It would be a plot worthy of James Bond villains Auric Goldfinger, Ernst Blofeld or Francisco Scaramanga: Destroy the world's solar industry by providing cheap capital and demand-creating subsidies so solar companies overexpand, then pull the plug on capital and subsidies so the industry goes bankrupt.

Except that this is the real world, and nobody needs Scaramanga to steal the Solex Agitator, the MacGuffin essential to solar energy production in the fictional world of "The Man with the Golden Gun." In the real world, well-meaning governments, financial markets and the solar industry can do the damage all by themselves.

The price of polysilicon, the raw material for conventional solar cells, has dropped by 56% this year, to levels not seen since 2003, and is down by 93% since 2008.

And yet, if all the factories now under construction or on drawing boards get built, global supply of polysilicon will climb to 500,000 tons by 2014 from 266,000 tons in 2011.

The average margin at the companies that turn polysilicon into solar cells (or that use other technologies for producing photovoltaics) fell to 0.1% in the third quarter of 2011 from 13.7% in the third quarter of 2010.

Image: Jim Jubak

Jim Jubak

And yet, the 10 largest makers of silicon panels doubled their manufacturing capacity last year.

It's no wonder that Jifan Gao, the CEO of Trina Solar (TSL, news), China's third-largest producer of solar panels, said in a recent interview with Bloomberg that two-thirds of the players in the industry will go bust or be acquired between now and 2015. Similarly, recent report from Australia's Macquarie predicted 66% of polysilicon producers would fall victim to the shakeout that has just begun.

For example, in China the sector could be reduced to as few as four players over the next three years from 35 known producers today. Further down the supply chain, the companies that manufacture solar panels face a similarly disastrous near-term future. The spot price of solar panels has tumbled 40% this year. Q-Cells (QCLSF, news), once the world's biggest maker of solar cells, has said it's open to a takeover.

And yet, this looks like the solar winter that comes before the solar spring. That's because a number of factors suggest the elusive goal of price parity between solar-generated electricity and electricity from coal, oil, natural gas and uranium is in sight:

  • The drop in the price of polysilicon, which makes up about 25% of the cost of a finished solar panel.
  • The intense price competition among solar companies.
  • The improving efficiencies in manufacturing.
  • Advances in the efficiency with which solar panels convert sunlight to electricity.

Exactly when parity will arrive is a matter of intense argument and speculation, but in some places -- those with lots of sunshine and higher costs for electricity from other sources -- it could come as early as 2015.

Installed global solar capacity was 36 gigawatts at the end of 2010, according to the U.S. Energy Information Agency. By the end of 2020 global capacity will be 20 to 40 times that level -- or 720 to 1,440 gigawatts, consultant McKinsey & Co. projects. Investment banker Piper Jaffray projects growth to a global installed capacity of 800 gigawatts by 2020.

Whoever gets through this winter will make a lot of money come solar spring.

3 ways to buy the trend

So how do you play this as an investor?

I think you've got three choices:

  • You can wait until 2013 or 2014, or perhaps even 2015, to see which companies survive and buy those shares then. That has the advantage of knowing who got shook out during the shakeout and who remains in a position to profit from the turn in the market.
  • You can buy a portfolio now of solar companies that look like they'll be survivors, in the belief that your winners will win big enough to make up for your losers. That's the kind of strategy that investors in high-risk areas such as venture capital pursue. It has the advantage of staking your claims now when prices are depressed, but it has the disadvantage that you could pick a portfolio full of turkeys and miss the upside even though you're diversified.
  • You could combine those two strategies by waiting for the shakeout to progress for a year or two and then buying a portfolio of likely survivors.

I'd vote for No. 3.

But no matter which of the three you decide to pursue, you'll need two things: a method for deciding which companies are likely survivors and a method for deciding which companies will be likely to profit in the transition from solar winter to solar spring. Those are the ones you want to own.

Continued on the next page. Stocks mentioned include LDK Solar (LDK, news), Canadian Solar (CSIQ, news) and Conergy (CEYHF, news).