Edward Stringham has written a great book, Private Governance, and a great introductory essay to this Cato Unbound. He argues that private governance, meaning non-state governance, can enforce complex contracts and often does so instead, or even in spite, of government. His research ranges from historical, to modern: at the emergence of stock markets, the Dutch government, for example, banned many types of contracts which were nevertheless enforced by the stock exchange; in the modern world, PayPal deploys sophisticated and effective anti-fraud mechanisms.
Such mechanisms work because of the logic of club goods. Clubs turn interactions into a repeated prisoner’s dilemma, where the discounted future value of cooperating is higher than the value from defecting. As such, it is in the interest of participants to continue cooperating.
While I would strongly recommend the book, I will take this essay to delineate how far the logic of private governance can be extended, something that Stringham did not do. First, is there a set of contractual interactions to which the club good model is not easily applicable? Second, how well does private governance deal with violence?
The success of private governance depends on whether the previous actions of participants are easily identifiable. If so, cheaters will be avoided and cooperators will be interacted with again. However, there are a class of people for whom their previous actions are not easily identifiable.
Imagine you are an entrepreneur in the third world. You have started a business, but cannot grow it because you are capital constrained. Banks are unwilling to lend you money because the government cannot be trusted to recover capital if you are late during repayment. Because you are a new entrepreneur there is not enough information about your ability and willingness to repay loans for the bank to simply trust you.
If we ignore the government failure of enforcing the banking contract, it is also apparent that a private governance mechanism cannot solve this need. And a recent paper by David McKenzie suggests it is stronger than usually recognized.
McKenzie examines the results of a business competition in Nigeria where a randomized selection of 729 firms were given an average of $50,000. After three years he found, “Surveys tracking applicants over three years show that winning the business plan competition leads to greater firm entry, higher survival of existing businesses, higher profits and sales, and higher employment, including increases of over 20 percentage points in the likelihood of a firm having 10 or more workers. These effects appear to occur largely through the grants enabling firms to purchase more capital and hire more labor.”
This suggests the private credit market in Nigeria is deficient in ways that have a strong effect on business creation and growth. The success of this policy experiment suggests there were long hanging entrepreneurs, so to speak. If so, then why wasn’t the market able to come up with a governance system to ensure they received credit which would have helped their businesses grow? As this example demonstrates, private governance may work far better than economists have believed, but it does not seem to be a replacement for all types of public governance.
The second question I would like to raise is how well private governance deals with violence. Stringham is often thought of as an anarchist in libertarian circles, having edited Anarchy and the Law. While he does not endorse anarchy in Private Governance, he does include a chapter on private police and a chapter arguing that Hayek would have been an anarchist had he followed his reasoning to its logical conclusion.
In this instance I think his book would have been improved if it were more clear about how far private governance may extend. For example, the logic of club goods only applies to police if they are operating in the shadow of the state. Private police forces in San Francisco, for example, are unable to extract resources through violence. While they provide superior services to the SFPD, it is unclear how they would act without the oversight of the state.
What might be termed the hard question of private governance is how violence is resolved outside the shadow of the state. David Friedman famously theorized that competing dispute resolution agencies and their police forces could reach an equilibrium in which no group held a monopoly on the use of force. Stringham had previously argued that proprietary communities, bundling real estate ownership with law enforcement under a single legal owner, could provide a stable form of governance.
However, the theory of clubs, at least as it is presented in Private Governance, does little to support either line of anarchist thought. Without a state, violence would be the defining relationship between groups and individuals rather than repeated social interaction. The relative prices are fundamentally different when exclusion means a violent defense of property, rather than non-inclusion in future business dealings. The implications we can draw from 17th century stock markets are limited in circumstances like these.
One common argument for free markets is they support low-frequency preferences, leading for example to more than 23 types of deodorant and 18 types of sneakers. However, it is a probably a good thing there is no market for violence, thereby suppressing the satisfaction of low-frequency preferences (I’d like to thank Vlad Tarko for this line of argumentation). Take, for example, the Ku Klux Klan. After the Civil War, it was founded with the goal of oppressing newly freedmen and suppressing their allies. Such actions have no place in a civilized society. In modern America there are still people with noxious views about race, religion, and immigration. It is a good thing that their low frequency preferences for violence are not satisfied.
I expect the response of an anarchist would be to argue that such violent preferences are weakly held, at least compared to the subjects of the violence, and they would be bargained away in an anarchist equilibrium. While such a scenario is conceivable, we cannot realistically identify the plausibility without evidence. Until such evidence is available we should admit ignorance.
That being said, this response is not so much a critique as an attempt to find the limitations of Stringham’s arguments. My only suggestion would be for him to do so in his book, instead of implicitly asking his readers to this work. Nonetheless Private Governance is an excellent book which I expect to cite often and strongly recommend to any economist or learned citizen interested in all forms of governance.