Which of the three major Dow Jones averages, the Industrials, the Transports or the Utilities, has provided investors with the greatest total return so far this decade? The Industrials have lost you money at an annualized rate of -0.0596%, the Transports have made you an annualized 4.1972%, and the Utilities have made you an annualized 9.9918%%. How does that rank in the no-contest department?

The returns so far in 2005 have been even more lopsided: Utilities are up 16.54%, and the Industrials and Transports are down 4.45% and 7.62%, respectively. Such a disparity in returns might be explicable if we were in some sort of bear market flight-to-quality in stocks or if the two principal costs of utilities, fuel and money, were falling. Neither is the case.

In fact, the outperformance of utility stocks has to rank alongside

the growing contango in crude oil and the bullish flattening of the yield curve as one of the great conundrums in this too-confusing market environment.

Let's switch index providers from the aggregated and price-weighted Dow Jones Utility index to the S&P utility indices. We can break these down by market capitalization to the large-cap

S&P 500

, the mid-cap S&P 400 and the small-cap S&P 600. We can also examine electric, gas and multibusiness utilities separately.

For the sake of completeness, Standard & Poor's also has water utility subindices, but none of these were statistically significant for the story to follow.

Triumph of the Counterintuitive

Now let's return to an analytic technique first introduced here in

February and revisited several times since, the calculation of relative industry group betas, or measures of relative volatility, against a set of primary market factors. This analysis will tell us, for example, whether the relative performance of a group such as the S&P 500 electric utilities to the S&P 500 itself has a statistically significant relationship at the 90% confidence level to a market factor such as natural gas. A table of results is presented below.

The results are as astonishing to the statistician as the discovery of flying pigs. Let's take the line item that does not have any entries, but should: natural gas. Because natural gas prices are a primary cost for electric utilities and it is not much of a stretch to say they should be significant for gas utilities, we should expect these groups to have some effect on their relative performance. None exists.

A related and perhaps greater surprise is the set of positive betas to crude oil. The results are unequivocal: The more crude oil prices rose, the greater the outperformance of all sizes of gas utilities, the S&P 500 electric utilities and the multibusiness S&P 400 and S&P 600 utilities.

It gets ever stranger when we move to financial variables. The flattening yield curve is the feature thrill ride at Conundrum World, and simple financial theory says that all debt-heavy firms should see their margins pinched as short-term financing costs rise. But the line for the forward rate ratio, a measure of yield curve steepness, is replete with positive betas. This indicates the various utilities' relative performance rose as the yield curve flattened, even as the line for two-year note yields is filled with negative betas.

Conclusion: Even though utilities were hurt by rising short-term yields, the drop in long-term yields -- note the negative beta entries on the 10-year note yield line -- helped them even more.

The positive betas to falling long-term rates make sense for debt-heavy firms. But that should mean they benefit from rising inflationary expectations as defined by the yield spread between regular 10-year Treasury notes and TIPS. No, the betas here are negative for the electric utilities. These are surely the only asset-rich major bond issuers hurt by rising inflationary expectations in recorded history.

As they say on the late-night TV ads, "But wait, there's more." You should expect rising volatility as measured by the VIX to benefit high-octane technology and industrial stocks more than the allegedly staid utilities. No, all of the betas against the VIX are positive.

Copper prices are an excellent proxy for industrial activity; can we explain the strong performance of utilities by this measure? No, the betas for the mid- and small-cap electric utilities are negative.

Credit to the Rescue?

Well then, can we throw all of the above out the window as statistical claptrap and focus on utilities' business quality? Let's return to a measure introduced here in

May, the cost of credit default swaps for a given economic sector. The logic, as stated at the time, is that stocks can get hammered without affecting the corporation's bonds, but if a corporation's bonds are in trouble, the stock will get dragged lower except in the case of a leveraged buyout or other corporate event.

If we map the cost of a weighted average five-year CDS index for the S&P 500 utility sector plotted inversely against the price of the

Utility Select Spyders

(XLU) - Get Report

, we see how this principle held between May 2003 and March 2005. As CDS costs declined, the stock rose; jumps in the CDS costs in periods such as August 2003 and June 2004 halted the stock's rise.

The CDS index reached its narrowest point in March 2005 and nearly doubled going into May. It is now failing to narrow further. What has the XLU done as the bonds started flashing red? Why, it has kept going up.

Utility Credit Risk Rising

Click here for larger image.

Source: Bloomberg, Howard Simons

The battle cry, "It can't go any higher!" usually emanates from the throat of someone about to lose a lot of money in a short period of time. If the stampede into utilities continues, it will do so without me trying to stop it -- or participate in it further. Year-to-date returns in excess of 20% for the likes of

TXU

( TXU),

Edison

(EIX) - Get Report

,

FirstEnergy

(FE) - Get Report

and

AES

(AES) - Get Report

are nice to have but hard to duplicate, especially when all of the ancillary indicators noted above are flashing red.

Not everything that goes up is a bubble. But any asset whose price becomes driven by money chasing performance hits bubble status sooner or later. Utilities, which then included Enron and its camp followers such as

Dynegy

(DYN)

,

Reliant

(RRI)

and Mirant, crashed very hard between 2000 and 2002 from levels not too far over where they are now. The time to buy the index was after that crash, not after this rally.

Howard L. Simons is president of Simons Research, a strategist for Bianco Research, a trading consultant and the author of

The Dynamic Option Selection System. Under no circumstances does the information in this column represent a recommendation to buy or sell securities. While Simons cannot provide investment advice or recommendations, he appreciates your feedback;

click here

to send him an email.

TheStreet.com has a revenue-sharing relationship with Amazon.com under which it receives a portion of the revenue from Amazon purchases by customers directed there from TheStreet.com.