I've always been perplexed by the cycles of merger and acquisition activity. The investment bankers who advise on these deals are some of the most highly paid people in the financial industry. They hold MBAs and other degrees from the world s most prestigious universities. They ve spent their careers analyzing markets and studying the results of mergers. And yet those advising the buyers in these deals routinely seem to violate the simplest proposition of successful investing, which is to buy low, not high.

Many pundits hailed the surge in M&A activity last week, when proposed deals totaled nearly $90 billion, as a bullish sign for the market. August is on track to be the best deal month of the year. Investment bankers and takeover lawyers are no doubt licking their chops in anticipation of the big fees and bonuses likely to follow, but is this anything for the rest of us to get excited about?

A look at the numbers suggest that deal volume is, if anything, a reverse indicator of market direction. Consider that the all-time record for merger deals in a single year $4.3 trillion, according to data-tracking firm Dealogic came in 2007. That s the same year the Dow Jones Industrial Average hit its all-time high. The feverish pace continued in 2008, until deal volume fell off a cliff after the collapse of Lehman Brothers and stock prices plunged. November 2008 marked the lowest level of M&A activity since 1995, when Dealogic started tracking deal volume.

Stocks bottomed in March 2009. So what happened to M&A activity in 2009, when stocks were undeniably cheap? Deals were few and far between, reaching a total for the year of only $1.3 trillion. The pace of deal making actually increased as stocks got more expensive.

Similarly, in 2000, when the Nasdaq hit its all-time high, global M&A volume reached $3.1 trillion, just behind the then-record of $3.13 trillion set the year before. In 2002, when stocks hit a post-bubble low, deal volume was only $1.27 trillion. (All M&A data are from Dealogic.)

So far my theory has been panning out, since the market has declined for four straight days in the wake of all the recent deals. And even though the market was off its highs for the year when those deals were announced, most of them were planned weeks or months ago when stocks were at recent peaks.

What explains this seeming paradox? I spoke to several M&A bankers who made these points:

For every buyer you need a seller, and high prices naturally attract sellers. (The same is true of real estate and art, where volume also surges at market tops.) More companies, especially desirable ones, are willing to put themselves up for sale when they think they re getting a good price as opposed to a fire sale. With the exception of the few overtly hostile deals (like BHP Billiton (BHP) bid for Potash), large deals usually depend on some degree of cooperation.

Companies looking for growth don t want to buy into a declining market. In March 2009, consumer sentiment was terrible, the banking system was fragile, and the economy seemed to be in freefall. The risks of buying seemed too steep. Many of these deals are leveraged, which magnifies the effects of declining revenues and profits. One banker told me the story of a chief executive who proposed a deal a year ago. One board member replied, Give me a break. You can t predict your next quarter, and you think you can tell me what another company will earn for the next ten years? From the evidence, corporate buyers and their sophisticated advisors are no better at calling market bottoms than anyone else.

Banks and other sources of capital are affected by the same factors. A year ago, no one wanted to lend. Even stable, blue-chip companies like Pfizer (PFE) and Merck (MRK) had to worry about securing funding for acquisitions. And when stock prices are depressed, stock transactions are unattractive. The target s stock price may be low, but so is the stock of the acquirer.

Wall Street and its clients are as prone to the herd instinct as anyone else. When CEOs gather at their country clubs and resorts this month, what do you think they ll be talking about? Deals. They and their boards measure performance by comparing themselves to their peers. And if everyone else is buying, who wants to be left behind?

Whatever the explanation, none of these reasons have any direct bearing on the future course of the stock market. When valuations are driven by M&A values as they were during the merger boom in 1987 and again during the leveraged buyout boom of the late 1990s-- multiples and prices can surge higher. But inevitably, there s a reckoning when prices aren t based on fundamentals like earnings.

My view is that deal volume doesn t amount to much of a crystal ball for the broad stock market. The only sure way to profit from merger activity is to buy shares in the advisors who stand to reap big fees. My sources all told me that deal chatter has picked up dramatically this summer. If the talk follows through, the big banks and investment banks are sure to benefit.