I was an extremely unlikely candidate to write a book with a sensational title like "Dow 36,000"--a book that argues that "stock prices could immediately double, triple, or even quadruple and still not be expensive."

For six years ending this past June, I had written a financial column for The Post whose theme, if I could sum it up in two words, was: "Cool it!" Once and then twice a week, I would hammer home themes such as buy and hold, keep it simple, and history counts. I warned readers about getting caught up in the initial public offering and Internet frenzy (but still said high-tech stocks could be bought judiciously). I urged them to stay away from day trading. I leaned toward value stocks--shares that were cheap because investors shunned them--and harped on keeping expenses low and diversification high.

So what was a nice, middle-of-the-road, circumspect kind of guy like me doing as the co-author of a book that justifies price-to-earnings ratios of 100?

My conversion was evolutionary. Like everyone else, I had been watching the market rise and rise. There were occasional dips, but the general direction was up. Sharply. Stocks nearly quintupled in the 12 years ending in 1994. Then in each of the next three years--1995, 1996 and 1997--the Standard & Poor's 500-stock index, a proxy for the broad market, returned more than 20 percent, an unprecedented event in financial history.

And, while the market was rising, nearly all the pundits were saying that it couldn't last. Stocks were overvalued, or "fully valued." Typical was a Money magazine cover story of August 1997 with the headline: "Sell Stocks Now."

On the direction of prices, I was agnostic. As a believer in efficient-market theory, I rarely questioned the current level of the market, and I felt that its next immediate step would be part of a "random walk." In other words, tomorrow's movements are unpredictable based on today's information.

Anyway, it didn't much matter. If you were a long-term investor, the key was the incontrovertible historical fact that over the years stocks were returning an annual average of 11 percent. You can't get that with bonds or money-market funds.

Still, I was puzzled. The S&P's price-to-earnings ratio--that is, the number of dollars it took to buy a share of the average stock--was soaring, well above the norm of 15, up to 20, 25 and beyond. What on earth was happening?

While I was writing for The Post, my office was actually at the American Enterprise Institute, a Washington think tank, where I am a resident fellow. In early 1997, I got a new neighbor at AEI, Kevin Hassett, who had previously taught at Columbia University and served as a senior economist at the Federal Reserve Board. Kevin's specialties were tax policy and corporate finance, and he had published dozens of papers on these subjects in peer-reviewed academic journals, but his true love was the stock market, and we got to talking about the same concerns: Why was the market defying gravity? Why wasn't the traditional model for valuations, the one that says that P/Es can't exceed 15 or 20, working?

Almost immediately after coming to AEI, Kevin was diagnosed with cancer--Hodgkin's disease. He was getting regular chemotherapy treatments (which proved successful), and he had lots of time to think about our stock market puzzle. (He also took the opportunity to bone up on biotechnology and to invest very profitably in the industry's stocks.)

Early in 1998, our discussions finally reached a tentative conclusion. It went like this: The old model for valuing stocks had been repudiated by the facts of the past 20 years, much as Ptolemaic astronomy had been repudiated by new discoveries in the sky.

For our new model, we started at the very beginning, with the notion that the value of any asset is ultimately determined by the cash it puts in your pocket. That's how you price a rental condo or a bond or a dry-cleaning store. Shouldn't stocks be priced the same way?

The least understood fact about stocks--as I had learned in writing my column--was that, over long periods of time, they were no more risky than U.S. Treasury bonds. Both Kevin and I had been impressed with the work of Jeremy Siegel, a professor at the Wharton School of the University of Pennsylvania, who proved this point conclusively in his book "Stocks for the Long Run." We thought Siegel's research should be taken to its logical next step. If stocks and bonds are equally risky, then they should produce the same returns--that is, they should put the same amount of cash in investors' pockets over time.

But they didn't! Stocks produced far more cash. Something was out of whack, as if a $1,000 bond from one solid company paid $70 a year in interest while another bond from an equally solid company paid $120. So we asked a simple question: How high do the prices of stocks have to climb for their cash flow to be equal to the cash flow from bonds?

We did the arithmetic and found, much to our surprise, that stocks would have to triple or quadruple. P/E ratios could easily be 100.

We first made our views known on March 30, 1998, with the Dow Jones industrial average at 8782. The article was headlined, "Are Stocks Overvalued? Not a Chance."

