Sunday, September 26, 2010

Socialism is Not the Issue; Central Planning Is

People accuse Obama of being a socialist, but that misses the point. The issue isn't socialism, per se. It's central planning. We have abundant examples from the Soviets and their satellites that central planning doesn't work. Central planning in the Soviet Union resulted in antiquated food distribution systems in which 25% of farm output was lost on the way to the centralized distribution center. It led to factories making nothing but left shoes in size 8. It led to the Soviet bromide, "We pretend to work, and they pretend to pay us."
My background is in software development. In software we have learned that creating big, centralized designs for complex large-scale systems is usually impossible. We've found that it works better to start with a small piece of a system, get that working, and then evolve the design of the larger system incrementally.
In software, compared to government, we have the advantage that our solution is "pure" -- our sole objective to make the system work. In politics, the solution is never "pure." It's muddled by various political calculations--attaching unrelated riders, funneling pork to the home district, cashing in favors, and so on.
Central planning of complex systems is simply beyond what we as humans are capable of, especially when the people who are supposed to be solving the problem are continuously sidetracked by issues that have nothing to do with the problem they're trying to solve.
Last Spring NPR ran two stories back to back that made an interesting comparison. The first story was about how the Obama administration was revising No Child Left Behind to give more latitude to the states. They had concluded that the issues involved in No Child Left Behind were too complex to plan centrally, and so they wanted to give the states more ability to optimize at the state level.
The next story NPR ran was about ObamaCare and how the Democrats were still hopeful that they could get the single payer provision put into the bill.
The juxtaposition of these two stories struck me. Is there anyone in America who thinks that medical care is less complex than education? If not, why does anyone think that we can centrally plan medical care if we can't centrally plan education?
The free market isn't perfect, but if you think of the issues as centralized vs. decentralized decision making, a thousand people making a thousand decisions that are locally optimal in their best interests is going to produce better results than 1 person trying to decide what's best for each of those thousand people from some centralized vantage point.
Absent that, at least pushing decision making down to the states on health care would make a lot of sense. Let's do some experiments at the state level, launch some pilot programs, see what works, and see where the problems are. Let's not standardize and bureaucraticze our ignorance before we even know what we're talking about. During the Obamacare debates there were many comments along the lines of "This won't work because New York tried it and it failed" or "This is a good idea because it worked in Oregon" or "This was proposed in Massachusetts but ultimately rejected." There were many comments about the efficacy of various state programs, but no one seemed to draw the obvious conclusion: The results of "experiments" with healthcare at the state level have been inconclusive so far. That conclusion does not imply that it's time to centralize planning for healthcare. It implies that we need to experiment more before we commit to any One Big Program.
The fallacy of trying to central plan complex entities is easy to see if you think of other complex entities. Think about trying to "plan out the internet" in a centralized way. Would a centralized planning body ever come up with Facebook, LinkedIn, Bloggin, or developing an iPhone to access all of that?
Going forward, I think one good litmus test for national government initiatives is:
Does the initiative require God-like omniscience to be successful? If it does, then has the sponser obtained a commitment from the Supreme Being to be directly involved in the initiative? If not, then we shouldn't do it.
None of this implies that government can't take an active role where it's needed. I think it comes down to where you want to start. Do you want to start with the free marketplace and decentralized decision making, and correct the problems that arise from that foundation? Or do you want to start with government-controlled markets/industries and centralized planning and correct the problems that arise from that? History shows that starting with the free market and correcting its excesses works best.

No comments:

Post a Comment