COLLECTIVE ACTION THEORY
186: “the theory of collective action is the central subject of the political science”
“The term “social dilemma” refers to a setting in which individuals choose actions in a interdependent situation….The reason that such situations are dilemmas is that at least one outcome yields higher returns for all participants, but rational participants making independent choices are not predicted to achieve this outcome…..Since the (187) suboptimal joint outcome is an equilibrium no one is independently motivated to change their choice, given the predicted choices of all others”.
187: outline of this chapter: First “literature positing a host of structural variables presumed to affect the likelihood of individuals achieving collective action to overcome social dilemmas” Second, “how a theory of boundedly rational norm-based human behaviour is better foundation for explaining collective action than a model of maximizing material pay-offs to self. If one has posit that individuals can use reciprocity and reputations to build trust in dilemma situations then one can begin to explain both successful an unsuccessful efforts to overcome social dilemmas through collective action [“impossible to explain success in this if “we continue to treat the model of rationality that has proven successful in explaining behaviour and outcomes in competitive market settings as a universal theory of human behaviour”]. Third, will examine the linkage between the structural measures and the core individual relationships….
188: Structural variables that do not essentially depend on a situation being repeated:
a) Number of participants involved,
b) Whether benefits are subtractive or fully shared (public goods vs cpr).
c) Heterogeneity of participants,
d) Face-to-face communication,
e) shape of production function
Situations where repetition of situation makes possible the impact of additional structural variables
f) information about past actions,
g) how individuals are linked
h) whether individuals can enter or exit voluntarily
188: Number of participants
M Olson argued that as the “size of a group increased, the probability of a group achieving a public good decreased …..[Mancur Olson hypothesis] as group size increases the noticeability of any single input to the provision of a public good declines…..Second, coming to an internal agreement about coordinated strategies in larger groups involves higher transaction costs”.
189: some theorists posit the opposite prediction: Bates and Shepsle (1995) trying to understand the age grade organization in Africa “developed a formal model of a a three period overlapping generations public good game. A corollary of this model generates a prediction that the provision of public goods is positively correlated with group size since the more individuals in an age set,the easier it is to produce any particular level of a public good…….”Thus how size might affect the likelihood of cooperation depends on how other structural variables are affected by the size of a group”.(subry mio)
189: Olson included dilemmas where it was difficult to exclude potential beneficiaries whether they had contributed or not. “Unfortunately, Olson’s analysis confounded situations where the consumption of benefits by one individual subtracted benefits from others with situations where consumption was non-subtractive in nature( characterized as having full jointness of supply (seee ostrom and ostom,1999). Marwell and Oiver(1993) conclude that when “ a good has pure jointness of supply, group size has a positive effect on the probability that it will be provided”.
189: CPR are subtractable by nature have in common with public goods that have problems of free-riding, but they also include the problems of overharvesting and crowding. In a CPR environment an increase in the number of participants holding other variables constant, is negatively related to achieving social benefits. (190) Thus in a CPR environment, whether size has a positive impact ,a negative impact, or any impact, is dependent upon how other variables are affected by a change in the number of participants.”
190: The heterogeneity of participants
“Olson argued that if there were one or a few individuals who had much stronger interests in achieving a public good…the probability of a group achieving a public good increased even though the good was still likely to be underprovided {See footnote 5 “hegemonic stability theory (keohane 1984,kindelberger1973)posits that heterogeneity promotes cooperation because large actors are endowed with more resources (including the power to coerce others)…..the theory predicts that when there are a limited number of larger states dominating international relations the collective good of peace is more likely to be provided”. Others have speculated that heterogeneity in assets, information, and payoffs are negatively related to gaining a cooperators’ dividend due principally to increased transaction costs and the conflict that would exist over the distribution of benefits and costs to be borne…..The impact of heterogeneity on levels of collective action achieved frequently interacts with the shape of the production function for a good and thus will be discussed further below.”
