Learning to Love the Bubble

I just got back from vacation, where I did a lot of reading. (You can check that out here, if anyone cares)

I've been taking a break from explicit economics and politics books for a little while, since the debate has been feeling a little stale since the election. I think a lot of people are feeling the same way about domestic issues and fiscal issues in particular, and I don't think that it will last very long. But despite my efforts, I keep getting drawn back into that old aphorism: you can talk the boy out of the economics, but you can't take the economics out of the boy.

So, you can't imagine my delight when, after going through books on the evolution of technology and nonlinear systems, I stumbled across this gem, a conversation with "practicing theoretician" Bill Janeway. While the speech covers some liberal standards like the role of government, it also has some wonderful points about financial instability, speculation and complexity.

Here's the basic message: an economy is structured around its technologies.* That technology is supported by basic scientific research. Since much of that research does not show any short- or medium-term economic function (number theory took 55 years to lead to Google), some major institution, either the government of state-supported monopolies, has to provide it. While this kind of research is a public good, it can be considered relatively wasteful from a short-term, profit-oriented economic standpoint. There's not many opportunities for profit-seeking firms in competitive markets to make this investment.

But without these investments in fundamental research, technological growth stalls out.** We're seeing some of that right now, since we've become stingier and stingier about fundamental research since Reagan. During the Bush years, public funding for research universities declined by 20 percent. The current fiscal environment is obviously not helping anything.

Everyone here has heard this argument from me in the past. It's standard liberal boilerplate. But here's where this starts to get interesting.

The process of translating fundamental research to technologies, i.e. the basic economic process, is inherently nonlinear. An apt metaphor is phase transitions in a thermodynamic systems. For example, we can look at a piece of ice and say that it's going to melt, but we have incredible challenges when trying to predict the physical structure of the puddle of water that will emerge once that ice melts. What's more, basic substances exhibit all sorts of surprising behavior during phase transitions. Hydrogen is probably a superhot liquid metal on Jupiter. At certain levels of pressure and temperature, water (normally a non-conductive substance on its own) has incredibly high energy densities and conductive properties.

We can say similar things about entrepreneurship and its role in the greater economy. We know that technology drives economic progress, and we know that technology is essentially the application of fundamental scientific research to solving human problems. But it is nearly impossible to say what form that research will successfully take. It was not possible, in the 1960s, to see how cryptography will translate into America's largest online retailer, but that's exactly what happened.

Because of this fundamental uncertainty, economies must depend on speculation in order to spur the economic process of innovative entrepreneurship. This process is chaotic. Three out of four startups fail. Those that succeed offer returns that follow power laws instead of any sort of normal distribution. As Jack Altman explains, almost all of positive returns in the venture capital industry is concentrated among of very small number of firms. As he writes,

Venture capital is one obvious manifestation of power law distributions, but we see this phenomenon all over. In industries susceptible to the “superstar effect”, such as sports, movies, or politics, the outcomes of top performers tend to follow a power law curve. The best baseball player makes a lot more than than the 100th best player, who makes much more than the 1,000th best player. “A-list” actors do dramatically better than “C-list” actors, who do dramatically better than struggling artists in Manhattan. The president has much more power than senators, who have much more power than local officials, who have much more power than me.

When you take your time to think this concept, one conclusion is inevitable: bubbles are a natural part of financial activity. We've seen that in countless economic experiments already. Anytime there's uncertainty, people will become speculative. When they speculate, group psychology and feedback loops will cause asset prices to dramatically escalate from "fundamental value." People will fixate on their ability to buy and sell with the momentum, "flipping" assets instead of investing in a traditional buy and hold pattern. In fact, the appearance of bubbles are likely a necessary feature of reaching any sort of price equilibrium in the market at large. People need to get burned before they learn.

What's more, bubbles are an important and healthy part of the economy. They encourage innovation, as they flood emerging fields with venture cash. Even after they burst, the innovation they spurred remains in place, and the few companies that win the bubble race become mainstays of the economy for years to come. This includes Amazon and Google from the tech bubble, just as RCA, GE and IBM were important components of the postwar US economy. In both historical cases, speculation helped create the foundation of longterm growth.

Of course, this isn't meant to be an apology for all bubbles, as some clearly threaten the financial system more than others. But the inescapable fact is that speculation is an essential part of all market economies. We should be designing an economic system that embraces shocks instead of one that avoids them. In other words, we should be seeking ways of making economies robust. Recent research from the Fed mines this field, George Soros generated a lot of interest talking about something similar, and there's a field forming around "complexity economics" that essentially advocates the same thing.

Although complexity economics has been mostly hype for awhile, it has a persistent habit of reasserting itself into the national conversation. Maybe conditions are right for a better understanding of the economy as a dynamic system, with policies designed to encourage dynamism instead of surpress it. If not, at least complexity serves as a welcome break from the monotony of the current national economic debate.

*The details of this can be found in The Nature of Technology by W. Brian Arthur.
**This is a riff on Tyler Cowen's argument in The Great Stagnation.