Introduction to Collective Action Problems

It would be simple if the problems in the world were the fault of certain evil individuals. By getting rid of those individuals, we could greatly improve things. But social problems are much more pernicious and more deeply embedded in the world when they are the outcome of the actions of good people. How could that be? How can good people produce much wrong? A lot of the time, it’s because of their environment, which structures their individual behavior into collective action problems.  

I've already written a (not totally) tongue-in-cheek post about a particular collective action problem, but this post is more of an introduction to the topic. Collection action problems are a very common theme on this blog because so many different and seemingly unrelated political issues can be distilled down to the problematic dynamic that results when many individuals’ actions each play only a marginal role in determining whether the outcome those individuals commonly experience is good or bad. The issues that make up the enduring dispute over political economy and the size and shape of the state, those that pretty consistently pit liberals against libertarians and conservatives, are often reducible to problems of collective action and how best to address them. These issues are ripe for controversy because the market system that we usually employ to distribute our collective duties and resulting benefits, uncoordinated self-interested action by rational individuals leading to optimal outcomes for all as though chosen by an invisible hand, does not apply as neatly in these cases. Instead it leads to an avoidably bad outcome. This is not to say that the faintest example of a collective action problem calls for any and all government action, but just that there is more reason to be wary of market outcomes in such cases and so more reason to consider alternatives.

A collective action problem (otherwise known as a coordination problem, the Tragedy of the Commons, a free-rider problem, an externality, a public good, the Paradox of Thrift, possibly even a Prisoner's Dilemma, among sure more labels) is a situation in which multiple people, through their informed, self-interested, rational behavior that they undertake as individuals in an anarchic environment without coordination among them, produce an outcome that is suboptimal for all of them, and that is not even Pareto-optimal. In other words, not only could each individual have attained a better outcome, but moreover this is true for all the individuals simultaneously; so they were not even forced into their actual outcomes because some of their better alternatives of were somehow incompatible with each other, as in a Zero-Sum Game. To the contrary, collective action problems are so pernicious because there does exist a single alternative possible outcome that is universally an improvement, or at least non-injurious. In short, if everyone had acted differently, all could be better off and no one worse off.  

But they all took their best available course of action, and still endured an avoidably bad result. In collective action problems, what is responsible, rational and virtuous when done as an individual, becomes reckless, harmful and vicious when done by a group. Private virtue is public vice. As such, collective action problems are examples of the Fallacy of Composition (also known as the Part-Whole Fallacy, i.e. what is true of the part need not be true of the whole and vice versa): what is good for the individual to do might be bad for the society to do.    

***

The cause of this unfortunate tendency in group behavior is not any deficiency in the individuals involved, as made clear by the inclusion of “informed” and “rational” in the definition above, but instead is the system of incentives that emerges when: 1. individuals commonly face a single impending outcome of multiple possible consequences that vary in desirability, 2. that must be addressed by a large enough effort to render each individual’s contribution only a marginal fraction in comparison, and 3. when they address these concerns individually and without coordination in an anarchic environment.

The mechanism that produces this unlikely-seeming outcome is the temptation to free-ride on the efforts of others and attempt to enjoy the benefits their contributions provide without paying one’s own share of the costs. The thinking goes that if one is simply one of many, one’s contribution toward the desired outcome is negligible; if I alone choose not to contribute and pay my share of the costs toward furthering a social end, and everyone else’s actions are fixed, then my action won’t determine the outcome. Either enough people will have paid their share of the costs and the effort will succeed, or too many thought to free-ride and shirk their share of the cost, and the effort will fail. In either case, my own miniscule contribution toward the effort changes nothing either way. So why should I pay the costs if the outcome I enjoy is fixed either way?

This first-order free-rider thinking can spark a self-reinforcing cycle of free-riding. Some of those who paid up initially may reconsider after having seen the free-riding example of others, whether out of resenting being one of society's suckers or even just from learning of the possibility. This dynamic is self-sustaining because the increase in free-riding that results from the awareness of its initial examples and accompanying resentment just provides more visible examples to produce even more resentment and awareness. Moreover, as more people choose to free-ride, the outcome for those who do pay is an increased price for the service or diminished quantity or quality (or having to bear some other cost to cover the implied funding gap), or some combination thereof. This only increases the incentive to stop contributing and free-ride instead. At some point, even those honorable and conscientious people who originally paid up (and who care enough for the collective purpose to not be tempted by the option of free-riding, including after the free-riding of others has made them resentful and driven their share of the cost up and the quantity or quality of the services they enjoy down) may stop believing in the sustainability and feasibility of the project; at that point, they face an additional incentive to stop paying: that the project will fail anyway and their contribution would be wasted. Why pay to further a goal when the attempt is doomed to fail anyway? A common endeavor yielding a common benefit stricken by this dynamic is not long for this world. 

