I say it is impossible. But in a comment on a previous post, Justin Rietz argues not at all.
I read your paper "Feasible Globalization" and I believe the premises and therefore the conclusions of the paper (and hence your post) are flawed.
In your paper, you state that many non-market institutions are needed in order for markets to work effectively. You then provide examples of current non-market institutions that supposedly play this role. First, I would argue that institutions that protect the right to private property are hardly “non-market” and there are many examples of private property being protected without government involvement yet with the agreement of the private citizens involved. Moreover, one may just as well argue that markets have worked in spite of many of these institutions, and that many economic problems blamed on unregulated markets should in fact be blamed on government interference (the Federal Reserve is a prime example, but to be debated at another time).
Moreover, few of the institutions cited could not be replaced with market mechanisms. Consider that court rooms are being unilaterally "privatized", evidenced by the significant growth in private arbitration. Many have found it to be as fair, more efficient, and in the end, less costly than the government legal system.
Ultimately, the paper tries to justify the support of protectionist policies by turning the issue on its head. Given the pre-existing conditions we find in much of the world, the move towards global free trade will be long and arduous, mistakes will be made, and those in power will continue to attempt, some times successfully, to shape policies that will benefit their constituents at the expense of others. But isn't this exactly the reason we want free markets - to wrest control of our economic future from inefficient, self-serving, and corrupt governments and instead allow consenting individuals and private parties to engage in mutually beneficial exchange?
And the important bottom line:
We should not denounce free trade because it is incompatible with nation-states, but rather denounce the corrupt, authoritarian, and mercantilist governments that are incompatible with free trade and hence free society.
First. I want to thank Justin for the thoughtful response. He has gone to the source, read my longer article, and written a critique directed straight at my arguments. No smirky one-liner; no cut-and-paste from a post at another blog; no ad hominem. This is the model of what an exchange in the blogosphere should look like. (Paine--I love your verse too!)
Further, Justin's response clarifies where the disagreements lie. The next best thing to coming to agreement in a debate is understanding why we disagree. Justin believes it is possible to envisage a market society where there is no (or minimal) government, and most of what we normally think of as public functions (regulation, stabilization, contract enforcement, property-rights protection, legitimation through social insurance and redistribution) either do not take place (redistribution) or are carried out through private arrangements.
My mind cannot stretch that far. We know from game theory that self-enforcing agreements become impossible to maintain as the number of participants increases and mobility rises--precisely the conditions under which markets deliver the goods. The alternative is third-party enforcement of contracts. And I would rather rely on a democratic state than on the mafia to do the third-party enforcing.
So I see the markets versus the state argument to be pointless. It is markets and the state. Getting the balance right is where the action lies.
UPDATE: Sami B asks for references on the game theoretic approach to institutions. The best single piece, also quite accessible, is by Avinash Dixit of Princeton. It also provides references for further reading. It is here.
UPDATE2: Marius has some really neat and deep things to say about why a democratic state is not the mafia, and should be preferred to it. Read his comments in the thread below.
UPDATE3: I should acknowledge some literature to which Peter Leeson and others have contributed on how social cooperation can be maintained without third-party enforcement even when the standard game-theoretic preconditions do not exist. Here is how Leeson puts it:
Standard folk theorem-type mechanisms, such as reputation, can secure 'good' conduct when populations are small, agents are socially homogeneous, and individuals have roughly equal physical strengths. However, as I have pointed out in previous research, such mechanisms break down when these conditions are violated.
Unlike Rodrik, however, I believe that economic logic and evidence support the contention that anarchy is capable of overcoming both of these obstacles, which would otherwise prevent widespread cooperation without government.
This argument is the focus of my research. "Social Distance and Self-Enforcing Exchange" (forthcoming in the Journal of Legal Studies) deals with how anarchy solves the problem associated with large and socially heterogeneous populations. "Trading with Bandits" (forthcoming in the Journal of Law and Economics) deals with how anarchy solves the problem associated with threat of violence when agents have disparate strengths.
These papers argue that the breakdown of folk theorem-type mechanisms of cooperation under conditions of large numbers, social distance, and disparate strengths does not prevent individuals from developing alternative (non-folk theorem based), private institutional arrangements to enable cooperation and make exchange agreements self-enforcing in the absence of government.