Greenspan, You're No Bill Gates

Testifying yesterday before the Financial Crisis Inquiry Commission, former U.S. Federal Reserve chairman Alan Greenspan defended his 21-year tenure at the Central Bank, saying it wasn't he who inflated the housing bubble that caused the Great Recession.

While mistakes were made, Greenspan admitted, he claimed to have been correct 70 percent of his time at the Fed, which sounds pretty good. The unanswered question in all this finger-pointing and next-day quarterbacking, though, is whether being right 70 percent of the time is good enough.

It isn't.

That's not just my opinion. It's a central conclusion of the 1977 Stanford University computer science dissertation of Charles Simonyi, longtime head of Microsoft's application group, the father of Microsoft Office, and now a celebrated space tourist.
Simonyi's paper on what he called "metaprogramming" -- technical process management by a strong central authority -- proved mathematically that a central manager of a complex technical process could be successful as the sole decision-maker only if the metaprogrammer was correct at least 85 percent of the time.

As a strong Fed chairman, Greenspan gleefully assumed the role of metaprogrammer for the U.S. economy.

But seventy percent correct, while it may sound good, actually isn't good at all, as Simonyi proved. Seventy percent is terrible. What's amazing is that Greenspan still doesn't understand this today.

This knowledge was available during the entire period of Greenspan's time at the Fed and many influential people read the paper.

After reading Simonyi's dissertation, for example, Microsoft founder Bill Gates hired the Hungarian to work at Microsoft where together they rigorously implemented the metaprogrammer system. Thus metaprogramming was proved not only mathematically, but also in practice, where it created thousands of Microsoft millionaires, made Gates the richest man in the world, and turned Simonyi into a billionaire with a $40 million one-bedroom house.

Any would-be metaprogrammer who couldn't make the right call at least 85 percent of the time, Simonyi calculated, would lead to an unstable system that would eventually -- and inevitably -- fail.

This doesn't prove, by the way, that the Fed chairman-as-metaprogrammer model (perpetuated today by current chairman Ben Bernanke) was bad. It just proves the Greenspan was the wrong man for the job.

Too bad Greenspan (and the various Presidents he served under) didn't know any of this.

If only they had called me.
Read Full Story

Find a home

Powered by Zillow