Tragedy of the Commons and Sustainable Thought: A Network Perspective
Given a shared, non-renewable common access resource and a self-interested population, it is commonly found that despite everyone in the population knowing about the concept of “tragedy of the commons”, it is in everyone’s “game-theoretic” best interests to continue using from the resource to the point of depletion. Simply put, every actor views his possible actions as “continue usage to the point of depletion” or “curtail usage while everyone else continues as normal”, and the first option evidently has a higher short-term payoff.
There is also an interesting question with regards to more abstract environmentalist ideas: given a population of people who are perfectly educated on the consequences of unsustainable behavior, why do such unsustainable behaviors persist? I find that both the classical “tragedy of the commons” and this more abstract social effect (without regard to a concrete resource) find their roots in similar network models.
Consider a network of people in a population, where each person is connected to some number of people that represents a small fraction of the total population. We assume that the population is quite large. Each person has the option to practice sustainable (S) or unsustainable (U) behavior, and the network initially starts out as uniformly practicing U. Due to the long-term nature of environmental consequences and the independence of actors in the population, the game-theoretic considerations may only take short-term payoffs into account. To this end, one may develop a two player “game” on the edge in the network graph where the “strategies” are S and U for each player. The payoffs in this simplified model are as follows:
(U, U): 3, 3
(U, S): 3, 0
This is because we assume that unsustainable behavior has some nonzero “selfish” benefit (perhaps people enjoy idling their car to warm up), but people who practice sustainable behavior while their neighbor practices unsustainable behavior are likely to feel that their efforts are unreciprocated or meaningless, or they may feel like they are not getting their “fair share” of comfort. Additionally, there can be an instrumental loss of utility for the sustainable agent: if everyone else is traveling by car and the agent has to settle for a bicycle, he might have limited opportunities for employment as a consequence of his lack of mobility around a community. Essentially, the fundamental idea here is that there is a benefit to copying others’ decisions.
(S, U): 0, 3
By symmetry.
(S, S): 4, 4 or 2, 2 depending on choice of model
Again, there is a benefit to copying others’ decisions. People in a sustainable community are more likely to psychologically internalize the benefits of living sustainably, so they will reap a benefit from copying the decision to be sustainable.
The payoffs for coordinated sustainability might be lower or higher than coordinated unsustainability. This is an empirical quality of the community itself and how much it values the idea of sustainability, morally or instrumentally (or analogously, how much it values the benefits of unsustainability).
This model is sufficient to analyze the spread of the “trend of sustainability”. According to the model proposed by Easley and Kleinberg, a node in a network plays the above game with all of its neighbors, and its strategy (S or U) is determined by whatever maximizes payoff. The initial network state (uniformly U) is stable, as no node can independently switch to S without decreasing their payoff (as the payoff for S, U is 0, 3). This holds even if each person has perfect knowledge of the fact (obviously not in these quantitative terms) that the payoff for coordinated sustainability is 4, 4 or even 100, 100. If one, or even ten people in a network decide to switch to sustainable behavior (perhaps they do not care about the payoffs predicted by the model, having been swayed by a particularly heart-wrenching advertisement), then a node A that plays U will have a certain proportion of its neighbors playing S. However, due to the large network size (and the somewhat-large number of neighbors that A has) and hence the large remaining number of neighbors playing U, the payoff to A from playing S is still not as great as the payoff from playing U. Essentially, the few people that switched to sustainability were not enough to surpass the threshold proportion of neighbors for an arbitrary node in the network to switch to S. As real life is likely faithful to this model, this serves as an explanation for why people choose to not be sustainable despite knowing that it is “the right thing to do”—it all stems from the benefits of copying others and the lack of benefits from doing otherwise.
The model extends trivially to the tragedy of the commons—just replace S with “use resource sparingly” and U with “use resource inconsiderately”.
There exists evidence of this model of network behavior occurring in real communities. This analysis (http://blog.gbntc.com/why-environmental-and-sustainability-programs-are-growing-in-popularity/) of the growing popularity of sustainable and environmentalist thought attributes this spread to the prevalence of higher-education programs centered around sustainable development and the increasing instrumental utility of sustainability. According to the above network perspective, one can interpret the prevalence of sustainability education as points in the network where the trend of sustainability is introduced, and the multiplicity of these “injection points” promotes spread of the trend through the network, allowing nodes to surpass their threshold of neighbors practicing sustainability, leading to that node “switching to sustainability”. The increasing social utility of sustainable action can be interpreted as the payoffs given above changing—either coordinated unsustainable behavior is no longer as beneficial, coordinated sustainable behavior is more beneficial, or uncoordinated behavior favors the “S strategy” increasingly (as opposed to the payoff for S, U simply being 0, 3). All this serves to reduce the threshold value for when a node is compelled to switch from U to S.
Additionally, there exist real-world interpretations of certain intricacies of the network game-theoretic model, namely the relative utility of sustainable behavior versus unsustainable behavior. There is increasing research on what it means to have a “culture of sustainability”, as well as how to develop one in a community (https://www.fastcompany.com/3020115/5-ways-to-create-a-culture-of-sustainability-in-any-company). Essentially, if a community values sustainability, one might see the payoffs for (S, S) be greater than the payoffs for (U, U). The reverse happens when a community does not see sustainability as an ideal. This of course has consequences, from a network model perspective, for the threshold proportion of sustainable neighbors of a node in the network for that node to become sustainable. Essentially, the importance of developing a culture of sustainability lies mostly in increasing the ability of sustainable “trends” to spread.
In sum, the current state of sustainable affairs in communities can be attributed entirely to the fact that sustainable behavior is a cost-benefit analysis for a single agent, as opposed to a piece of beneficial information that spreads through networks unimpeded. Hence, the spread of sustainable behavior must follow the network model proposed above, which relies on the principle that copying the behavior of others is beneficial. Understanding social network structure from this point of view will help guide future initiatives for sustainability and environmental awareness.