Hey Kids! Transactions Costs!

In this macro-brutalizing post, William Buiter lays down many passages I agree with. I don't believe it was offered in this spirit, but I think Buiter does a servicable job of explaining why my fave econ gurus — F.A. Hayek, James M. Buchanan, Douglass North, and Vernon Smith — will make you smarter than your local macro textbook. Buiter:

[M]ost of the New Classical and New Keynesian macroeconomics assumes away the problem of contract enforcement.  This problem is especially acute in trade over time or intertemporal trade, where the net value to each party to a contract of fulfilling the terms of the contract varies over time and can change sign.  In a world with selfish, rational, opportunistic agents, able and willing to lie and deceive, only a small set of voluntary transactions will ever be observed, relative to the universe of all potentially feasible transactions.

The first set of voluntary exchange-based transactions we are likely to see are self-enforcing contracts – those based on long-term relationships, repeated interactions and trust.  There are some of those, but not too many.  The second are those voluntarily-entered-into contracts that are not self-enforcing (say because interactions between the same sets of agents are infrequent and market participants have a degree of anonymity that prevents the use of reputation as a self-enforcement mechanism) but are instead enforced by some external agent or third party, often the state, sometimes the Mafia (sometimes it’s hard to tell who is who).  Third party enforcement of contracts is again often complex and costly, which is why it covers relatively few contracts.  It requires that the terms of the contract and the contingencies it contains be third-party observable and verifiable.  Again, only a limited set of exchanges can be supported this way.

The conclusion, boys and girls, should be that trade – voluntary exchange – is the exception rather than the rule and that markets are inherently and hopelessly incomplete.  Live with it and start from that fact.  The benchmark is no trade – pre-Friday Robinson Crusoe autarky.  For every good, service or financial instrument that plays a role in your ‘model of the world’, you should explain why a market for it exists – why it is traded at all. Perhaps we shall get somewhere this time.

[…]

The future surely belongs to behavioural approaches relying on empirical studies on how market participants learn, form views about the future and change these views in response to changes in their environment, peer group effects etc. 

Yeah! Do some real science, economists!

Let me point out that Buiter badly oversteps his line of reasoning when he says that voluntary exchange, as such, is the exception rather than the rule, because trade is in fact a ubiquitous feature of human society. The baseline is not “no trade” but the far from negligible level of exchange in hunter-gatherer societies. What Buiter might have said had he not himself wasted so much time with macroeconomics textbooks is this: There are transactions costs. These limit the trades it makes sense to make. But a vastly greater number of trades can be made (and vastly greater gains from exchange realized) when transactions costs — such as the costs of enforcing complex contracts — fall. The complex order of spatially and temporally extended impersonal exchange is the exception, not the rule. And that's why almost all of humanity has been poor for almost forever.

But the main idea here is right, and it's good to see a central banker grasp it: economics that leaves out transactions costs simply assumes what humanity has only rarely managed to approximate. Modern economies are weird, pulsing, unsteady achievements of ongoing cultural evolution. Economies certainly aren't machines governed by physics-like regularities. Nor is “an economy” the creature of specious nationalist bookkeeping studied in textbooks. There are lots of things we need to grasp if we are to “get somewhere this time.”