So the collective action problem is a state of affairs in which many agents, when acting individually, face a perverse system of incentives, which rewards the socially irresponsible and punishes the socially responsible, and leads to an avoidable sub-optimal outcome for all of them.

***

Although it will become clear on this blog that many, many different political issues can rightly be characterized by collective action problems, here are just a few notable examples:

The idea of the Tragedy of the Commons is used to show the undesirable consequences that result from the collective action problem faced by individuals who share a common resource without coordinating their use of it. For example, the common use of the town green to graze all the residents’ livestock will end in tragedy if the residents do not coordinate. What incentive does one resident have to prevent his animals from grazing as much as possible (and to thereby limit the animals' size and value) in order to preserve the useful life of the town green, if everyone else does not do likewise and therefore the green is rendered barren just as fast? In this circumstance, he who does act virtuously will be disadvantaged compared to those who do not.  

In more recent thinking, the guy who wrote the book on collective action problems, Mancur Olson, used the example of political lobbies: why does public policy so often favor the interests of the few over the interests of the many? The collective action problem facing the few is not as great as the one facing the many; it is easier to organize few than many into lobbies and special interest groups. On public policy issues that pit the interests of industry and production against the interest of consumers, the benefits to consumers if policy makers side with them are diffuse while the costs to industry would be concentrated. Thus, industry has a greater incentive to organize and lobby for their interests than do consumers. Moreover, the costs of organizing producers is lower than that of organizing consumers since there are fewer of the former than the latter.

The idea of riding a public transportation trolley without paying for it by hopping on the back or jumping the turnstile gives us the name “Free-Rider Problem” for this whole concept.

Economic thinking discusses externalities, both positive and negative, and public goods. Externalities are effects felt by third parties, not party to the deal that caused those effects, and also therefore not consenting to them. Externalities can be positive and desired, or negative and undesired. Positive externalities, also called “public goods”, entail a collective action problem in their being systemically underfunded without coordination; some people would rather free-ride on the efforts of others to produce these public goods, and others choose not to produce them at all because they cannot recoup their costs or earn enough profit to justify foregoing a different opportunity cost. Commonly, public goods are those goods that are deemed to be non-rivalrous (as in they can be enjoyed by multiple different people at the same time) and non-excludable (non-payers cannot be excluded from enjoying them). Similarly (although probably not strictly part of the definition of public goods), goods and services with high fixed costs and low marginal costs are taken to be good candidates for public goods. Negative externalities meanwhile are overproduced without coordination because they impose undesired effects on unconsenting third parties, which means that responsible parties avoid paying some of the costs of their activity and their marginal cost is less than their social cost. In the many cases in which the negative externality degrades the quality of a common resource, like in the case of environmental pollution or noise, attempts to individually resolve the problem encounter a collective action problem: why should I be the only one not to pollute? why should I not litter this trash right now when it's too small to make a difference?

Again in economics the concept of the collective action problem is illuminating when applying it to some disastrous macroeconomic phenomena: demand-side recessions and depressions. In recessionary economies, people have an incentive to cut their spending and increase their saving. But because one person’s spending is another person’s income (and vice versa), everyone cutting their spending to increase their saving leads only to lower spending and saving overall. This is the Paradox of Thrift, a clear case of when private virtue becomes public vice. Moreover, the spending that is needed to revive an economy in depression or recession, whether by consumers to spur more investment, or by industry and companies to invest and hire more labor, is held back by the collective action problem. Each individual consumer or firm has an incentive to not increase their spending while hoping others do. Individual agents that do increase their spending unilaterally are disadvantaged compared to those that do not.

There are obvious examples of significant collective action problems relating to our need of a hospitable climate and environment. On climate change, every individual person and company faces perverse incentives to free-ride or otherwise shirk their contribution to reducing global emissions, but so does every individual country have an incentive to not unilaterally act to reduce their own emissions. But there is more than just climate change: the problems of overfishing and overgrazing, groundwater cleanup, acid rain, even littering and many more are examples of collective action problems. That this area provides so many examples is only natural, as there is a single planet for many individuals, and compared to which individuals and their actions are often seen as insignificant.

