I’d been planning on reading Elinor Ostrom for a while, possibly (though less than probably) before she and Oliver Williamson won the econ Nobel (there not being a poli-sci fauxbel, it was a decent enough fit). Peter Boettke had been writing about her and the “Bloomington School”, but it seemed best to go to the source. A lot of the work she’s known for centers on common pool resources, so “Governing the Commons” seemed the best single text to go for.

I’d first like to get out of the way the canard that Ostrom somehow debunks Garett Hardin’s “tragedy of the commons” or shows it to be a non-issue. She regards it as one possible outcome, and some of the examples in the book are of people that failed to prevent it from happening. Her complaint is that policymakers jump to assume firstly that it will happen if they don’t intervene, and secondly that their intervention will fix things. The people on the ground who actually use the resource often notice the problem and come up with systems that they have the capability of carrying out (for the most part) for themselves, and outsiders intervening sometimes makes things worse. One of my links earlier suggested that Ostrom is arguing for anarchy, and anarchists can certainly rely on some of her arguments, but that would not be an accurate representation of her position. The beneficiaries of common pool resources often make use of government resources (as in her first study, California water basin management), but they can also achieve government backing and still fail (as with some of those very basins).

The framing Ostrom tackles is that the only solution to common pool resource (CPR) problems is for the government to take it over and regulate usage, or for it to split it up into property owned by different users. One problem she brings up is enforcement costs, with the (assumed) distant central authority having a poor ability to detect whether a violation of its rules has ocurred, giving rise to both false positives and negatives in its punishment. Some game theory shows that monitoring problems can result in vastly different equilibriums in these modified prisoner’s dilemnas. The users of the resource tend to be closer to the action and are often motivated to monitor behavior, so she concludes they often have an advantage in enforcing rules, and the optimal rules for enforcement are often something other than preventing anybody else from coming across your square of property (pastoralists often move their animals wide ranges over seasons, so aping stationary farmers can be a mistake). There are times when Ostrom seems to treat individual monitoring as costless, but the point about the varying monitoring costs of different rules shows that not really to be the case. The monitoring doesn’t have to be by the individuals either, they can outsource the job to government or hired private actors. Ostrom notes that the extra monitoring required when dividing up a resource into parcels of private property is avoided if it is all assigned to one agent. However, an arrangement like assigning the current beneficiaries stock in a corporation holding rights to the entire resource pool is little discussed (one federation had to because a new law permitted only “legal persons” to have water rights). It occurs to me that one possible difference between amalgamation into a resource-exploiting corporation and the cooperative arrangements discussed is that clients of the resource (or the outputs for which it is a factor of production) would deal directly with the corporation, rather than the individuals drawing on its resources. That would give rise to a sort of monopsony/cartel, and if those are not expected to be stable that might explain why the people studied did not settle for that degree of coordination. Of course, some of the cooperatives pre-date modern corporate law (quiz for regular readers: what three books would I normally cite here?), and many of the people involved may prefer being independent producers rather than owners/employees of a firm. I have discussed the problems of limited-purpose public-goods providing entities unable to adapt in response to changing circumstances before, but Ostrom’s research finds many organizations maintaining flexibility in their rules to manage CPRs for a very long time.

One surprising bit is where Ostrom writes “Systematic empirical studies have shown that private organization of firms dealing in goods such as electricity, transport and medical services tends to be more efficient than governmental organization”*. I read that a bit after reading Timothy Lee on the “mirage of free-market roads” (along with some other comments on utilities) and Cowen on the comparative performance of private water supply. We should also not forget Robin Hanson‘s writings on how “showing that we care” and refusing to think about death leads to more wasteful medical expenditure than a tightfisted government bureaucracy like the NHS might engage in. Ostrom cites Robert Netting on a set of conditions under which communal ownership makes sense: “(1) the value of production per unit of land is low, (2) the frequency or dependability of use or yield is low, (3) the possibility of improvement or intensification is low, (4) a large territory is needed for effective use, and (5) relatively large groups are required for capital-investment activities”.
*She cites a review, De Alessi 1980, “The Economics of Property Rights: A Review of the Evidence.”

