Andrew Smithers told investors to dump stocks just as the market peaked three years ago. Thirty-five hundred

Dow

points later, he hasn't changed his tune.

In his book

Valuing Wall Street, published in early 2000, the founder of London-based advisory firm Smithers & Co. had some urgent advice for investors: "By far the most important thing you can do with your money is get it out of stocks. What you do with it then is a second-order problem."

It might be easy to suggest this was a lucky call, if not for the

q

ratio. First discussed by Nobel laureate James Tobin in 1969 and modified by Smithers and co-author Stephen Wright,

q

is a way to measure the difference between where stock prices are and where they should be. Looking back, it arguably has been the most accurate long-term predictor of market movements.

What does the

q

ratio -- and Smithers -- tell us about U.S. stocks today? Brace yourself.

Because

q

measures the value Wall Street assigns to stocks compared with the cost of actually creating those businesses (replacement costs), the proper

q

ratio in theory would be 1. As of Thursday's close of 868.52, the

S&P 500

had a

q

ratio of 1.50, meaning the index would have to fall about 33.5% to 577 to get to a

q

of 1. It would have to fall even further to revert to its historical average of 0.65.

According to Smithers, investors still don't grasp how overvalued domestic stocks are. "American investors are still in a bull market, psychologically. They are still trying to figure out how to make the most money, when they should be more interested in not losing it."

Based on his reading of

q

, coupled with his opinions on the economic malaise and the effectiveness of policymakers here and abroad, Smithers maintains that the most likely return for U.S. stocks over the next five years is very poor.

Investors, especially those with a shorter-time horizon, should be out of the market. "The economic growth rate is below trends, which means falling profits -- it's not that the economy will disappoint, but profits will," Smithers said. "This is not something American investors have taken on board."

Q and Its Critics

To explain how

q

works, and why some say it's outdated, we must define it.

q

is the number you get from dividing the total value of the stock market by total corporate net worth, as measured by the

Federal Reserve's

data on replacement costs. (Smithers & Co.'s

Web site has a page that can determine

q

for you.)

Applying this formula over the past century,

q

correctly signaled that markets were extremely overvalued in the late 1920s, undervalued in the late 1940s, overvalued in the late 1960s and undervalued in the early 1980s. In the late 1990s,

q

reached its highest levels ever, signaling to Smithers that the market was 2.5 times overvalued.

There are two sides of the equation for

q

: stock prices and corporate net worth. Because stock prices move much more rapidly than net worth, the history of

q

is one of reversion to the mean. In other words, when

q

is high, corporate net worth usually doesn't surge to catch up with prices -- stock prices fall.

If

q

has been so accurate, why doesn't everyone use it? The most notable criticism -- raised by, among others, Tobin, who died last year -- is that

q

weighs only tangible assets like plants and inventory, leaving intangibles such as goodwill, brands, human capital and intellectual property out of the equation.

In other words,

q

doesn't take into account things like the virtual monopoly of Windows at

Microsoft

(MSFT) - Get Report

or

Intel's

(INTC) - Get Report

microprocessor or the brand name of

Walt Disney

(DIS) - Get Report

. These intangible assets have become an increasingly important component of the U.S. economy, which makes

q

increasingly less relevant, critics say.

According to a 2001 study by Philadelphia Federal Reserve economist Leonard Nakamura, when adding in the trillions of dollars in intangibles, the bear market has driven the value of the stock market below its long-run equilibrium.

Smithers' reply: "The aggregate value of intangibles is zero." For all the "goodwill" out there that doesn't get added into the equation, there is an equal amount of "ill will," as Smithers puts it -- as demand for products falls off, products become obsolete and so on. "If people have been doing so wonderfully well with all these intangibles, why haven't profits kept rising?"

Beyond Q: Make Policy, Not War

As if

q

weren't enough to make Smithers gloomy, he also takes a somewhat dim view of the odds that monetary policymakers will save the day. Every major asset bubble during the past century has been followed be a major recession -- except, so far, the most recent.

On the economic and policy front, "things aren't looking too good at the moment," Smithers said. Echoing statements he made in a recent research report titled "Policy, Not Iraq, to Blame for Economic Weakness," Smithers said the popular conception that Iraq is the primary drag on the economy is wrong. "If the U.S. economy is growing below trend, it is going to bounce back above trend when Iraq is over."

While central banks, especially the Fed, have lowered interest rates substantially, a postbubble economy responds weakly to lower rates. Smithers said central banks should be pushing the government for another big short-term tax cut and enacting other measures to stimulate demand that will "give the world economy a big boost" during the next 18 months.

He said the Republicans in Congress are wrongly trying to sell a dividend tax cut as a stimulus, when it will, in fact, have a more gradual effect. And given the Senate's recent decision to hack at President Bush's proposed tax cuts in a time of war, the odds of a major tax-cut stimulus are slim.

Smithers said economies will underperform and keep profits depressed as policymakers are stuck in a period of "trial and error." He doesn't believe the Fed will suffer from the same "policy paralysis" that has bedeviled Japan's central bank. However, the political and economic climates make it unlikely that policy initiatives can keep the economy or the markets from weakening further.

What Investors Should Do

Smithers makes no claim to being a market timer, or a stock picker. He considers his area of expertise to be valuations, and based on those parameters, he is staying away from equities. He also said the U.S. and U.K. look "absolutely mad" -- which is to say, they aren't optimal investments. Over the next three years, "something will make money. Deutsche mark bonds may be a good bet," Smithers said. "But only hindsight will tell you."

The other suggestion he makes may seem odd, given that Smithers counsels mutual fund firms for a living. He suggests sensible investors shouldn't delegate their investments to individuals and firms that don't have the same interests. "You cannot expect your interests to be in tandem with fund managers who are required to buy shares when the market is overpriced," he said.

Does this mean investors should take all their money out of the stock market? Well, I have a very long-term investing horizon -- it'll be at least 30 years until I retire -- so I still believe stocks are my best bet for the long run. Nonetheless, I believe the U.S. market remains overvalued and I have lightened up from some areas that seem especially rich (U.S. large-cap growth, for instance).

Smithers, meanwhile, is heeding his own advice and staying away from equities. "I can't say that what I do is guaranteed to be the right answer," he said. "Nonetheless, these are still markets in which the wise man tries to preserve wealth rather than make it."

If you want to read more of Smithers' thoughts, check out his firm's Web site. His book is also available in paperback.