The response came fast and furious. The financial establishment was much displeased. We were, after all, going right to the heart of what Wall Street analysts, journalists and scholars had cherished for decades--a set of rules about how to value stocks. "This stupid article does not make sense," Bob Brusca, then chief economist at Nikko Securities Co., was quoted by the New York Post as saying.

He was right. Our theory made no sense at all if you were stuck with the model that was dominating finance. But if you accepted our new model, it explained, as no other theory had, just what was going on in the stock market.

The price of a stock is determined by three things: the profits a company is making (and the rate those profits are growing), prevailing interest rates, and something called the "risk premium."

The first two elements really did not change much in the 1980s and 1990s. Profit growth has been roughly in line with postwar history. Interest rates bounced around, but the rates that count--the real, or after-inflation, rates--stayed about the same.

But the risk premium! That was where the change was occurring. A risk premium is the extra return that investors demand when they put their money into an asset whose returns are volatile--and, in fact, can be negative. For instance, to borrow money in the market, a shaky company has to offer high interest rates on its bonds--say, 10 percent, while the U.S. Treasury can borrow at 6 percent. The difference--4 percentage points--is a risk premium.

Stocks carry a risk premium, too, again in relation to the Treasury rate, which is a kind of benchmark. Calculating the premium involves some tricky math and some assumptions, but most economists believe that the historical risk premium for stocks over bonds has been about 7 percent. What has happened in the past two decades is that the premium has fallen to about 3 percent. A falling risk premium means a rising price.

To put it simply: Investors have become more comfortable with stocks and less worried about risk. Meanwhile, actual risk is falling because of better fiscal and monetary policy, globalization, peace abroad and better corporate management, among other factors we recite in the book. Investors are finally waking up to the fact that stocks are a terrific buy. So they are bidding up the prices.

Kevin and I believe this process will end when the risk premium is about zero and the Dow is about 36,000. At that point, stock returns will level off. Prices and dividends will still rise, but much more slowly.

Last summer, we signed a book contract. Immediately, the market tanked, the Dow falling to below 7500. But, to tell the truth, I wasn't particularly worried, and I said so in my columns. Stocks have since come back by nearly 50 percent, and I think they will keep rising, even though there will be downdrafts (buying opportunities) along the way.

We say in the book that the Dow should be at 36,000 today, but that realistically the process should take about five years. That's a growth rate of roughly 25 percent annually, not so outlandish considering the record since 1982. And, of course, the Dow is up more than 20 percent this year already.

Our theory is rooted in the idea that investors have become more rational. Economists have tried for a long time to understand what they call the "equity premium puzzle"--why is it that stocks return so much compared to bonds? Or, to put it another way, why is the risk premium so high?

Their answer is that investors behave irrationally. They see the short-term volatility of stocks and project it out to the long-term, like someone who notices that it has rained for three days and starts building an ark. Now, investors are getting more reasonable. While many market analysts think that investors are currently off their rockers, we think that, to the contrary, they are finally becoming rational--recognizing the true value of stocks for the long term.

Which brings me back to what I said at the beginning about calm and caution. In the second half of the book, we present strategies for taking advantage of the Dow 36,000 Theory.

Readers of my column will be comforted to find that the advice is not much different from what I have espoused for the past six years: buy and hold, look for value, stay diversified.

I also trot out my favorite quote from Benjamin Graham, the great mentor of legendary investor Warren Buffett: "The investor's worst enemy--is likely to be himself." In other words, what you do before and after you buy stocks is more important than the stocks you choose. For that reason, we urge most investors to get help from good brokers and investment advisers, more for hand-holding than for stock-picking.

Post readers will also note a few changes: I like growth stocks more than I used to; I think that the long-term investors should be in stocks for a minimum of only five, rather than seven, years; and I believe that older investors can have more stocks in their portfolios and fewer bonds.

In the book, as in my column, names are named. Among our favorite stocks are Cisco Systems Inc. (CSCO), the Internet infrastructure maker; Automatic Data Processing Inc. (AUD), payroll services; and Tootsie Roll Industries Inc. (TR), log-shaped candies; and Fannie Mae (FNM), mortgage finance. We also calculate what we call the "perfectly rational price"--or PRP--for many companies (for Fannie, it's $315 a share) and show how you can figure it out for yourself.

But despite the big numbers, the book does not stray far from the message I delivered during six happy years writing for The Post: Keep cool, keep invested. Even after the Dow hits 36,000, stocks will remain a wonderful place to stash your money.