190: Face to face communication: “Given that non-cooperative game theory predicts that communication will make no difference in the outcome of social dilemmas, the repeated findings of a strong positive effect that communication has on the outcomes of collective action experiments is a major theoretical puzzle….(191) In other words communication is used for “moral suasion”… Kerr and Kaufmann-Gilliland(1994) conclude that communication in general helps a group gain a sense of “solidarity” and that face-to-face communication enhances the likelihood that individuals will keep their promises to cooperate….When they are in a repeated situation, they use the opportunity for communication to discuss deviations from promises made in a highly critical and moralistic tone.”
191: The shape of the production function
“The production function that relates individual actions to group outcomes may take any of a wide diversity of forms (se figure 8.1 p. 192). Russell hardin (1976) was among the first to argue that when the shape of the production function for a public good was a step function, solving social dilemmas would be facilitated since no good would be provided if participants did not gain sufficient inputs to equal or exceed the provision point (k). Until the benefit is actually produced, it is not possible to “free ride” on the contribution of others….This type of production function may create an “assurance problem” rather than a strict social dilemma. For those who perceive their contribution as critical, not contributing is no longer the unique Nash equilibrium. Closely related to this attribute of the production function itself are the sharing formulas that may be developed by participants to make each person of the entire group, or a designated minimal contributing group, feel that their contribution is critical (van de kragt et al,1983). By agreeing that each person will contribute a set proportion of what is believed to be the total cost of obtaining a good, the individuals in such a minimal contributing set face a choice between contributing and receiving the benefit (assuming others in the minimal contributing set also contribute), or not contributing and receiving nothing.”
192: Marwell and Oliver (1993) focus on non-linear functions and distinguish between third-order production functions that are decelerating and those that are accelerating. In the decelerating case (d in fig 8.1) while every contribution increases the total benefits that a group receives, marginal returns decrease as more and more individuals contribute. When contributions are made sequentially, the initial contributions have far more impact than later contributions. The example they use to illustrate such a production function is calling about a pothole in the neighbourhood…….With an accelerating production function initial contributions make small increments and later contributions yield progressively greater benefits. “Accelerating production functions are characterized by positive interdependence: each contribution makes the next one more worthwhile and thus, more likely” (Marwell and Oliver,1993,63). Protest activities where mass actions are needed in order to gain a positive response involve accelerating functions…(193) The theoretical predictions depend sensitively on the particular shape of the production function, on whether all participants are symmetric or have different levels of assets, on the sequence in which individuals contribute and on the information generated by each action…. The key is whether the initial contributions are made and this is somewhat less likely with homogeneous group than with a heterogeneous group who may have some members with high levels of interest and who would be more interested in contributing the initial inputs.”
193: Repetition of interactions: Information about past actions
“In families and small neighbourhoods, where interactions are repeated, reputations for always voting or always contributing to political campaigns can be built over time and group members can build up a level of trust about other participants (seabright 1993)….Various ways of monitoring the actions of participants increase or decrease the availability and accuracy of the information that individuals have concerning the particular actions of known individuals (or types of players) in the past (Janssen 2004)
194: How individuals are linked
….have stressed the importance of how individuals may or may not be linked in a network when confronting various types of social dilemmas (Granovetter,1973; Cook and Hardin,2001). They have posited that individuals who are linked in a network where A contributes resources to B, B contributes…to C and C contributes resources to A – or any similar unidirectional linking- are more likely to contribute to each other’s welfare than individuals whose resource contribution goes to a generalized pool from which all individuals obtain benefits. The reason given for this expectation is that individuals in an undifferentiated group setting can expect to free ride for a longer period of time without reducing their own benefits that when contributions have to be delivered to someone in the chain of relationships in order for benefits to eventually come to them…. Creating a particular type of network may change the structure of the game from an n-person PD to an Assurance Game (Yamagishi and Cook,1993).
