the Imperceptibility of mancur olson’s logic of collective action
Although game theorists have often misread it to show otherwise, Mancur Olson’s 1965 Logic of Collective Action does not assume or reinforce the Prisoner’s Dilemma analysis of small-sized, medium-sized, or large-scale groups.
According to Olson, the size of a group matters; according to game theory, the same PD logic applies independent of group size.24 Olson’s analysis hinges on the causal negligibility of actors’ impact on realizing collective goals, rather than the clear external cost imposed on others in the two-person and small-n PD game. By contrast, the public goods and tragedy of the commons argument presented by rational choice theorists concludes that a large-scale failure of collective action is the limiting case of a Prisoner’s Dilemma extended from two individuals to a very large number of people.25 According to this view, the two-person dilemma that rational self-interest necessarily leads to a mutually impoverishing Pareto suboptimal result will only be multiplied when each additional individual is added. The23 Schelling’s model, as in uniform multi-person PD games used for “resource dilemmas,” assumes that payoffs reflect salient objective and interpersonally transferable resources, “Hockey Helmets,” 1973, 386. Note that the PD model is similarly applied to the case of two or more actors in the development of a free market from duopoly and oligopoly; for discussion, see Tuck, Free Riding, 2008, 169-171.
24 This section takes up an insight developed by Richard Tuck in contrasting Olson’s Logic of Collective Action (Cambridge, MA: Harvard University Press, 1965) with the rational choicebased Prisoner’s Dilemma model routinely applied to public goods and the tragedy of the commons. See Tuck, Free Riding, 2008.
25 See Tuck for discussion, Free Riding, 2008, 28.
multi-person, large-scale collective action must therefore necessarily fail in accordance with this same logic. However, this is not Olson’s argument.The stakes of explaining how collective action fails, either because of the Prisoner’s Dilemma logic of predatory gain extended from two individuals to countless individuals on the one hand or because of the breakdown of instrumental agency in very large groups on the other, are high. The former reinforces the irrationality of cooperating at all levels of social organization. The latter argues that the solutions to global resource dilemmas may build on the tactics small and medium-sized groups have developed to collaborate effectively. If we accept that the two-person PD model accurately reflects two-person interactions in bargaining, exchange, as well as a state of nature, then it seems prima facie evident that it must extend to social relations with numerous individuals who make a binary choice over whether to contribute to or defect from an interaction with a tangible rewards PD payoff structure.
Reading Olson closely confirms that the Prisoner’s Dilemma analysis is not only largely irrelevant but also contradictory to his overarching thesis and that Olson himself acknowledges that he does not employ game theory.[550] Olson argues that small and medium-sized groups are able to cooperate successfully, but that even those actors who have tools transcending conventional instrumental rationality, including moral regard for others and solidarity, will be stymied by the collective action problem once their size crosses a threshold that renders any single individual’s actions causally inappreciable. Olson’s work fits comfortably with the conventional experience that small groups (such as cartels) and medium groups (such as trade unions) are sufficiently capable of voluntarily cooperating and must be curbed by legal restraint.[551]
Olson presented a mathematically formalized tripartite categorization of collective action problems: small, medium, and large.
Small groups may have a privileged ability to cooperate if their rate of profit exceeds the number of members. For these small groups, each individual is rewarded by a profit margin that multiples his contribution to such an extent that no individual is tempted to withhold his contribution.[552] Consider the following example of a “privileged group,” using Olson’s terminology, which is readily able to cooperate. Olson’s analysis incorporates the impact each individual’s contribution may have on the total resource supply and its yield. Suppose that it has ten members, each of whom contributes $1,000, and as a function of the total collective input of $10,000, each receives a share of $10,000 as a result of a tenfold yield. If a single defection, leaving only $9,000, were to produce only an eightfold yield, so that every individual would receive $7,200, then it is easy to see that the defector would have been better off by contributing. Had she contributed, her total profit would have been $1,800 higher. Such a payout structure clearly does not reflect the Prisoner’s Dilemma model, and thus Olson’s privileged group falls outside of the standard game theoretic treatment of the failure of cooperation.For mid-sized groups, the Prisoner’s Dilemma model seems apt, particularly if agents are only concerned with material rewards, not the processes by which rewards are generated. In this case, even though all members gain more from contributing than the cost of their contribution, each has a greater material incentive to free ride. Olson speaks to this reasoning: “Though all of the members of the group therefore have a common interest in obtaining this collective benefit, they have no common interest in paying the cost of providing that collective good. Each would prefer that the others pay the entire cost, and ordinarily would get any benefit provided whether he had born part of the cost or not.”[553] However, he neither acknowledges nor promotes this logic.