Of significant note is that the effort to eliminate or alleviate poverty through individual voluntarily schemes has been thought of as a collective action problem, and rightly so.  This has been the justification in economic circles for government intervention into the economy to relieve poverty, for if people spend their resources to relieve poverty, everyone gets the benefit, so regular free-riding thinking applies. The benefit of poverty reduction can be the virtuous one of simply an improved moral state of the universe, or it can be more like fewer unpleasant interactions with the poor and a lower public health threat; some interpretations are more selfish than others but the problematic collective dynamic persists regardless.

In world affairs, the problem of the threat by excessive militarization, especially nuclear armament, is commonly conceived of as one of collective action known as the Security Dilemma. Each individual state has an incentive to be as armed as possible compared to its rivals, and for its rivals to be as little armed as possible. So a prudent state, acting on those incentives, builds up its military capability. That however spurs other states to respond rationally to the increased threat and follow suit, building up their own militaries and undoing the original state's security advantage. This of course has the potential to become a reinforcing cycle, as the original state rediscovers the incentive to arm itself upon the elimination of its original security edge and kicks off another round of military buildup, which inspires another similar round in its competitors, and so on and so on and so on. And when it comes to demilitarization, each individual state has an incentive not to disarm first: that would put them at a disadvantage compared to everyone else.

Even the idea of the Prisoner’s Dilemma is a simplified form of a collective action problem, one with just two participants.

So having covered just a few examples, the diversity of collective action problems is now obvious. It stems not only from the different real world issues involved, but from some key ways in which collective action problems can differ. Some are situations that entail existing harms in the present (like destructive storms from climate change or a depressed economy) and the problem consists of an inability to make progress toward remedying them. Others are situations that entail minimal harm at present, or even none at all, but which promise substantial harms in the future if the existing course of action is maintained (like the exhaustion of natural resources or deficit-caused interest rate spike); in these cases, the problem consists of an inability to make the necessary changes to avoid catastrophe. In addition to varying by timing, collective action problems can vary by whether they thwart efforts to avoid a harm, either in the present or future (like the examples of climate change, or a badly performing economy, or natural resource exhaustion, or a spike in interest rates), or whether they thwart efforts to produce an independent positive benefit (like the creation of a public transportation trolley or road and highway system).* Finally, collective action problems vary by the nature of the costs and benefits involved, in that they are either binary "all-or-nothing" possibilities (like a trolley system that either does or does not service a certain segment of the track), or instead they are not binary and the involved costs or benefits can vary by matter of degree or number of units (like carbon emissions that can be reduced by greater or lesser amounts).

***

The point in analyzing and identifying collective action problems has to be in the interest of resolving them. 

A standard solution to many problems of collective action is privatization. If we abolish the problematic commons, privatize the commonly held assets and divvy it all up into individually held private property, then we have solved the problem: incentives are realigned more appropriately since people no longer necessarily face a common fate and can no longer free-ride on the efforts of others. This "Coaseian" solution to just clearly establish property rights and let the invisible hand do the rest is attractive because it allows for more efficient outcomes, especially compared to the alternatives of taxation or regulation (or tort law, in the case of negative externality causing private injury).

This market solution applies very neatly to economically-conceived collective action problems like externalities, but can not resolve all collective action problems. Some commons can't be privatized or are extremely costly to do so: how to delineate different pieces of private property in the ocean, in the sky, or further out in space, especially in a way to stop activity in one plot from bleeding through to another nearby plot? It's possible that some commons, while in theory could be privately owned, might cost more to privatize than to regulate or even than to just suffer through the collective problems. This is to say nothing of collective action problems based on a commons that is not a commodity, like national defense: how to defend only one of two adjacent neighbors?

So privatization and the market can not be our only tools to resolve collective action problems or some problems will go unsolved. The other type of solution is coordination and planning among individuals to replace the anarchic environment in which they act and realign their incentives. They must construct a system that coordinates their individual simultaneous actions and that enables individual participants to verify the participation of others, with enforcement measures to ensure against free-riding that kick in when one’s level of participation lags behind others’ (i.e. one won’t be singularly compelled to participate if there are others who are also not participating).

***

Footnote:

* By "independent positive benefit", I mean a good defined independently from a specific harm, such that the good cannot be characterized as a solution to any problem. Sure, the absence of a good is worse than its presence, so such an absence could perhaps be called a problem or a harm, but only relative to its presence. Before the benefit became a possibility, its absence was just the way things were and therefore not viewed as a problem. So I am attempting to delineate the difference between producing a benefit that solves an existing and recognized problem and producing a benefit that does not address any such recognized problem. This may seem like an arbitrary distinction, but I think it helps drive home the variety of types of collective action problems.