T.J. Rodgers, Man on the Margin
Jude Wanniski
April 11, 2001


Memo To: T.J. Rodgers, CEO, Cypress Semiconductor
From: Jude Wanniski
Re: ĎA Computer Would Do Better Than the Fed'

It is no surprise to me that you wrote Saturday's NYTimes op-ed, "A Computer Would Do Better Than the Fed." Of all the chief executives in the computer industry, you have been the most interested and cerebral in thinking about the economy. Bill Gates may have many more billions than you in his net worth, but he never seems to have given a thought to the world of public policy. You keep popping up now and then with ideas about taxes or trade that always seem to have a seminal twist. Remember I called you some years ago to congratulate you on some other op-ed you'd written, although I cannot remember what it was. Your idea in the Times, that we should be able to program a computer to manage "the discount rate" instead of having the Federal Reserve board of governors do it, was inspired. I agree with you completely. Except I would do it with the most powerful computer known to mankind, which is mankind itself. You may be able to design a computer that could outwit the world chess champion, but you could never, ever design a computer with the power of all the people in the world who are able to express their opinion on Federal Reserve policy through the marketplace.

Mankind in the market could do this when our central bank was on a gold standard, because it was given a means of communicating with the Fed, via the dollar/gold price. As long as we were on the gold standard, the Federal Reserve could not make a mistake at the end of the day, when it came time to balance its books. Alexander Hamilton had made the same argument you made in a different way when, as our first Treasury Secretary, he told Congress that the dollar might be managed in an okay way by several wise men, as long as we were at peace, but whenever there were threats to the nation requiring an increase in taxes, it would be so much easier to print money than to raise taxes that the system would break down. With gold, he said, the bank always knows when it is printing one dollar too many, because it will show up somewhere in the economy and the person holding it will come to the bank and ask for gold. That little-bitty signal is all the Fed needs to know at the end of the business day. If there is nobody at the gold window with surplus dollars and nobody at the dollar window with surplus gold, you know you are perfect in your management of the dollar.

Did you ever hear the story about Francis Galton, the Victorian genius who hit upon the idea of a regression to the mean? Galton, a cousin of Charles Darwin, made dozens of important discoveries, including the use of fingerprints to track down criminals, but none was more important than his discovery that large numbers of people can act like modern computers. It was Peter L. Bernstein, the old Wall Street guru, who told me about Galton going to a London fair and noticing a bull on display, with a prize given to the person who could come closest to guessing its weight. After the event, he asked for the slips of paper containing the guesses, added up the number of pounds guessed and divided by the number guessers. He found that the number was exactly the weight of the bull. Nobody had guessed it exactly, but a large number could get it exactly right. Jack Treynor, who taught finance at USC, tells me he would get his students to guess the number of jelly beans in a jar, with the same result. But he would agree with you that you will not get the same result if you ask the students to take a second guess, after they have heard the first guesses, without hearing the answer. The first guess is superior. Treynor meant this as a proof of the financial market, but I saw it as a proof of the political market as well.

The idea goes way back, I subsequently found. As George H. Sabine, the late, great historian writes in his History of Political Theory: "It is possible to argue, Aristotle says, that in the making of law the collective wisdom of a people is superior to that of even the wisest lawgiver... Men in the mass supplement each other in a singular fashion, so that by one understanding one part of a question and another part, they all together get around the whole subject."

Finally, you can check with Paul Volcker, who was chairman of the Fed from 1979 to 1987, the period of great monetary inflation and deflation, when gold climbed from $240 to $850 and back down to $350. (It is now $257.) At one of several meetings I had with him in that period, I told him he would do better if he added liquidity when the gold price began to fall and subtracted liquidity when it began to rise, letting the market do his work for him. Volcker stared at me for a moment and said, "You want to turn me into a robot." Yes. "Exactly," I said. "Exactly."