Don't believe the rosy forecastsDecember 30, 2010: 10:56 AM ET
Most economists and pundits predict a continued upward trend for most assets next year, but they will eventually be proven wrong.
In one of my classes at the University of Chicago Business School in the 1970s, the eminent statistician Harry V. Roberts liked to tell a story showing the homespun wisdom of his colleague and idol, economist Milton Friedman. The great monetarist was part of a panel evaluating a PhD presentation on forecasting growth rates for the U.S. economy. The candidate painstakingly described his methodology for fitting a broad array of variables––global capital flows, future exchange rates, immigration trends, and sundry other factors––into a computer model that, presto, spat out a prediction for the following year's GDP.
According to Roberts, Friedman delivered a brief, cutting critique––probably in the same nasal monotone I remember when, decades later, he returned my long-distance calls collect. Declared Uncle Miltie: "Why would this extremely complex model, based on factors that are themselves hard to forecast and could easily be wrong, produce a better number than taking the growth rates for the past five years, and dividing by five?"
For Roberts, the anecdote amounted to a parable on the pitfalls of economic forecasting. Friedman also liked to use the aphorism, "Predictions are extremely difficult, especially when they're about the future."
Friedman's lesson isn't that forecasting is impossible, but that the best prediction is usually the basic assumption that prices and growth rates will go back to their historic averages, or in economic parlance, "revert to the mean." What's difficult is guessing when that will happen. Indeed, the timing is truly unpredictable. But it invariably does happen.
Now, we're in the heart of the predictions season. The forecasts for the New Year from the pundits on Fox Business Network, CNBC, Bloomberg TV and dozens of websites vary a bit, but the overall message is overwhelmingly the same for most assets: stocks will remain on a roll, generating double-digit gains for 2011. The prices of gold, oil and other commodities will continue their upward march. As for bonds, rates will keep rising, but slowly. And almost no one has a good word to say about the housing sector, where prices have fallen for months, and according to the prediction mill, are destined to follow the same downward trend.
The problem with these predictions isn't that they rely on complex economic assumptions. It's just the opposite––they're really not forecasts at all, but extrapolations. The pundits are telling us that recent trends will simply keep rolling.
They could be correct, but only for a while. In the longer term, these forecasts will prove wrong, for a simple reason. Most assets are already selling at prices far above their historic averages. As economic gravity takes over, they'll inevitably return to those benchmarks -- meaning stocks, bonds and commodities have a long way to fall.
So let's briefly examine these assets, and look at the factors that govern their long-term value. For commodities, it's production cost. For stocks, it's the multiple of price to average earnings. For real estate -- the surprise in this package -- it's the cost of owning versus the cost of renting.
From 1983 to 2004, gold prices averaged around $400 an ounce. Today, the price for an ounce stands at $1390, or 3.5 times its historic average. The rub is that the production cost of gold is still in the $400 range. It's a combination of speculative demand and a temporary shortage of supply because of the long lags in mining the yellow metal that accounts for the surge. The spread of gold ATMs and ads luring people to sell their jewelry are sure signs of a bubble. Look for gold prices to drop sharply, and silver, copper and other metals to follow.
A similar scenario is threatening oil prices. The cost of production even in deep water wells is less than $60 a barrel, yet prices are over $90. Hence, the highest cost producer is earning 50% margins. We saw what happened when builders were making huge profits––they kept building until housing prices collapsed. Oil will follow the same pattern.
How about stocks? The best measure of the whether stocks are cheap or expensive is the price earnings formula devised by Yale economist Robert Shiller, which divides the current S&P price by a ten-year average of inflation-adjusted earnings. By smoothing earnings, Shiller avoids the error of judging that equities are cheap when profits are unusually high, as they are today.
Today, the Shiller PE is a lofty 22.7 -- that's more than 40% higher than its long-term average of 16. Indeed, stocks could keep rising for months or even longer. But that would make them simply more overvalued than they are today. In other words, a Friedmanesque reversion to the mean does not signal a rise in equity prices at all, but a sharp drop. The only question is when it will happen.
Let's move on to bonds. The sudden rise in yields on the 10-year Treasury from 3% in early December to 3.49% yesterday is a chilling reminder of the high risk to bond prices at these extraordinarily low rates. Despite the recent rout, the prices of 10-year Treasuries have plenty of room to decline. If yields return to their historic average yield of 6%, Treasury prices would drop by 20%.
Housing priced right
Believe it or not, the one asset that isn't overpriced, at least in most cities, is the most vilified: housing. The 30% nationwide decrease in values since the 2006 peak has brought the cost of owning a home to where it should be––back in line with the cost of renting the same house or condo. The fall in prices in most markets is over, or nearly over, and if it isn't, housing will become a better bargain with each downward tick.
That doesn't mean that housing prices will soar. But it does signal that homes are finally priced about right, and that it's an excellent time to move out of that rented flat and buy one.
The murky economy
It's extremely hard to forecast what changes we'll see in economic policy. Those changes may not affect the health of our economy this year or next, but they could greatly influence its future course. A prediction made by one of Friedman's heroes, the legendary Austrian economist F. A. Hayek, demonstrates the treacherous challenge of charting public policy.
In 1979, when inflation was raging and the Federal Reserve was deploying cheap money to battle rising unemployment, Hayek gave a speech declaring that the Fed could not be trusted managing the money supply. Hayek stated that it would always succumb to political pressure to use cheap money to create jobs, and that the effort would bring not prosperity, but greater inflation.
Hayek was wrong. Three months before Hayek issued his warning, Paul Volcker became chief of the Fed. Over the next several years, Volcker defied Hayek's predictions by reversing course and taming inflation.
So here are the best predictions for the New Year: Prices of stocks, bonds and commodities will gravitate towards their long-term averages––meaning the odds are they'll go lower. As for economic policy, if Hayek can be wrong, why try to forecast the truly unpredictable?
Also on Fortune.com: