How important is cooperation in a market society?
First, what is cooperation? Suppose a number of individuals occupy a common social and geographical space. They have a variety of individual interests and things they value, and they have outcomes they'd like to bring about. Some of those outcomes are purely private goods, and some can be brought about through private activities by each individual. These are the circumstances where private market-based activity can bring about socially optimal outcomes.
But some outcomes may look more like public or common goods -- for example, greater safety in the neighborhood or more sustainable uses of resources. These are outcomes that no single individual can bring about, and -- once established -- no one can be excluded from the enjoyment of these goods. (Public choice theorists sometimes look at other kinds of non-private goods such as "club goods"; see Dennis Mueller, Perspectives on Public Choice: A Handbook.)
Further, some outcomes may in fact be private goods, but may be such that they require coordinated efforts by multiple individuals to achieve them efficiently. An example of this is traditional farming: it may be that the yield on one individual's plot is greater if a group of neighbors provide concentrated labor on weeding this plot today and the neighbor's plot tomorrow than if each of us do all the weeding on our individual plots. The technical conditions surrounding traditional agriculture impose a cycle of labor demand that makes cooperation an efficient strategy.
This is where cooperation comes in. If a number of the members of a group agree to contribute our efforts to a common project we may find that the total results are greater -- for both common goods and private goods -- than if we had each pursued these goods through individual efforts. Cooperation can lead to improvement in the overall production of a good for a given level of sacrifice of time and effort. This description uses the word "agree"; but Robert Axelrod (The Evolution of Cooperation) and David Lewis (Convention: A Philosophical Study) observe that many examples of cooperation depend on "convention" and tacit agreement rather than an explicit understanding among participants.
So cooperation can lead to better outcomes for a group and each individual in the group than would be achievable through entirely private efforts.
Cooperation should be distinguished from altruistic behavior; cooperation makes sense for rationally self-interested individuals if appropriate conditions are satisfied. A cooperative arrangement can make everyone better off. So we don't have to assume that individuals act altruistically in order to account for cooperation.
So why is cooperation not ubiquitous? It is in fact pretty widespread. But there are a couple of important obstacles to cooperation in ordinary social life: the rational incentive that exists to become a freerider or easy rider when the good in question is a public good; and the risk that cooperators run that the endeavor will fail because of non-contribution from other potential contributors. There is also often a timing problem: it is common for the contribution and the benefit to be separated in time, so contributors are even more concerned that they will be denied the benefits of cooperation. If Mr Wong is asked to weed today in consideration of assistance from Mr Li in harvesting the crop four months from now, he may be doubtful about the future benefit.
The basic logic of this situation has stimulated a mountain of great social science research and theory. Garrett Hardin's "tragedy of the commons" (Managing the Commons) and Mancur Olsen's The Logic of Collective Action: Public Goods and the Theory of Groups set the negative case for thinking that cooperation is all but impossible to sustain. Elinor Ostrom's Nobel-prize winning work on common property resource regimes documents the ways in which communities have solved these cooperation dilemmas (Governing the Commons: The Evolution of Institutions for Collective Action). Douglas North essentially argues that only private property and binding contracts can do the job (The Rise of the Western World: A New Economic History). And Robert Axelrod has made the case for the rational basis of cooperation in The Evolution of Cooperation: Revised Edition. He argues that there are specific conditions that enhance or undermine cooperation and reciprocity; essentially, participants need to be able to reidentify each other over time and they need to have a high likelihood of continuing to interact with each other over an extended time. (His analysis is based on a series of experiments involving repeated prisoners' dilemmas.)
A market can "simulate" cooperation through enforceable contracts; so, for example, a peasant farming community could create a legally binding system of labor exchange among households. And organizations can create quasi-binding agreements for cooperation through "memoranda of understanding" and "inter-governmental agreements" -- written agreements that may not be enforceable through legal remedies but nonetheless create a strong incentive for each party to fulfill the obligations of cooperation. However, quite a bit of the opportunities for cooperation seem to fall outside the sphere of these formal and semi-formal mechanisms for binding agreements.
Informal cooperation needs some kind of institutional or normative setting that encourages compliance with the cooperative arrangement. So there has been an energetic debate in the past twenty years over the feasibility of non-coercive solutions to cooperation problems; this is an area where the new institutionalism has played a key role. And in the real world, we do in fact find numerous sustainable examples of informal cooperation. Individuals work in community gardens; foundations join together in supporting urban renewal projects; villagers create labor-sharing practices. But it is an interesting question to consider: are there institutional reforms that we could invent that would allow us as a society to capture more of the benefits of cooperation than we currently realize?