Out of the spotlight, solar will thrive

By Jim Jubak

Government neglect could make 2008 and 2009 the best years ever for solar companies -- and for those who invest in them.

That's right. Long-term prospects for the solar industry are actually brighter because the energy bill that President Bush signed Dec. 19 didn't launch a crash program to expand solar-energy use or even extend solar tax credits that will expire in October.

The solar industry had a problem in 2007, and it wasn't a lack of demand. Global solar production, measured by the megawatts of power that solar cells and modules produce once they're hooked up to the grids, climbed 57% in 2004, according to investment bank Jefferies International. And it increased 30% in 2005, 35% in 2006 and a projected 13% in 2007.

See a problem there? Young growth industries facing a virtually untapped market shouldn't show slowing growth rates. If solar has the potential its supporters say it does -- and I agree it does -- the industry's growth rate should be accelerating, not dropping. (Global solar installations, which always lag equipment production and which are growing from a smaller base because of that lag, climbed a projected 60% in 2007.)

Not enough raw materials

So what happened? Beginning in 2005, demand for silicon from solar-wafer, solar-cell and solar-module makers overwhelmed supply from the companies that provide silicon to the solar and semiconductor industries.

(Solar wafers are processed silicon that's been cut into pieces, the first stage in producing solar cells. Wafers are doped with materials such as boron and phosphorus to turn them into cells that will generate flows of electrons when exposed to sunlight. When cells are placed in metal frames protected with rubber or plastic and embedded in a protective coating, they become solar modules.)

Companies that turn silicon ingots into solar wafers couldn't get enough crystalline silicon to meet orders from solar-cell makers. That forced makers of solar cells and modules to idle capacity. Utilization rates, the percentage of production capacity that a company is using, fell among solar-cell makers to 70% in 2005 from 86% in 2004. And the problem kept getting worse in 2006 and 2007. German wafer and cell maker ErSol Solar Energy (ERSLF, news, msgs), for example, cut its forecast for 2007 production to 55 megawatts from 70 megawatts, a 21% drop, because of the wafer shortfall.

That shortfall in supply had a predictable effect on silicon prices: They went up. Way up. Companies able to arrange long-term contracts with suppliers saw the silicon raw material required by wafer and solar-cell manufacturers jump in price to $45 a pound by mid-2007 from $20 in 2003. On the spot market, cell manufacturers who couldn't sign up a stable, long-term supply were paying $95 a pound.

The logjam won't last

Those higher prices have had the salutary effect of bringing more companies into the business of making silicon for solar wafers and encouraging existing suppliers to expand. After climbing just 14% in 2007, silicon production is projected by Jefferies International to climb 43% in 2008, 50% in 2009 and another 50% in 2010. The industry bottleneck that restricted production and produced higher prices for raw materials will be broken this year.

And that's critical because the big barrier to growth in the solar industry isn't a lack of subsidies from the U.S. government but the cost of electricity produced from solar cells. Right now, it costs about 30 cents to produce a kilowatt-hour of solar electricity versus the 15 to 18 cents retail customers pay for most kilowatt-hours in the United States. So, solar isn't yet competitive with other technologies for generating electricity.

But solar is gradually closing the gap. Solar-wafer makers are getting more wafers out of a kilogram of silicon by making wafers thinner. The industry is on track to get 35 wafers out of a kilogram of silicon by 2010, up from 29 wafers per kilogram now. That 21% increase is a huge cost savings because silicon accounts for about 30% of the cost of a solar module.

At the same time, solar-cell makers are getting more energy out of their cells. Solar-cell efficiency is projected to go up to 17% from the current 15% by 2010. Combine that with the manufacturing savings and the cost of solar electricity falls almost 30% by 2010, to about 21 cents a kilowatt-hour.

Price parity coming soon

I think you see the punch line coming. As electricity from solar gets cheaper and electricity from conventional sources gets more expensive, at some point the cost of solar-generated electricity reaches parity with the retail cost of electricity.

Hemlock Semiconductor, the world's largest producer of silicon, projects parity around 2012. (Hemlock is owned by Dow Corning, Mitsubishi Materials (MIMTF, news, msgs) and Shin-Etsu Chemical (SHECF, news, msgs). Jefferies International is a bit more conservative: It sees parity in the sunshine-rich U.S. Southwest in 2013.