Garett Hardin’s alternative to privatization was government control, which Ostrom also finds imperfect, in part because she tends to assume the central government will be distant (literally and figuratively) from the CPR, resulting in both poor monitoring and understanding. An example she gives is forest nationalization, which has led to bribery of government monitors and overexploitation of forest. Readers of Jared Diamond (or Steve Sailer) will recall that the Dominican dictator Rafael Trujillo was an exception who succeeded in preventing poaching of the extensive nationalized forests. “Have a high quality government” is correct but useless advice in many cases, and whether relying on a devoted agency, enforcement of basic property rights or even just letting locals alone to enforce norms with social sanction and Ellicksonian “self help”, corrupt or inept (Acemoglu would add “extractive”) governments can ruin things. Quality of government involvement isn’t just a matter of being in the right polity, there was wide variation in California water management by basin. Beginning with the relatively small Raymond basin, water expropriators went about a process of litigation and legislation that resulted in an agreement ensuring that sustainable amounts of water would be withdrawn. This was repeated in the West and Central basins (later combined into one water replenishment district) with varying degrees of difficulty. San Bernadino country attempted a similar process with the Mojave Water Agency, but the size of the county and the number of resources managed led to divergent interests and constant squabbling rather than a stable agreement. This would appear to be a case of leaping too quickly into a large & unwieldly arrangement, rather than building up from a more functional scale. Within Sri Lanka there was also divergence between a corrupted fishing arrangement on the coast (with an implicit critique of the quality of post-independence governance with its colonial predecessor), and a once-dysfunctional irrigation program which managed to turn things around.

Among users of the common pool resource, sometimes there are simple rules for appropriation that apply the same to everybody, and sometimes there are more detailed systems of “ownership”. With modern economies that can mean tradeable rights to certain amounts of a resource, sometimes with its own scrip. The amount of ownership is not always equal (long established use tends to persist and become formalized, resulting in an incentive to establish your use early), and being a dominant user of a resource (as in some of the smaller California water basins) gives a large enough stake to take initiative in establishing rules that prevent a race to exhaustion. What Bowles would call “guard labor” is often employed to prevent illicit appropriation, but reputational incentives appear to have a larger effect than the actual value of the fines in many cases. Penalties should start out low and rise with repeat offending, maintaining a perception of a fair & enforced system in which those who strive to abide by the rules are not made “suckers” by cheaters. The penalties set are not always optimally efficient, but the transaction costs to reach agreement on such penalties may be prohibitive.

For those wanting to set up their own CPR management systems, Ostrom provides some design principles. Those consist of clearly defined boundaries, congruence between appropriation/provision rules and local conditions, collective-choice arrangements (stakeholders can influence rules), monitoring, graduated sanctions, conflict-resolution mechanisms, minimal recognition of rights to organize (where even the usually high-quality government of Canada falls short) and (for CPRs part of larger systems) nested enterprises. The work of both recently departed Ostroms (Vincent & Elinor) is characterized by an interest in federalism & polycentric institutions. This makes it attractive to those pushing the anarcho-capitalist envelope at GMU, even if the Ostroms weren’t inclined that way themselves (the same can be said of the above mentioned Robert Ellickson). Sailer has cited Ostrom’s work on the benefits of homogeneity in resource appropriation, using the example of Vietnamese immigrant shrimpers who ignored the social sanction of locals, making it difficult to ensure only a sustainable amount of shrimp was taken. However, while homogeneity may have benefits, it is not required. The Sri Lankan farmers who succeeded in managing irrigation were (at certain geographic scales) diverse in religion/ethnicity, socio-economic status (including rights to the CPR), recency of settlement, and particular needs with regard to the resource. The key was to have a nested structure. Many of these diverse farmers did fail to cooperate, but on the left bank of the Gal Oya irrigation project the government hired the overeducated & underemployed from farming backgrounds to be institutional organizers, first getting relatively homogenous neighbors using a small resource unit (a distributory canal further divided into field canals) to cooperate. There wound up being four tiers of organization, with the smallest level comprising 12 to 15 farmers, but the bureaucrats & academics behind the project made sure to get their industrial organizers working with that minimal group to cooperate before adopting any more formal organization or initiating the rehabilitation of canals.

The most common environmental problem discussed nowadays is global warming, which along with deep sea fishing (which gets some discussion in this book) has pretty much all the characteristics the book mentions as predicting a failure to coordinate. Ostrom herself was fairly optimistic about bottom-up efforts. To me the problems of cooperation on that scale seem insurmountable enough that the unilateral nature of geoengineering makes it the most feasible response. The massive top-down risky nature of it would likely frighten Ostrom. Rightfully so.