Leeson makes three points in response to my critique.
- In theory, self-enforcing agreements can indeed scale up, if they rely not on punishment in repeated play, but on signaling, whereby “honest” types separate themselves from “dishonest” types.
- The large volume of international trade is evidence that self-enforcement without government is in fact possible.
- The correlation between income per head and size of government arises from the fact that more prosperous countries can afford to pay the higher costs that larger governments impose.
Let me say a few words about each of these points.
1. Although I have not studied the papers that Leeson refers to, I think I understand the theory that Leeson has in mind. I do not agree that this line of reasoning evades the scaling up problem. Successful signaling relies on separating, rather than pooling, equilibria. In other words, “honest” traders have to incur costs that “dishonest” traders would find too onerous to bear in order to effectively signal their type. That means some participants have to be shut out of the benefit from trade in order for the remaining to benefit.
You might say that this is just as it should be: if there are honest and dishonest traders, it is only the former that deserve the benefits of economic exchange. But as soon as we put it this way, we understand what is wrong with the signaling argument in this context. People are not born honest or dishonest: enforcement problems arise in practice not because honesty is an attribute of some individuals and not of others, but because all individuals have the incentive to behave opportunistically when circumstances allow them to do so. Relying on signaling in these settings, even when it works, leaves many wasted opportunities for economic exchange.
2. On international trade, I had made the point that much of it still relied on the shadow of the state. What I meant by that is not the esoteric New York Convention at all. What I was thinking of is that traders can still sue each other in their respective legal jurisdictions, or in fact in third-country jurisdictions. Arbitration decisions too are typically enforceable in some legal jurisdiction. What keeps traders honest is not simply the fear that they will be excluded from future commerce, but also that their assets can be taken away through legal judgments.
The other point I had made is that because these legal and quasi-legal arrangements for contract enforcement work more poorly in the international arena than they do domestically, international exchanges are still subject to much higher transaction costs. Anderson and van Wincoop have estimated these transaction costs to be around 40 percent in ad valorem terms. And indeed there is a large literature on international trade that talks about “missing trade” and “border effects.” Trade volumes may be high, but not anywhere near as high as they would be without the transaction costs imposed by jurisdictional discontinuities at the national border.
3. On the correlation between size of government and level of incomes, I am happy to accept that there may be some reverse causation. But what exactly does that mean? It means that people in rich countries demand more government! Public services are a luxury good, to put it in economists’ language. Why would that be, if government came only with rent-seeking, corruption, and other costs?
I am also happy to accept that more government is not necessarily better government, although I would not rely on the Heritage Foundation index to make that point. There are some big governments that do a lot of bad things (North Korea) and some big governments that do mostly good things (Sweden). I just have a hard time with the doctrinaire view that identifies government with only the former model.