The post is brilliant in its simplicity. By taking NREL data, the average price of a solar module can be seen to be dropping at 6%/year. Extrapolating forward, while making the assumption that the installed cost will continue to be twice the module cost (which may not be true as improvements to module cost have seemed to be happening faster than installation improvements - but we can hope), as well as making a couple of other reasonable assumptions on interest rates and cost of electricity, he demonstrates through a pretty simple NPV analysis that by 2015, solar will have reached grid parity in Minnesota.
That's pretty striking, and he didn't provide what the solar resource was like in Minnesota (although, it might be pretty good - there is a latitude issue, but I bet that there aren't many clouds).
I like this analysis because it is so simple, and it's fairly easy to change any assumptions that you might disagree with and see what there effect is. This graph from Solarbuzz shows that module prices have started to work through the silicon shortage (it had to happen sometime!) and prices have resumed their downward trek. This might push out the grid parity time an additional five years - or, due to the billions of investment dollars that have gone into solar in the last few years (and given First Solar's announcement of <$1/W manufacturing cost) the price reduction curve could accelerate from 6%/yr to 10%/yr or beyond. At any rate, this is exciting as I believe that since electricity is a commodity, there will be a sharp tipping point once the economics work out, and this is all foreseeable in the next few years.
