Not what they expected. That's the take on the stock market's sudden decline this August. The third year of a president's term is normally a good one for stocks — and history shows that when Congress is in recess, as it is now, markets almost always rally.

One product makes clear exactly how unusual this year's slide has been, and offers a clue as to why 2011 broke the rules. It's called the Congressional Effect Fund. Founded by Wall Streeter Eric Singer in 2008, the fund is premised on the idea that equity markets dislike a hostile Washington, tolerate a friendly Washington, but prefer an inactive Washington above all.

It follows that stock-market rallies would come most often when Congress is idled — in recess, at home, in the districts. From 1965 until early this summer, the Standard & Poor's 500 Index, Singer's proxy for stocks, rose 17 percent while Congress was out of session versus only 0.9 percent while Congress was working in Washington.

In one study, four scholars took a step back to look at a century of returns — from 1897, just after the Dow Jones Industrial Average was founded, to 1997 — and found that average daily returns when Congress was out of session were almost 13 times higher than when it was in. Their explanation: "Perhaps the market enjoys the temporary certainty exhibited by the absence of Congressional decisions."

Singer is blunter. About Washington's impact on the economy, he says simply: "Congress subtracts value." His fund seeks to profit from this dynamic by holding Treasuries and other neutral bonds during Congressional sessions, and S&P proxies, such as S&P Index futures contracts, when Congress is on holiday. From its inception in late May 2008 to June 30, 2011, the fund had annualized returns, net of fees, of 2.8 percent versus 0.5 percent for the S&P overall.

Singer has also tried to capture policy uncertainty and its relationship to equities through other formulas. Unified governments, in which the president and Congress belong to the same party, are ones capable of promulgating policy and then spending like crazy. The Congressional Effect theory would suggest that such governments are bad for returns, and split governments are good. It indicates that the benefit of public spending for certain sectors is outweighed by the damage caused by the arbitrariness of such spending.

The data bear that out, too. A unified Washington, whether Democratic or Republican, sees lower returns than a split one. We don't know that Congressional action is actually bad for markets; we know only that markets think it is.

There have been years when the Congressional Effect, as defined by Singer, hasn't occurred: 1978, 1981, 1997 and 2009. In these years, more market rallies, or fewer drops, occurred when Congress was in session than when Congress was out. These exceptions don't undermine the general thesis about stock owners being wary of state activity. One way government can intervene less than it has in the past is by reducing its own presence in market transactions. A capital-gains tax cut would be the most direct form of such decreased involvement. In two of those outlier years, 1978 and 1997, Congress and the White House promulgated cuts in the capital-gains tax. The 1978 reform, as some readers may remember, was the passage of the Steiger Amendment, which effectively halved the rate on such profits; in 1997, the Clinton-Rubin-Gingrich tax cut lowered the capital-gains rate to 20 percent from 28 percent.

That all brings us to the question: Why so bad, this August? One answer is that a storm of government activity is now doubtless forthcoming: more fiscal changes of an unpredictable form. The supercommittee recently assigned by lawmakers and the White House to plan legislation to resolve the nation's budget problems makes some sense as a practical matter, but its outcomes are deeply uncertain. And markets may be deeming the supercommittee just another example of Washington's capacity to dominate the rest of the economy.

Another explanation for darkening skies — admittedly, a speculative one on my part — might have to do with the Federal Reserve's Aug. 9 pledge to hold interest rates near zero through 2013. Some observers will argue that such a decision doesn't constitute "action." But the ever-rising price of gold and the dollar's new low against the yen last week suggest that some in the markets view the Fed's commitment as aggressively inflationary — that is, highly active, in that it will drive down the value of the dollar at home and abroad. A sharp decline in equity prices since the Fed's announcement also indicates that the central bank's move means, at least to some, a broadening of the range of possible interest rates — including extremely high borrowing costs — in the future.

The bottom line is very simple: The best thing Congress can do to help the markets is nothing at all. There are signs that even the Republicans, the party of "less government," have yet to fully internalize this concept. Presidential candidate Mitt Romney has lately been ripping President Barack Obama for taking a 10-day holiday on Martha's Vineyard; he argues that Obama should abort his holiday and hurry back to the Oval Office.

But the Congressional Effect theory suggests Obama is right to enjoy his island stay, and might even lengthen it.

(Amity Shlaes, a senior fellow in economic history at the Council on Foreign Relations, is a Bloomberg View columnist. The opinions expressed are her own.)

© Copyright 2011 Bloomberg