194: The possibility of choosing whether to play or not (Entry and exit)
“[when you have previous information about the players] individuals will choose partners so as to increase the frequency with which cooperative outcomes are achieved. This gives individuals a third choice in a social dilemma game. Besides deciding whether to cooperate, then can decide whether to “opt out”……The symbols are used by participants to remember which type of player cooperated in the past….Given the capacity to recognize trustworthiness in others and the capacity to withdraw from playing a game at all, cooperation levels rise over time and reach relatively high levels in populations composed of 100 players. With 1,000 players cooperation levels are lower unless the number of symbols that can be used recognize trustworthy is increased.
195 TOWARDS A MORE GENERAL THEORY OF HUMAN BEHAVIOR
“To have one theory –rational choice theory- that explains how individuals achieve close to optimal outcomes in markets, but fails to explain voting or voluntary contributions to political campaigns, is not a satisfactory state of knowledge in the social sciences. Simply assuming that individuals are successfully socialized into seeking better group outcomes does not explain the obvious fact that groups often fail to obtain jointly beneficial outcomes (Dietz et al,2003)”
“In other words, the context within which individuals face social dilemmas is more important in explaining levels of collective action than relying on a single model of rational behaviour as used in the classical non-cooperative game theory…. As Alchian (1950) demonstrated long ago competitive markets eliminate businesses that do not maximize profits. Further markets generate limited, but sufficient, statistics needed to maximize profits.”
“A broader theory of human behaviour view humans as adaptative creatures (jones,2001) who attempt to do as well as they can given the constraints of the situations in which they find themselves (or the ones that they seek out) (Simon,1955,1957,1999)…They adopt both short term and long term perspectives dependent on the structure of opportunities they face. Multiple models are consistent with a theory of boundedly rational human behaviour.”
196: Heuristics and norms
“In most everyday situations individuals tend to use heuristics –rule of thumb- that they have learned over time regarding responses that tend to give them good (but not necessarily optimal) outcomes in particular kinds of situations. In frequently encountered, repetitive situations, individuals learn better and better heuristics that are tailored to a particular situation. With repetition and sufficiently large stakes, individuals may learn heuristics that approach best-response strategies (Gigerenzer and Selten, 2001)..From a simulation (See Morikawa et al 1995)..they predict that the heuristic is of most value to individuals who are moderately disposed to cooperate rather than holding either of the extremes. Their simulation also generates the prediction that the heuristic will be most valuable when social dilemmas occur among those in close proximity and that the probability of there being some very cooperative groups of agents increases with the size of the population”
In addition to learning heuristics individuals also learn norms. “By norms I mean that the individual attaches an internal valuation –positive or negative- to taking particular types of action. Analytically, individuals can be thought of as learning norms of behaviour that are relatively general and fit a wide diversity of particular situations…. [delta parameter that is added to or subtracted from the objective costs of an action or an outcome]….The strength of the commitment made by an individual to take particular types of future actions (telling the truth, keeping promises) is reflected in the size of the delta parameter. After experiencing repeated benefits from their own and from other people’s cooperative actions, individuals may resolve that they should always initiate cooperation in the future”.
197: “ James Cox and colleagues posit that individual behaviour in a particular setting is affected by an individual’s initial emotional or normative state and then by direct experience with others in a specific setting…”
“Fairness is also one of the norms used by individuals in social dilemma settings.The maximal net return to a group may be obtained in a manner that is perceived to be fair or unfair by those involved –using the general concept that
197-198: Contingent strategies and norms of reciprocity
Explaining better social dilemmas “if one assumes that individuals enter situations with an initial probability of using reciprocity based on a calculated strategy that reciprocity leads to higher outcomes or based on a norm that this is how one should behave....In either case, individuals learn to use reciprocity based on their prior training and experience. The more benefits that they have received in the past from other reciprocators, the higher their own initial inclinations. The more they have faced retribution the less likely they estimate that free riding is an attractive option”.
