A Moore's Law for solar should predict the rate at which the $/kWh price of photoelectricity falls and tie this into market growth. This prediction would partly be based on learning curve potentials but mostly on market conditions. There are several things at work.
The Float Effect Rides High Down Incentive Lane
There's a distribution of incentives in the world market. These incentives have inspired solar developers to move into Germany, Spain, the Czech Republic, California, New Jersey etc. The higher the incentive the more developers move in. You'd expect this to push down system prices but prices tend to float at a level higher than input costs would have you expect.
The best (or worst) example of solar incentives increasing prices comes from Spain. Their particularly high Feed-in Tariff coupled with high insolation, floated system prices to up over 6 Euro/Watt in 2008 - A nosebleed level compared to today's average prices of about 3 euro/Watt in Germany. There were several factors in play that led to these high prices. One could say it was all because of the silicon shortage which pumped up module prices. This certainly had something to do with the high prices but remember that the silicon shortage (at least the severity of it) was attributable to incentive programs like Spains' in the first place. Supply and demand for silicon, inverters, labor etc. feed into the float effect but are not the primary driver. The primary driver of the float effect is the incentives themselves, the high FiT in this case. If a solar installer knows the FiT rate, they can infer an acceptable IRR for their customers and from this know how much they can get away with charging. If there's any sort of supply shortage, a high incentive and a limiting time factor the installer holds all the cards and can set system prices at will. Spain's FiT program ended badly. There's no market left. There's no way to see if the Float Effect is still playing out there. We have to look at Germany for this.
The Float Effect vs. The 2009 Recession
Germany provides a better snap shot of the Float Effect at work. We've seen system prices float down with the FiT over several years, crash in early 2009 due to limited liquidity (a discontinuity) and equilibrate by the end of 2009. We saw average system prices in Germany step down by about 9% with the year end degression. What was the year end degression you ask? 9 percent. If Germany remains the driving market, the Float Effect predicts that the July 1st degression will lead to a 13% step down in prices, followed by another 3% step down in September and another 13% step down in January. Will Germany remain the driving market? For 2010 definitely, for 2011 probably.
The Float Effect vs. The Competitive Market
The Float Effect becomes most useful when we start visualizing the longterm competition of photoelectricity with the grid. To be continued...