Buy (Some) Utilities
How do you capitalize on the country's insatiable appetite for electric power? Begin by identifying the producers that don't burn lots of expensive gas.
It's a financial crisis to rival the real estate bubble of the 1980s that bankrupted 750 thrifts. About $150 billion of debt paper written by electric utilities and independent ("merchant") power-plant operators is now trading at deep discounts. One in ten utilities now ranks as a junk debtor, half are almost there, and the Dow Jones Utility Index closed out 2002 at a seven-year low.
So now is a time to buy utilities. But buy only the ones that have steered clear of speculative foreign investments, Enronlike trading debacles—and natural gas.
Four years ago we estimated that the manufacture and operation of semiconductor circuits and datacom networks already accounted for well over 10% of our national electricity consumption, and we predicted that digital demand would propel rising electric use for a long while to come. It was a surprising and controversial assertion, but subsequent research has borne out our conclusions. (For links to critiques of our study and to our rebuttals, see the online version of this article.)
With half of all U.S. power generated by coal, FORBES titled our story "Dig More Coal, the PCs are Coming" (May 31, 1999). Depending on whether you believe government's numbers or the industry's, electricity consumption is up between 7% and 10% since then—despite a recession that sharply depressed industrial output. Stock prices of pure-play coal companies have risen about 20%, while the DJIA was dropping 10%.
Why didn't power companies thrive during this period of rising demand? Because many of them, instead of investing capital in their coal plants, got caught up in a wild scramble to burn more gas. Utilities had built far too few plants in the early 1990s, and the cushion of excess capacity had fallen dangerously low (see chart below). Regulators then opened up interstate transmission lines to competition. New coal plants take at least five years to build, and so, faced with an unexpected shortage, utilities and the new merchants all rushed to build new gas-fired plants, which take only six months to two years. Gas now generates 19% of our power (consuming roughly a third of the country's gas) and accounts for 35% of generating capacity. Unregulated merchants now supply about a third of the nation's power.
Gas-fired power, however, is two to four times as expensive as coal or nuclear. The wise course of action is to use gas to cover just the spread between baseload demand and peak demand (typically, about 15%) and to provide another 10% as a reserve against unexpected outages. Then use other fuels to cover the baseload. That minimizes the combined cost of fuel (high for gas, low for coal) and capital (high for coal, low for gas). But the market has swung far past the optimum mix. Today's generating capacity glut is almost all in plants that burn expensive gas.
Gas prices aren't coming down. Domestic gas production has remained virtually flat since 1996, and while Canadian imports are up, gas can't easily be imported from any farther afield. Capacity limits in gas pipelines are putting further pricing pressure on gas-fired power.
So there's now a huge opportunity to generate power under the sky-high umbrella set by gas. Old plants can be juiced up like hot rods, and that's now happening. Some 80 gigawatts—about 10% of current total U.S. capacity—will be squeezed out of existing coal and nuclear plants in coming years.
Improved maintenance and management allowed top-notch nuclear operators like Entergy and Duke, over the last decade, to raise the run time of their plants from about 60% of the hours in a year to 90%. Nuclear plants are so overbuilt that they can also readily be run 5% to 15% hotter. Federal safety regulators approved 17 such "uprates" in the last year alone and will process another 30 to 50 applications in the next five years. Exelon uprated two of its big nukes last year, eight in 2001.
Coal operators have pushed their run times up about ten points (to around 70%) in the last decade; another ten points is still achievable. Refurbishing plants with new furnaces, boilers, and turbines will add even more power to the grid. A few years ago Detroit Edison was set to replace turbine blades at its Monroe, Mich. plant—a fix worth about 70 megawatts of new power, with no increase in fuel consumption or emissions—but the Clinton Administration treated rebuilt plants as "new sources," requiring prohibitively expensive new scrubbers. New rules recently announced by the Bush Administration won't. Those changes will be worth about 40 gigawatts of additional capacity over the long run, if they survive in the courts.
Even oil is beginning to look interesting again. Warren Buffett put $1.5 billion into Center Point and TXU—whose older gas-fired facilities can also burn oil. Keep in mind, also, that the country is divided into ten major—and largely autonomous—grids, and conditions vary considerably from region to region. The Mid-Continent Area Power Pool (MAPP) grid in the Midwest, for example, is still seriously short of peaker capacity, so it makes sense for a coal utility like Wisconsin Energy to invest in two new 545-megawatt gas-fired units.
Better wires will facilitate further shifts from peak capacity to base. Here again much of the action is in upgrading existing facilities. Last year New York's Marcy substation was outfitted with silicon switches that boosted the capacity of existing wires by 240 megawatts. To be sure, parts of the utility business are subject to regulated, cost-plus pricing. But the deregulated fraction is growing, and a good portion of the gains to be made from upgrades of old plants will flow into bottom lines.
The portion of gross domestic product fueled by electric power passed 50% around 1978; it's 65% today. All-electric information technology now accounts for somewhere between 25% and 60% of GDP growth, and the fraction continues to rise year by year. Domestic electrons have no shelf life, they face no off-shore competition, and barriers to entry in the baseload market are substantial. Utilities and merchants—the ones with long-term horizons, coal and nuclear capacity, and a stick-to-the-knitting business philosophy—are going to prosper.
Looking Beyond Natural Gas
Here are six coal-and nuclear-rich utilities whose stocks are worth buying. These fuels are reasonably priced, and not subject to the volatility of gas. While most of the companies listed have gas facilities, too, they have very strong baseload capacity in regions that certainly need it.
Do all things digital consume just 3% of the nation's electricity? Or somewhere in the range of 8% to 13%? We first published our estimate, the latter range, in Forbes ("Dig More Coal: The PCs are Coming," May 31, 1999).
Our critics responded that we were too high by a factor of three to ten. Lots of people got interested; even Doonesbury caught up with the debate on October 23, 2000, (https://www.doonesbury.com/strip/dailydose/index20001023.htm) by way of Presidential candidate George W. Bush, who had cited one of our estimates in a speech (https://www.bcse.org/climateact2.htm). Our estimate continues to circulate, and our critics continue to maintain the right number is no more than 3%—most recently in the Wall Street Journal. ("Bold Estimate of Web's Thirst For Electricity Seems All Wet," Dec. 5, 2002, www.wsj.com). But we stand by our numbers. To read more about this subject, and a rebuttal to the 3% claim, log on to Peter Huber and Mark Mills' Digital Power Group at www.digitalpowergroup.com.
This piece originally appeared in Forbes
This piece originally appeared in Forbes