Former Federal Reserve chairman Alan Greenspan remembers the stock market boom under his tenure fondly. He wrote in his 2007 memoir: "People would stop me on the street and thank me for their 401(k); I'd be cordial in response, though I admit I occasionally felt tempted to say, 'Madam, I had nothing to do with your 401(k).' It's a very uncomfortable feeling to be complimented for something you didn't do."

For a man never short on ego, he may have been underestimating himself.

In one of the most startling studies I've ever seen, researchers from the Federal Reserve this week measured how much the stock market is influenced by...the Federal Reserve. Their conclusion: "Since 1994, [stock] returns are essentially flat if the three-day windows around scheduled FOMC announcement days are excluded."

The Federal Reserve announces what it's going to do to interest rates eight times a year at Federal Open Market Committee meetings. These are scheduled in advanced and well-publicized, so investors know exactly when the goods are coming.

Since 1994 (when the Fed started publicizing its moves), the S&P 500 has risen from 450 to 1300. But remove the 24 hours just prior to FOMC announcements, and returns fall to almost nothing:

Source: Federal Reserve.

The researchers note: "More than 80 percent of the equity premium on U.S. stocks has been earned over the twenty-four hours preceding scheduled Federal Open Market Committee (FOMC) announcements."

Might some of this be a coincidence? Highly doubtful. My colleague Matt Koppenheffer took market data going back to 1994 and randomly removed 136 days (eight per year for 17 years, or one for every FOMC meeting). The difference, compared with the unmolested market returns, was trivial. We ran the simulation several hundred times, removing different sets of random days. Nothing came within a third of the skew caused by removing the days shortly before FOMC meetings.

This doesn't mean Fed policies have been responsible for the majority of the market's rise since 1994. After-tax corporate profits have increased fourfold since 1994, or more than double after inflation. Cellphones in 1994 were brick-like boxes, less reliable than carrier pigeons. Today you can stream live video on an iPhone. That's real innovation and progress that accrues value to shareholders. The Fed has played an enormous role in boom-and-bust cycles and has inflated asset prices for decades, but it's hardly the only driver of long-term market returns.

And when the researchers looked at the returns of bonds and currency exchange rates -- whose values should be more sensitive to Fed policy -- in the day before FOMC announcements, they didn't "find any differential returns ... on FOMC days compared with other days." That's not to say Fed policy didn't inflate the values of those assets -- indeed, almost by definition, it did -- but the effects weren't concentrated into a single day like they were in the stock market.

Why is that? I think it's a good example of how short-term and headline-driven stock markets behave. The stock market has been dubbed "the bond market's idiot kid brother," which seems appropriate here. Back in the '90s, CNBC hyped what it called the "briefcase indicator," using a live video feed of Alan Greenspan's morning walk from the Fed's parking lot to his office to gauge the thickness of his briefcase. If it was thick, it meant he was studying hard and preparing to make big changes. If it was thin, his mind was clear, and interest rates were likely to be left alone. This went on for years before people realized how ridiculous the theory was.

But the amount of short-term, carnival-barker thinking that permeates markets has, I think, gotten progressively worse over the years. Bond markets tend to process information in measured, calculated ways. The stock market, unable to think past next Tuesday, takes the binge-and-forget approach. Over the long haul, stocks are driven by business fundamentals, but any given day's movements are almost entirely headline-driven.

Just after I read the Fed's report, a news article crossed my screen. "Nearly two-thirds of the 30 investment experts surveyed by CNNMoney agree that the Fed's latest move to extend its so-called Operation Twist policy was warranted," it read.

Can you blame them?

Check back every Tuesday and Friday for Morgan Housel's columns on finance and economics.