Government support in the form of rebates to buyers of solar equipment (the U.S. approach) or in the form of guaranteed above-market prices for the purchase of electricity from solar-power generators (the European approach) is crucial to reaching that point. Without the economies of scale created by the demand growth generated by these subsidies, solar costs won't fall fast enough to hit parity on that schedule.

Continued: Why neglect is good for the industry

So why do I say U.S. government neglect, which puts existing federal subsidies in jeopardy and postpones plans to raise subsidies, is a good thing for the solar industry?

Two reasons:

• Existing subsidies in Germany, Spain, South Korea, Greece and individual U.S. states such as California will generate enough demand to keep solar efficiencies climbing and solar costs dropping this year.

• The delay in increasing U.S. subsidies avoided creating more demand than the industry could meet during the silicon supply crunch. A big surge in demand would have produced runaway silicon price increases and the kind of disruption that could set back an industry for years.

With the bottleneck likely behind it, the industry looks like it's ready to ramp up to meet a surge in demand that will result from an extension and increase in U.S. subsidies and new directives in China to increase the use of solar power. The Beijing government has decreed that utilities generate 5% of their electricity from renewable sources by 2010 and 10% by 2020 (and hydroelectricity doesn't count). That could result in 6 gigawatts of annual demand -- roughly 50% more than total global installed solar-power systems in 2007 -- if even 25% of those renewables were solar.

Beijing hasn't yet provided any money to back up its decree, making the real rate of adoption slower than projected, but I still think the global solar industry can expect to see demand from China gradually kick in over the next few years.

Nuclear power to benefit, too

And then there's the potential boost from nuclear power. I know, I know. Nuclear and solar are normally placed on opposite ends of the power-supply spectrum. But in a world increasingly likely to need a crash solution for reducing carbon emissions to combat global climate change, solar and nuclear are privileged front-runners. They both easily tie into the existing electrical grids, so nobody needs to build another distribution network.

The end product, electricity, can be used to reduce emissions in the carbon-intensive transportation sector. And that end product works equally well with proposed solutions such as electric or hydrogen-powered cars (where carbon-free electricity is essential for the production of hydrogen).

At the moment, I'd say nuclear and solar are dual-tracked. They're both seen as possible solutions by different constituencies. Nuclear has an edge in many government circles because it seems best-suited for generating a lot of carbon-free electricity quickly. That's why efforts to expand nuclear power in the United Kingdom and the U.S. are getting more government attention than solar.

But there's a good chance solar could move up in the list of favored government rankings because nuclear may not be able to deliver on schedule. The nuclear industry has its own bottleneck: Because no nuclear plants have been built anywhere in the world for 15 years, the industry has a severe shortage of engineers, component fabricators and equipment suppliers. It's a lot harder to solve that kind of bottleneck than it is to simply increase the supply of silicon. The first takes training a generation of engineers; the second takes three years or so to build a factory.

Timelines for crash solutions

Look at the proposed government schedule in the United Kingdom, a country on the edge of giving the green light to a new generation of nuclear plants. A 2009 study would identify sites for new plants. Construction wouldn't begin until 2012. And construction wouldn't finish, depending on how many other countries around the world had also launched crash nuclear programs, for three to five years after that.

As crash solutions go, that timetable leaves a lot to be desired. I'd guess that by about 2010, governments around the world will be eager to hedge their bets by launching crash solar programs.

Current projections of 25% to 35% annual growth in solar installations for the next decade would turn out to be pretty conservative in that scenario.

My three solar picks -- wait to buy these until the market settles down -- are:

• Q-Cells (QCLSF, news, msgs), a German company that is the world's largest producer of solar cells from conventional silicon.

• Suntech Power Holdings (STP, news, msgs), the most promising Chinese producer of solar cells.

• First Solar (FSLR, news, msgs), a leading maker of solar cells using thin-film technologies that promise to cut the cost of solar cells.

