MICHELE NORRIS, host:
Economists use lots of fancy math to try to understand the economy. But beyond all those equations, the economy is real people making their own decisions. Many traders who buy and sell oil and other commodities make decisions using a method that's been compared to voodoo. Yet, as NPR's David Kestenbaum reports, the arcane practice somehow seems to work.
(Soundbite of people yelling)
DAVID KESTENBAUM: This is how we figure out the price of oil. In a cavernous room at the southern tip of Manhattan, men wearing bright colored vests stand inches from each other and yell.
(Soundbite of people yelling)
KESTENBAUM: It's the commodities market for crude oil at the New York Mercantile Exchange. Out of this buying and selling frenzy comes one of the miracles of the free market - a price. I expected the traders here to be experts in oil shipping schedules, the politics of distant places like Saudi Arabia and Azerbaijan.
But trader after trader told me, yes, they watch the news and all that. But there's an easier way to predict these markets. You print out the charts of where the market has been, the peaks and valleys. And draw lines off into the future. You use history to forecast what comes next.
Mr. RANDALL ROTHENBERG (Trader, Natural Gas): I actually use a protractor.
KESTENBAUM: This is Randall Rothenberg, who trades natural gas.
Mr. ROTHENBERG: There's a gentleman by the name of Gann who was a mathematician. So I use what are called Gann angles. From a particular edge, I draw angles and lines based on that on my charts.
KESTENBAUM: Rothenberg says sometimes, prices seem to be confined to channels. Other times, there seem to be magic numbers, support or resistance levels that function as ceilings or floors. The price will approach that number but then turn back. It turns out that all this has a name. It's called technical analysis. And according to textbook economics, it should not work.
Markets are supposed to be driven by the cold logic of supply and demand. If a hurricane is heading for the Texas refineries, the price of oil will go up because people think there might be a shortage. In a recession, like now, the price of oil drops because fewer people will be driving. Technical analysis has been compared to astrology and reading the entrails of a goat. And traders say, yeah, they've heard that, but it works. They don't know why, but it does. Anthony Grisanti, another trader, is president of GRZ Energy.
Mr. ANTHONY GRISANTI (President, GRZ Energy): There is something to these numbers. I mean, when a market gets to a number three or four times and reacts off that number, then it's a big number. So yeah, I don't think there is any analyst or anybody in the industry that would say just disregard the numbers.
KESTENBAUM: Which is the sort of thing that got Carol Osler interested. She's a professor of finance at Brandeis University.
Dr. CAROL OSLER (International Business School, Brandeis University): To economists who are used to thinking in market forces and grand theories, the notion that a sequence of lumps and bumps is going to predict a price is completely ridiculous. They think it's just like voodoo, and, I mean, I came at it with that impression initially.
KESTENBAUM: In the late 1990s, Osler and a colleague studied the currency markets, people trading the dollar, the yen, the mark. She says technical trading was really popular there. Reading the bumps in the market was a daily activity for many traders all over the world.
Dr. OSLER: I decided to start with the head and shoulders pattern, which seemed the most voodoo-ish.
KESTENBAUM: The head and shoulders pattern - a series of three bumps is supposed to be a bad omen indicating the price will drop. Osler and her colleague analyzed data going back to the 1970s and found, much to her embarrassment, that actually, it did work. The pattern of bumps was right about where the market was headed enough of the time that you could trade on that information and make money.
Dr. OSLER: My results said that this was a profitable trading strategy in the two biggest markets.
KESTENBAUM: And how big were the profits?
Dr. OSLER: How big were they? You know, I'd have to look at the paper.
KESTENBAUM: She looked it up after - somewhere from a 10 to 20 percent profit a year, better than average returns for the stock markets. She has no idea why this stuff works.
Dr. OSLER: When I first started looking into this, I read eight different technical analysis textbooks. And I was fascinated to look at the way the people who wrote them tried to justify this stuff. They used metaphors. So one of them was armies. OK, you know, they're slogging up the hill. And then the people at the top fight them back down. Another one was trains. I mean, it's just metaphors.
KESTENBAUM: One theory is that these patterns exist because we want them to be there. We've sort of willed them into existence. If people think the market will drop after a head and shoulders series of bumps, those people will sell, causing the market to drop. Andrew Lo is a professor of finance at MIT and author of the book, "A Non-Random Walk Down Wall Street."
Dr. ANDREW LO (Laboratory for Financial Engineering, MIT): Over time, these kinds of patterns ended up becoming almost self-fulfilling prophesies.
KESTENBAUM: So you sort of feel like, when you see these patterns in there, that's some kind - that's a manifestation of human psychology written into the pattern of the market.
Dr. LO: That's right. And, in fact, I think, if you look at the historical record of prices, I view that as a very rich fossil record of the different kinds of individuals that have engaged in these kinds of market dynamics.
KESTENBAUM: If you think of the market like a river, the overall flow, the big bends in the river may be governed by supply and demand, but the little eddies, the ups and downs, some of those might just be weird human behavior. David Kestenbaum, NPR News.
NORRIS: And you can find out more about how the small decisions we make drive the economy. Check out our Planet Money blog and podcast at npr.org/money.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.