in Musings


It’s amazing how much we trust “experts,” given how infallible they are sometimes.

In December 2000, the majority of investment banks forecasted that by the end of 2001 the dollar and the euro would be about equal in value. This list included Credit Suisse, Bank of Tokyo, RBC, UBS, and Deutsche Bank. The real exchange rate at the end of 2001 was only $0.88. Every bank overestimated.

In compensation for their overestimates the year before, the banks uniformly corrected their predictions downward. But the euro went up; the true exchange rate was 1.05, higher than any of the banks had foreseen. Surprised by the upward trend, the banks corrected their forecasts upward for 2003. Once again, the actual exchange rate was outside the range of estimates.

This continued on until 2010. Almost every year, the actual rate was outside the predicted rate.

Why do banks pay large amounts to entire departments for these meaningless predictions? For one, there’s an element of defensive decision making, where senior managers can point and say, “Well, this is what our mathematical models said would happen. It’s not our fault.” But two, there continues to be a large enough demand for these predictions that they’re supplied. Humans place an inordinate amount of trust in experts, and even desperately seek them out, so banks maintain the illusion.

The truth is anyone can become a market guru. Roger Babson is credited with correctly predicting the stock market crash of 1929, but what is less known is that he had been predicting the crash for years. Of course, no one remembered those misses after he was right once. Elaine Garzarelli predicted the stock market collapse in 1987, and four days later it really did crash. She became known as the Guru of Black Monday, but thereafter, her predictions about the market were right less often than a coin toss.

Warren Buffett often gives the following example: Consider 10,000 investment managers whose advice is equal to flipping a coin. After a year, 5,000 of them will have made a profit. The next year, 2500, and so on. After ten years, about ten managers will have gotten it right every single year. We would classify these ten people as gurus, attributing such a feat to their unique, deep insights of the market and some innate talent.

In my undergrad businesses school, we took a lot of complex finance classes teaching us the intricacies of portfolio optimization, where you have a chunk of money and want to invest it in a diversified portfolio. In the academic world, this involves fairly complex formulas and a host of assumptions to deal with. The professors promised us it would all come in handy on the job.

Well, it turns out there’s a one-line sentence that trumps all of these strategies, even beating a Nobel Prize-winning portfolio strategy. It’s this: Allocate your money equally to each of N funds. 

In a study, this rule of thumb was compared to a dozen complex investment methods. Seven situations were analyzed. In six of the seven tests, 1/N scored better than the others, and none of the other twelve were consistently better at predicting the future value of stocks.

The Nobel Prize-winning method isn’t a sham, it’s just that we live in a world of unknowns and uncertainties that can’t possibly be quantified in assumptions. Experts will tell you otherwise, but really, what is an expert?