And then you pay 10x market rate for the resulting electricity for the rest of the panels’ lifetime. Naturally they don’t say this up front, or anywhere on their website. But do the amortized cost analysis, with discounting. From SunRun’s page: >Overall, the cost of a solar system is usually between $8 and $10 per AC watt http://www.sunrunhome.com/learn_about_solar/cost_of_solar/ In line with every other figure I’ve seen (N.B. the panels alone are just 60% of this cost.) They also say the system will last for 18 years: http://www.sunrunhome.com/sunrun_home_solar/sunrun_total_solar/ A typical PV capacity factor, for a temperate climate at mid-latitude, with non-solar-tracking panels (e.g. roof mounted), is around 10%. As my reference example, I use this large commercial roof installation in Massachusetts, which is nice enough to provide actual generation statistics (you can get the raw numbers in a table, or a graph): http://www.sunviewer.net/portals/MoCA/ For the year of 2008, they generated 47,514 kWh (over 366 days), compared to a nameplate capacity of 52 kW; hence a capacity factor of 10.4%. Now let us apply the standard discounting model, as is explained in e.g. appendix 5 of the IEA cost study (they use both 5% and 10% discount rates). http://www.iea.org/publications/free_new_Desc.asp?PUBS_ID=1472 For these purposes, I will ignore maintenance costs, assume a constant rate of power generation over 18 years, and a single up-front solar panel purchase/installation, which is *not* discounted (because it is in year 0). The basic equation simplifies to: levelized cost of electricity = I / (? E(t) * (1+r)^-t) Where ‘I’ is the investment cost, ‘E’ is the electric generation (which for this purpose is a constant function of t, E(t) = E0), and r is the discount rate, and ‘t’ ranges over the useful lifetime of the solar panel, that is [0,T] where T=18 years. The ? might as well be replaced with a continuous integral, whose solution is a much simpler formula: E0 * (1 – (1+r)^(-T)) / (ln (1+r)) where ln is the natural logarithm (r>0 of course). (Easy to check that in the r->0 limit this is reduces the simple integral of E(t), that this, the total amount of electricity generated in the whole lifetime.) The LCOE equation becomes: LCOE = I* ln(1+r) / [E0 * (1 - (1+r)^-T)] So finally: at the lower discount rate of 5%, an investment cost of $9/W, an economic life of 18 years, and an electricity generation rate E0 of (0.10 * 1 W) = 0.88 kWh/year, the levelized cost estimate is **86 c/kWh**. (If you were to naively sum the electricity generation without discount, as many naive solar advocates do, you would get "only" 57 c/kWh.) For comparison, the average residential electricity cost in Colorado is 9.55 c/kWh (2009): http://www.eia.doe.gov/cneaf/electricity/epm/table5_6_b.html (Of course the R sector pays much more than average; industry for instance, only pays 5.94 c/kWh). In short, the expected real cost of rooftop solar power is an order of magnitude above the grid cost. (I emphasize this is end-user, retail cost – not the production cost for utility power plants. Those are far lower, e.g. for coal about [2c/kWh](http://1.bp.blogspot.com/_lfibbBnlKt8/SlDvz6DbO-I/AAAAAAAAAPw/hGT8_rouDGk/s400/US_Electricity_Production_Costs_blk95-08.png), or 1/60th the cost of solar generation). No financial trickery can hide this cost; either they pass it on to you, the victim, or to other victims, taxpayers, or they operate at a huge loss. It is an exercise for the reader to determine exactly what outrageous lies they use to claim that this is a money-saving purchase: http://www.sunrunhome.com/sunrun_home_solar/sunrun_power_plan/

This comment was originally posted on Reddit