I think your numbers are pretty optimistic. You're neglecting depreciation for one. Panels have a lifetime in the range of 20-30 years, inverters have a lifetime in the range of 10 years. 4% is also an optimistic interest rate for a loan that lasts the lifetime of the system. You may be able to get that for a ~5 year loan, but then you need to include the higher opportunity cost of the money that was used to pay that loan off which will be tied up for 20+ years.
Ok, I've updated the spreadsheet to properly account for discounted energy (discount rate: 4%) and degradation (at the rated 92%/25yr). It comes out to 6c/kWh, which is about 20% larger than my first number. Half of that increase was due to the degradation, the other half from the limited time horizon - and assuming the system will be worthless after 25yr (!!).
The cost of the system includes buying a new one in 30yr. I didn’t include deprecation of the panel, but modern panels such as Sunpower or Panasonic degrade at around 8% over 25 years, so maybe add a multiplier if 1/.92 to the cost to fudge that in.
As for the 4% interest rate, which in this case is really a time value of money discount rate, I’m comparing it to mortgage loan amounts (which is pretty much the average consumer's risk free rate of return). I’m assuming the buyer can afford a cash system.
I don't se how it includes the cost of a new system in 30 years. 2.3 * 30,000 = $69,000 and your cost listed is only ~$59k. 8% over 25 years is much better than I thought for the panels, but you still need to include about 10%/year depreciation on the inverters since you'll need to replace them every 10 years or so (at least that was about their lifetime last time I checked in Arizona, it may be different in cooler climates)
I'm including the 30% tax rebate and assuming solar panel installations will be 50% cheaper in 30 years which is pretty conservative given their price history.
Regardless, adding up the cost of the second one in 30 years doesn't make sense for calculating kWh prices, depreciation and cost of capital are the only factors that matter and you are only including one of those.
Sure, looking at my spreadsheet I also discounted the cost by the same interest rate after 30 years, (1/1.04)^30.
Agreed that including the second system doesn't really make much sense, but I wanted a quick-and-dirty estimate without having to individually discount each year's production (as you would if you had a finite time horizon).
You should always be paying cash whenever possible for the remaining system cost (after incentives). Failing that, you should be able to get a HELOC, home equity loan, or roll the system cost into a first mortgage at low (~4%) fixed rates (that might also be tax advantaged).
Higher interest rates would change the equation, but I don't know of people who are using high interest loans (like an unsecured loan) to finance residential solar.
