Another rough-and-tumble session on Wall Street ended with stock proxies well off their early lows thanks to a final-hour surge that had all the earmarks of month-end "window dressing" by fund managers.

The

Dow Jones Industrial Average

rose 0.7% to 8736.59 after having traded as low as 8537.10. The

S&P 500

gained nearly 1% to 911.62 after having traded as low as 889.82. The

Nasdaq Composite

closed off its low of 1307.01 but still ended down 1.2% to 1328.19.

Of course, there's no way to prove that money managers bought shares of their biggest holdings to help improve their monthly results, but it sure looked that way. Regardless, July was not a kind month for major averages, with the Dow falling 5.5%, the S&P 7.9% and the Comp by 9.2%. July was the fourth-consecutive losing month for the Dow, the first time that's occurred in 20 years, according to

Reuters

.

Today, the Dow and S&P were aided by strength in consumer names such as

Coca-Cola

(KO) - Get Report

and

Procter & Gamble

(PG) - Get Report

, as well as

Johnson & Johnson

(JNJ) - Get Report

and

Exxon Mobil

(XOM) - Get Report

.

The Comp was weakened by profit warnings from tech names

Nvidia

(NVDA) - Get Report

and

KLA-Tencor

(KLAC) - Get Report

as well as retailers

Hot Topic

(HOTT)

and

Children's Place Retail Stores

(PLCE) - Get Report

.

Beyond stock-specific news --

AOL Time Warner

(AOL)

and

Sealed Air

(SEE) - Get Report

were other lowlights -- the markets dealt with the day's disappointing economic data, which raised further questions about the timing of a recovery in corporate earnings and business spending.

The market's resilience to the data does support the notion that stocks put in some kind of bottom last week. However, the data similarly support the notion that stocks' recent swoon was forecasting some kinks in the robust economic recovery scenario, as

discussed earlier.

Valuing Valuations

A piece today about

price-to-earnings valuations drew

scads of email, much of it high-minded and thought-provoking.

Several readers suggested historic dividend yields should have been included in the table. For those interested, the information is available at Robert Shiller's

Web site. Also, the current dividend yield of the S&P 500 is 1.85% vs. the long-term average of slightly north of 4%.

Again, more evidence the current market ain't cheap by historic standards.

Still, much of the feedback focused on why we should not focus on P/Es exclusively (and I agree) or why high P/Es aren't necessarily a "bad" thing. Value managers, for example, prefer to buy cyclical stocks when P/Es are high, because it suggests earnings have bottomed, or soon will. "Think about it, does it make more sense to buy

General Motors

(GM) - Get Report

when it's losing $12 a share and on the verge of recovery or when they earn $6

per share at a peak," emailed one value fund manager.

Also, P/Es are not the only -- or even the prime -- metric money managers use to make decisions. Every industry has its own key ratios and I'm not just talking about industries without earnings; in REITs, for example, price-to-funds from operations is the key determinant. Also, EBITDA has historically been the key metric for media companies, although that's recently come under intense scrutiny.

Then, of course, there's the ongoing debate over what, exactly, "E" is these days and how it should be measured: forward or trailing, net or operating, core or reported, etc., etc.

Meanwhile, there's also the argument that P/Es are simply not a very valuable or accurate tool.

"By far the most common way to value a stock is by taking the product of the company's earnings and a P/E. This approach is as deeply flawed as it is widespread," Michael Mauboussin, chief U.S. investment strategist at Credit Suisse First Boston commented in a report yesterday. P/E ratios "add little or no insight into the investment process."

Mauboussin argued P/Es present a "circular case," since the earnings used in the ratio are also a determinant of the price. "The P/E doesn't determine value; rather, it derives from value," he wrote. "Price reflects future cash flows, which must incorporate growth, capital needs, risk, and sustainability of competitive advantage. These drivers are

implicit

in P/Es. But the P/E is not robust enough to guide investors when the expectations of these value-shapers change."

In a similar vein, Jeffrey C. Thomas, president of Alpha Asset Management, an Akron, Ohio-based investment advisory firm with over $50 million under management, suggested comparing P/Es to bond yields is also frivolous.

Prior to 1977, "there existed a negative correlation between bond yields and equity yields," Thomas noted. (A recent study by Smithers & Co., a Uw.K.-based investment advisory firm, came to a similar conclusion and found the negative correlation between bond and equity yields from 1948 to 1968 was "marginally stronger" than the positive one from 1977 to 1997.)

"I believe the last leg down in this bear market will be accompanied by the terror of portfolio managers realizing that Wall Street's equity valuation models are broken and not bent," Thomas concluded. In the future, "fair value" will be a "reasonably constant multiple"of S&P's core earnings, he forecast, rather than any function of bond yields.

Whether that comes to pass remains, of course to be seen. The above is intended so show how nothing on Wall Street is a simple as it may first appear.

For those looking for a "simple" conclusion, consider that none of the above changes the fact that the market's current P/E is far higher today then it was at past market troughs.

Aaron L. Task writes daily for TheStreet.com. In keeping with TSC's editorial policy, he doesn't own or short individual stocks, although he owns stock in TheStreet.com. He also doesn't invest in hedge funds or other private investment partnerships. He invites you to send your feedback to

Aaron L. Task.