Tit-for-tat (cooperate first, and then do whatever the others did on the last round). “Axelrod and Hamilton(1981) and Axelrod(1984) have shown that when individuals are grouped...and when the expected number of interactions is sufficiently large, reciprocating strategies such as tit for tat can successfully invade populations composed of individuals following an all-defect strategy..... Reciprocating strategies continue to limit what individuals can do who face others who do not cooperate. The only way of “punishing” defection is to defect oneself, which may lock participants into the deficient equilibrium......Hirshleifer and Rasmusen (1989) fid that the strategy of cooperate and then punish any defectors will increase to a polymorphic equilibrium in large populations if 1)defectors respond to punishment by a single player by cooperating thereafter and 2) the long-run benefits to the punisher exceed the costs they pay for punishing someone else.”
199: “Several of the heuristics or strategies posited to help individuals gain larger cooperators’ dividends depend upon the willingness of participants to use retribution to at least some degree. In tit-for-tat for example,, an individual most be willing to “punish” a player who defected on the last round by defecting on the current round....the grim trigger is a strategy that cooperates with others until someone defects, and then defects the rest of the rounds (Fudenberg and Maskin,1986). In repeated games where substantial joint benefits are to be gained from mutual cooperation, the threat of the grim trigger is posited to encourage everyone to cooperate. A small error on the part of one player or exogenous noise in the payoff function, however, makes this strategy a very dangerous one to use in large environments where the cooperators’ dividend is substantial.”
2000: “Further, the evolution of preferences that include benefits to others is more likely to emerge in populations where individuals are not anonymous and can use symbols to identify their type (Ahn,Janssen and Ostrom, 2004)
2000: The core relationships: reputation, trust and reciprocity as they affect cooperation
“..at the core of an evolving theoretical explanation of successful and unsuccessful collective action are the links between the trust that one participant (Pi has in the others( Pj....Pn) involved in a collective action situation, the investment others make in trustworthy reputations, and the probability of all participants using reciprocity norms (fig 8.2)....(201) thus reputations for being trustworthy, levels of trust, and reciprocity are positively reinforcing. This also means that a decrease in any one of these can generate a downward cascade leading to little or no cooperation”.
2000-2001: LINKING STRUCTURAL VARIABLES TO THE CORE RELATIONSHIPS
“...the task we now face is how to link external variables to an inner core of individual-level variables –reputation, trust and reciprocity- as this in turn affect levels of cooperation and net benefits achieved......one can confidently posit that in a small, homogeneous group interacting in a face-to-face meeting to discuss producing a public good with an accelerating production function, the costs of coming to an agreement will tend to be low and the probability that individuals keep their promises will be high..... “ “The combined effect of the structural variables in this example on reputation, trust and reciprocity is likely to overcome short-term material benefits that individual participants are tempted to pursue. In a different context –a large heterogeneous group with no communication and no information about past trustworthiness who jointly use common-pool resource- individuals will tend to pursue short-term material benefits and potentially destroy the resource.”
“Thus, using a broader theory of human behaviour that includes the possibility that participants use reciprocity and cooperate in social dilemmas when they trust others to do the same enables scholars to generate testable hypotheses based on a combination of structural variables as they interact to increase or decrease the likelihood of cooperation and net benefits occurring (see Weber, Kopelman and Messick 2004 for a similar effort).”
2002: “Further, in a small group with extreme heterogeneity face-to-face communication may lead to exacerbate conflict rather than reduction in conflict and agreement on new sets of rules. Instead of one large general causal model, one can develop specific scenarios of causal direction such as those posited above, that can be tested (Ostrom 1998) Thus an important next step in the development of collective action theory is more careful attention to how structural variables interact with one another....It is the combination of these variables that evokes norms, helps or hinders building reputations and trust, and enables effective or destructive interactions and learning to occur.....(see on transactions costs and shared norms 2002)
2003: CONCLUSIONS
A KEY LESSON OF RESEARCH ON COLLECTIVE ACTION THEORY is recognizing the complex linkages among variables at multiple levels that together affect individual reputations, trust and reciprocity as these in turn affect level of cooperation and joint benefits......Instead of looking at all the potential variables, one needs to focus in on a well-defined but narrow chain of relationships as recommended by Agrawal(2002) One can then conduct analysis of a limited set of variables that are posited to have a strong causal relationship.....
