Lynne Kiesling
In the post-election show of sleeve-rolling-up meeting between Barack Obama and John McCain, their main rhetoric revolved around how they could work together to “fix up the economy”. At the time I wrote about how that language rankled me (and Russ Roberts), because the economy is not a closed-system project, and politicians who have the hubris to believe that they can treat the economy like it’s a construction project will do us great harm. This hubris is what Hayek called the “fatal conceit”, but the phenomenon was apparent in the 18th century as well, when Adam Smith wrote in Theory of Moral Sentiments about the perils of giving “the man of system” power (as I discussed in this post from 2005).
In the intervening (pun intended!) two months, the prospect has gotten worse, with a proposed stimulus package approaching one trillion dollars that promises decades of debt to fund current spending that will have uncertain outcomes. Certainly this expenditure and debt obligation will “fix the economy”, won’t it?
In a very important article that I urge you to read, Max Borders joins me and Russ (and others) in observing that “fixing the economy” is precisely the wrong way to think about fostering economic growth and productivity. The economy is not like a house, it’s not like a machine,
The economy is an organic ecosystem.
Thus Max titles his article “The Economy Is Not A Machine”. In his explanation of why that’s the case, he invokes precisely the concept from which the name of this web site is derived:
But the whole idea of fixing, running, regulating, designing, or modeling an economy rests on the notion that, if the right smart guys are at the rheostats, the economy can be ordered by intelligent design. But the economy is no mechanism. There is no mission control. Government cannot swoop down like a deus ex machina to explain the inexplicable and fix the unfixable. Why? Because the knowledge required to grasp each of the billions of actions, transactions and interconnections would fry the neural circuitry of a thousand Ben Bernankes. This is what F. A. Hayek called the knowledge problem.
He then goes on to give an excellent description of why and how the economy is a complex adaptive system (one example of which is an organic ecosystem).
Both ecosystems and economies are distributed systems. In the former, billions of interdependent means-ends activities are a reflection of a billion preferences and choices. In the latter, species are dynamic and interwoven in a web of relationships. For both, the whole system is an ever-evolving cascade of change that is unfathomable to a single mind. Data snapshots may be useful for some things, but should not be intended as blueprints for government planners. Even sophisticated computer models will, like the old Philips machine, eventually fail. There are no oracles.
And the laws of ecosystems are not the mechanistic laws that allow us to predict with very good accuracy how machines will behave. Ecosystems evolve by trial and error, by experimentation. Any entrepreneur who has been responsible for a new product launch can tell you that markets are not mechanistic, and that distributed non-mechanistic nature aggregates up across markets into an economic that is a complex system.
His recommendation will sound familiar to you, because I have recommended it many times here and in my published work: design clear and transparent institutions that reduce transaction costs if you want economic growth.
By fundamentals I mean rules. Only rules can be the product of human design. These are the simple rules that lower “transaction costs,” which is a fancy way of saying help us trade with each other easily, minimizing conflict. We call these rules “institutions” (property rights, contract enforcement, and so on) — not legislation meant to regulate failure away, but to protect people from force, theft and fraud. For these are the rules that bring market discipline in a system of prices, profit and loss.
Policymakers cannot “fix” our organic economic ecosystem, but they can create an environment conducive to beneficial decentralized coordination, and through coordination, growth, by implementing institutions that reduce transaction costs and by eliminating institutions that throw up barriers to such coordination and exchange. If we could find a way for the politicians to profit from doing that, instead of profiting from promising to “fix the economy”, then we would see true, resilient, meaningful economic growth out of this recession.
“Fixing” the economy is like getting your dog “fixed.”
I disagree with some word choices here.
I agree that no one I know of currently has the ability to model, predict, or guide the economy. However, your description makes two assumptions that are not necessarily true and that I don’t know of anyone proving.
By saying it can’t be done, you assume no model can be made of it. This is very different from the statement that none has been made. I am familiar with some of the innate complexity speculations, and the theories about systems that can’t be meaningfully represented by anything less complex than themselves. However, I don’t know of any results that show this must be true for economics. (If such exists, it does call into question the utility of studying economics.)
Also, the assumption that minimizing transaction costs using clear and transparent institutions is the best possible course of action is hard to support. It requires a faith in the invisible hand of the market place that does not have to be true. In fact, if as you assert the task of producing a useful model is beyond us, then it is no more likely that the invisible hand works than any other assertion about the proper way to interact with the market place.
In specific, the kind of localized self optimization of many different actors that is assumed in your idea to be the best possibility for optimizing the economy is effective in some classes of problems, and worse than random guessing in some others. (It’s worse than this. In the space of all problems, the subset on which this method outperforms random guessing is of measure zero.) Without showing that the economy fits into the efficient class, it is a leap of faith to just decide it does.
In this case, the folks who are trying to fix the economy aren’t doing anything that is less defensible than letting market forces rule. They are just making a different (and also unjustifiable) leap of faith.
Pingback: What to make of the fiscal policy discussion? « Knowledge Problem