Craig Hunter, Vice President and General Manager of Energy Technologies at Intermolecular -- which specializes in semiconductor and solar R&D design and testing -- concurs in Greentech Media:
[T]he litmus test for any solar energy technology is its ability in the next 10-20 years to be deployed in hundreds of gigawatts per year, delivering electricity at $.05-.07/kWh, even in areas that aren't very sunny. Given the load factors of PV installations, not to mention the possible need for storage, we need to consider a target installed system price of no more than $1/Wp.Personally, I'm agnostic about whether a technological breakthrough might make non-PV solar power significant in the long term. But I'm sure that distorting subsidies are wasteful and unnecessary--the free market responds automatically to shortages and can be counted on to solve any energy crisis. Without subsidizing what might be dead-end technologies such as PV.
Panel prices have indeed come down significantly, but the PV "experience curve" (15% cost reduction for each doubling of production) is too slow, requiring us to get to 40-80GW per year production just to reach sub-$1/Wp panel pricing. And unless we see disruptive improvements in conversion efficiency, the "balance-of-system" costs (i.e., all the system costs other than the solar panel itself) will make it impossible to achieve a $1/Wp system price even if the panels are nearly free.