Continued: Updates and developments

Updates to Jubak's Picks

Sell Companhia Vale do Rio Doce (RIO, news, msgs): The problem isn't that 2008 will be so bad for Companhia Vale do Rio Doce but that 2007 was too good. Revenue climbed a huge 65% last year, and the increase looks a lot more modest in 2008, with Standard & Poor's putting revenue growth at 14%.

The stock's price-earnings ratio doesn't seem high at 14.4 times trailing earnings, but that's way above the eight-to-nine multiple it showed from 2003 through 2006. The huge load of debt the company has taken on to buy nickel supplier Inco and increase capital spending rapidly also makes this stock riskier in the current market. The company's long-term debt climbed to $17.5 billion in the third quarter of 2007 from $4.6 billion in the third quarter of 2006.

So I'm going to take my profits in the company, now known as Vale. I have a 56% gain since I added the stock to Jubak's Picks on April 17, 2007. (Full disclosure: I will sell my personal position in the stock three days after this column is posted.)

Buy Exelon (EXC, news, msgs): You don't have to wait for the world's new generation of nuclear plants to profit from nuclear energy. In fact, given all the real uncertainties that underlie the so-called nuclear revival, I'd bet that going with a company that owns operating nuclear plants that generate electricity now is not only safer but likely to be more profitable in the long haul.

Exelon's 17 nuclear units generate 18% of all U.S. nuclear power. It costs a coal-fired power plant between $30 and $50 to generate a megawatt-hour of electricity. At a natural gas-fired plant, the cost is $50 to $70. Existing nuclear plants generate electricity for just $15 a megawatt-hour. Nationally, the retail price of electricity averages $100 to $150 a megawatt-hour.

Nuclear plants' big edge on profit margins is likely to increase as efforts to combat global climate change impose extra costs on power plants that use carbon-based fuels.

Exelon had its best year for power generation from its nuclear units in 2007 thanks to a record 94.5% average capacity. The company seems to think these good times will go on for a while. It has raised the dividend it pays by 14% to an annual $2 a share and announced a $500 million increase to a $1.25 billion plan to buy back shares. (The shares paid a yield of 2.4% as of Monday.)

As of today, I'm adding these shares to Jubak's Picks with a target price of $92 a share by July.

Developments on a past column

"How to profit from rising food prices": World grain supplies will be lower -- and prices higher -- than projected earlier. That was the message in the U.S. Department of Agriculture's World Agriculture Supply and Demand Estimates report released Friday.

The report lowered projected corn production by 94 million bushels. That's not good when year-end corn inventory had declined by 360 million bushels. The projected price of corn in 2008 went up to $3.70 to $4.30 a bushel from $3.35 to $3.95 a bushel.

The department also raised the projected price of a bushel of wheat by 25 cents. The range for soybeans went up to $9.90 to $10.90 a bushel from the earlier $9.25 to $10.25, as the projected supply and year-end stocks were both revised lower. This news is good for agriculture stocks such as Deere (DE, news, msgs) and for fertilizer makers, but bad for producers of ethanol and, of course, for those of us who buy food. After the report came out, March corn closed up 20 cents a bushel at $4.95.

Meet Jubak at The Money Show

MSN Money's Jim Jubak will be among more than 120 investment and finance experts sharing buy and sell advice at The World Money Show in Orlando, Fla., Feb. 6-9. Invest four days dedicated to planning and refining your portfolio by attending the event's more than 320 workshops and panel presentations.

Admission is free for MSN Money readers. To sign up, call 1-800-970-4355 and mention priority code No. 009554, or register online.Editor's note: Jim Jubak, the Web's most-read investing writer, normally posts a new Jubak's Journal every Tuesday and Friday. Please note that recommendations in Jubak's Picks are for a 12- to 18-month time horizon. For suggestions to help navigate the treacherous interest-rate environment, see Jubak's portfolio of Dividend Stocks for Income Investors. For picks with a truly long-term perspective, see Jubak's 50 Best Stocks in the World or Future Fantastic 50 Portfolio. E-mail Jubak at jjmail@microsoft.com.

At the time of publication, Jim Jubak owned or controlled shares in the following equities mentioned in this column: Companhia Vale do Rio Doce. Hedid not own short positions in any stock mentioned in this column.