Instead, Olson assumes that mid-sized groups can achieve a cooperative outcome as a result of other-regarding preferences or normative commitments.[554] In a medium-sized group, an individual’s failure to cooperate is noticed by everyone because the total public good to be shared is appreciably diminished, as in the two-person PD. The individual who contemplates defecting then may reason that “he would be worse off when the collective good is not provided than when it was provided and he met part of the cost,” hence preferring mutual cooperation to lone defection.[555] The Prisoner’s Dilemma therefore does not pertain to Olson’s analysis of the failure of mid-sized groups. He doubts that this rationalization of free riding is sufficient because he understands that medium-sized collectives are able to effectively achieve joint objectives. He proposes that the ability of groups to collaborate successfully is a function of their size, and he focuses his inquiry on how the dynamics of participation change in relation to the ability of actors to influence the group’s collective outcome.[556]Olson homes in on how group size interacts with the relationship between the marginal cost of purchasing additional units of a public good and the marginal rate of production for that good. From a collective standpoint, it is easy to identify when each individual’s cost for a good, when collectively amalgamated, is less than the group’s ability to supply the good, in which case obtaining the good together is cost-effective from everyone’s perspective. Group size is important because the ratio of individual contribution to the total amount of public good provides each individual with a ratio between her own contributions and how much she receives in return as a function of this payment. For example, if an individual contributes $100 - leaving all others’ contributions and receipts at the same level - and the individual in question receives $110, or a 10 percent return, then it pays to contribute to the collective venture.
Olson highlights the conditions under which participation is directly in each individual’s interest:The marginal cost of additional units of the collective good must be shared in exactly the same proportion as the additional benefits. Only if this is done will each member find that his own marginal costs and benefits are equal at the same time that the total marginal cost equals the total or aggregate marginal benefit.[557]
Although an individual may contemplate or actually choose to free ride, this is not the point of Olson’s theoretical exploration. An individual may or may not contribute, and here neoclassical economics is agnostic as to whether or not an actor acts in accordance with narrow self-interest or other-regarding considerations that may encompass acting in accordance with shared norms. Olson is only concerned with determining at which point group size renders contributing to a joint venture economically unsound because individuals are no longer able to perceive any discernable impact from contributing. In fact, in referring to a Prisoner’s Dilemma-like situation, in which one agent may harm another by failing to contribute, Olson observes that a group’s ability to cooperate “depends upon whether any two or more members of the group have a perceptible interdependence, that is, on whether the contribution or lack of contribution of any one individual in the group will have a perceptible effect on the burden or benefit of any other individual or individuals in the group.”[558] The key attribute of the Prisoner’s Dilemma in all of its applications is that all individuals demonstrably affect others’ outcomes by their choice to cooperate or defect, hence their fear of being suckered. Not only is Olson’s treatment distinct from the free rider problem central to rational choice theory, but also his central thesis is that the failure of collective action arises from the eventual insignificance of any single actor’s ability to make a difference as the group’s size grows increasingly large.
This is the condition that obtains under perfect market competition in which no buyer or seller has sufficient presence to alter the trading price of a good.Olson’s Logic of Collective Action is widely misunderstood to be suggesting that groups face obstacles to achieving cooperative outcomes because of the Prisoner’s Dilemma logic.[559] A close reading of his argument, however, reveals that he did not doubt the power of mid-sized groups to collaborate as a function of their ability to act on resources beyond narrow self-interest.[560] Olson’s central point is that very large groups will fail to achieve collective goals because their size renders the causal efficacy of any single individual’s contributions impotent when evaluated from the perspective of any member of the group.
Three points loom large. First, it is astonishing that the Prisoner’s Dilemma bandwagon effect was so great that few theorists were aware of the sea change in thinking about the rationality of cooperation that attended this new fascination.[561] Second, not long ago, as Richard Tuck evidences in Free Riding, both empirical practice and elite theory accepted the rationality of collaboration, which has the structure of joint maximization.[562] Third, once pried free from the inauspicious Prisoner’s Dilemma logic, Olson’s analysis points to a unique challenge confronting large-scale global concerns, such as climate change, deforestation, and overfishing. Any attentive study of these arresting contemporary problems will face a very different puzzle if it turns out to be true that individuals are paralyzed by the sheer scale of the predicament. Olson beseeches us to be aware that “in a large group in which no single individual’s contribution makes a perceptible difference to the group as a whole or to the burden or benefit of any single member of the group, it is certain that a collective good will not be provided.”[563] The failure of large-scale collective action is more a function of the apathy or despair resulting from the insufficiency of instrumental agency than of strategic calculations for achieving self-gain despite others.