Early last week I was attending the annual North American meetings of the US Association of Energy Economics. Some of the plenary speakers and papers presented focused on the need for technological innovation, particularly in technologies that will reduce the use of fossil fuels for transportation. Some people argued for the political process to take the lead in funding and guiding such innovation. Indeed, this line of argument is one reason for the existence of the Department of Energy. This is a question I mull over frequently, and I usually come down to the conclusion that governments and political processes do a very poor job of picking technology winners, and thus that a dirigiste government technology policy is more likely to waste taxpayer money.
I thus read Mike Munger’s recent EconLib column with great interest. Mike introduces a great insight into the question of government technology policy: political processes work relatively well when the decision being made is uni-dimensional.
For political decisions, “good” simply means what most people think is good, and everyone has to accept the same thing. In markets, the good is decided by individuals, and we each get what we choose. This matters more than you might think. I don’t just mean that in markets you need money and in politics you need good hair and an entourage. Rather, the very nature of choices, and who chooses, is different in the two settings.
Put another way, political processes constrain us to sharing a common outcome, dictated by the median voter’s preferences, while market processes allow for a plurality of outcomes to coexist simultaneously. That’s why markets (and technology) have led to viable long tails, while politics does not.
What does this have to do with innovation? Mike illustrates with a story of the two Steves:
In 1976, there were two Steves, in a garage. One worked for Hewlett-Packard, and the other for Atari, in California. The Atari guy, Steve Jobs, had the garage, and the two Steves worked on a revolutionary machine. Or so they hoped. Near the end of 1976 the Steves, Jobs and Wozniak (the HP guy), released into the market a metal box attached to a TV screen, and tried to charge $666.66 for each one. The contraptions were called Apple I’s.
Suppose we had taken a vote at that point on this “personal computer” thing. Most people would have guessed that there was no future in these silly, overpriced boxes. ($666.66 is more than a satanic pricing point; it is also $2,350 in 2006 dollars. This for a paper weight with no fixed memory and 16k of RAM.) No government agency would have funded the thing, unless Messrs. Jobs and Wozniak were also big campaign contributors. And it is unlikely that there would have been a grass roots wave of support, since no one even knew who the Steves were or what they were trying to accomplish.
What if we don’t all want the median? That’s the fundamental question to ask with respect to innovation policy. Government policy is not good at tolerating heterogeneity of either individual preferences or outcomes. Furthermore, such innovation primarly occurs through the vision, creativity, and opportunism of non-median thinkers like Jobs and Wozniak. I think the common-median dynamic of politics and the nonconformist-innovator dynamic of markets are fundamentally irreconcilable. Mike’s excellent essay probes that insight and brings it alive.
Key quotes from his conclusion with respect to innovation, although I recommend your reading the whole thing:
[C]an I decide, and risk just my money, for great reward if I’m right? Or will we decide, and risk our whole future budget, on things we aren’t very good at deciding? …
The thing to keep in mind is that market processes, working through diverse private choice and individual responsibility, are a social choice process at least as powerful as voting. And markets are often more accurate in delivering not just satisfaction, but safety. We simply don’t recognize the power of the market’s commands on our behalf.