inheritance reflects prior education and other forms of cultural (rather than genetic) transmission. Individuals&mash;and, by extension, populations&mash;inherit behavioral orientations, strategies, or established practices in various combinations. Thus some individuals are naturally aggressive and others are shy; some speak Chinese, others English. As in classical game theory, specific combinations of strategies operating within particular contexts generate outcomes with payoffs, but evolutionary payoffs do not indicate players' utility, but instead show the fitness or reproductive viability of their employed strategies. Strategies that earn high fitness payoffs reproduce or transmit abundantly; low-payoff strategies (failed practices) fade away over time. Evolutionary game theory thus facilitates modeling of learning processes and other forms of selective social adaptation.
Behavioral game theory uses experimentation in contexts guided by game-theoretic precepts (e.g., prisoners' dilemma scenarios) to explore relationships between context, perceptions, and agents' actual behavior. Experimental findings may then be incorporated into the other three types of game-theoretic modeling&mash;for example, by informing representations of payoffs. Finally, epistemic game theory focuses on the cognitive dimensions of strategic interactions. Using classical reasoning with emphasis on the extent and limits of agents' prior knowledge and expectations, it illustrates how certain shared understandings can guide or correlate strategic decisions among multiple players. Epistemic game theory is especially useful for modeling the impact of institutions on strategic behavior.
In short, game theory offers an extraordinarily flexible and widely applicable set of modeling techniques. Its use informs not only political science and economics but also biology, psychology, sociology, anthropology, business strategizing, and policymaking. In political economy, game theory enables us to model multiple intricate micro-level transactions within small groups, as well as macro-level patterns among nations or across populations. We may then specify the microfoundations of exchange as strategic social interactions that arise within or among groups of purposeful agents, rather than as a mere summation of independent acts of maximization. In so doing, game theory fosters a truly social scientific modeling framework for political economy. Agents' choices and even their preferences may depend on and respond to anticipated activity or reactions from other parties&mash;all conditioned by social and institutional contexts. Even though game theory can incorporate such social influences, it also posits individuals (or unified organizations, such as firms) as critical decision-making or response units as it models goal-oriented and adaptive behavior. Game theory thus retains the core efficient rigor of economic or rational-choice logic&mash;but vastly enhances its domain and profoundly alters its implications. Modeling multiple facets of political economy becomes feasible.
Game-theoretic modeling permeates this book. As we shall see, the simple two-player prisoners' dilemma illustrates the core idea of a collective-action problem, and the analysis builds from there. Our approach to game theory mixes intuitive logic with relatively accessible mathematics. We use models to illustrate specific assertions or perspectives on political-economic interactions at multiple levels of analysis. Because it is never possible to model everything at once, the choice of modeling technique (e.g., the relevant type of game) will depend upon our analytical purpose and its accompanying questions. Much of the discussion in this text thus complements game-theoretic modeling with descriptions of problems, discussion of context, examples, intuitive arguments, relevant principles, and more general theory.
Social Network Analysis
Social network analysis offers a related modeling approach that focuses on connections among multiple agents. Social networks are configurations of relationships or communication pathways that, somehow, connect people. Virtually all economic, political, and social exchanges operate within social networks. Organizations are networks. A CEO, for instance, occupies a specific position within a complex corporate network. Political parties are networks, as are social clubs. Markets rely on, arise from, and sometimes embody networks of exchange among various buyers and sellers. Families and communities are networks. Network analysis encompasses all such entities. It allows specification and examination of the pathways that transmit information, ideas, influence, goods, services, and the like among specific groupings or vast populations of agents. Network analysis facilitates analyzing the origins and development of such pathways, the associated patterns of transmission among agents, possible micro- or macro-level impacts of such transmissions, and how the positioning of agents within networks affects the content or influence of their transmissions. Social network analysis thus provides another method for investigating how social context shapes the evolution, operation, and impact of social, political, and economic exchange.
Information economics explicitly addresses implications of costly, incomplete, and asymmetric information on economic behavior. According to Joseph Stiglitz (2002), information economics alters the prevailing paradigm for economic theory. Not only does asymmetric information introduce a new set of strategic variables, it implies that contracts and other forms of agreement may not be fully enforceable. Second-order CAPs follow. The ensuing enforcement problems, in turn, indicate that labor and capital markets routinely fail to clear. Furthermore, such markets face unavoidable (though often manageable) efficiency losses&mash;as well as exercises of power within exchange relationships. A political dimension of exchange thus emerges.
The game-theoretic distinction between imperfect and incomplete information speaks to the importance of information economics. Imperfect information connotes an equally distributed lack of knowledge regarding states of the environment or an equally shared inability to observe the actions of others. Such imperfect information underlies traditional risk analysis. Although agents may not know the exact values of certain variables, they do know the underlying probability distributions and can therefore maximize on the basis of expected values. Incomplete information poses more difficult problems. It connotes any information asymmetry among participants concerning conditions in their environment or any lack of knowledge among them concerning the characteristics, motivations (payoffs), or strategies available to other participants in their strategic interactions.
Information economics stresses that asymmetry permits strategic manipulation of information. In this regard, the concept of adverse selection (Akerlof 1970) reflects the intuitive notion that, prior to signing a contract or conducting exchange, sellers usually know more about the quality of exchangeable goods or services than do buyers. Such asymmetry can lead to no exchange (a complete market failure) if buyers lack confidence in quality, sellers lack confidence in marketability, or both. Inefficient exchanges of lower-than-expected quality are also possible. More fundamentally, adverse selection implies imperfect definition or understanding of the de facto property rights to be exchanged because rights related to quality are unclear.
Similarly, principal-agent models indicate problems of post-contractual asymmetric information (moral hazard problems). In these models, a principal contracts with an agent to perform certain services. Because providing service is costly to the agent, their interests differ. Additionally, the principal cannot fully observe or verify the agent's actions. The principal then faces an enforcement problem because the agent has an incentive to provide less effort, diligence, or information than specified by the contract. Resolution requires devoting resources to instituting an internal enforcement mechanism&mash;that is, one that operates within the exchange process itself. For example, bosses may fire employees who appear to perform less diligently than expected when they were hired. The related literature on mechanism design addresses how social mechanisms that specify incentives may or may not elicit accurate information and rule-abiding behavior. When such mechanisms successfully align the incentives of principals and agents, they can mitigate underlying conflicts, enhancing prospects for resolving their information-based second-order CAPs.
Excerpted from COLLECTIVE ACTION AND EXCHANGE by WILLIAM D. FERGUSON. Copyright © 2013 by Board of Trustees of the Leland Stanford Junior University. Excerpted by permission of Stanford University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.