After four years of having our photovoltaic array on the roof, and people asking me how long it would take for the system to pay for itself, I finally decided to do some calculations. I had put off the calculations because it seemed to involve too many assumptions to be accurate. So instead of going for 100% accuracy, I'm going for a reasonable estimate.
The PV system is estimated to pay for itself in 15 years if we take into account only the savings from the electricity generated by the panels, and 9 years if we also include the savings from changing to the new electric meter.
I will be updating these estimates each year until the system pays for itself, and will keep a record of the previous estimates. Click in the table to get spreadsheet details of each year's estimates.
| PV Payback Period | PV only | w/Meter |
| Year 4 (2007) Estimate | 18 yrs | 10 yrs |
| Year 5 (2008) Estimate | 18 yrs | 10 yrs |
| Year 6 (2009) Estimate | 16 yrs | 9 yrs |
| Year 7 (2010) Estimate | 15 yrs | 9 yrs |
Note that when we signed up with Akeena Solar, I expected the payback period to be 13.5 years. After the final design was set and the city forced us to put some of the panels on the north-facing roof, I expected the payback period to be 14 years. Those were my extremely crude estimates assuming 0% interest rates and no increases in PG&E electricity rates. Akeena did not give a payback estimate in terms of years. Instead they estimated a "tax free rate of return" of 7%, whatever that means.
As electricity rates go up in the higher tiers, the estimated payback period gradually gets shorter.
Calculations are done as if financing for the PV system were put on a second mortgage, rather than wrapped into our house mortgage.
The net cost of the PV system is assumed to be the cost of the PV system ($76,994), plus the cost of the new Time-of-Use meter ($277), less the $4/watt rebate ($29,472) and the 15% California Tax Credit ($7,528). Net Cost $40,271.
The amount of the loan is assumed to be the net cost of the PV plus the actual closing costs of the 5.75% interest rate mortgage we got.
The loan interest rate in the first two years is assumed to be 5.75%
The loan interest rate starting in year 3 is assumed to be 5%. While we did not refinance from 5.75% to 5% on the exact second anniversary of our PV system, it is close enough.
The closing cost for the refinance to 5% is not included in this analysis because we would have refinanced our mortgage at that time anyway, even if we hadn't had the PV system. The same could not be said of the 5.75% mortgage.
To account for the effects of mortgage interest deductions on income tax the incremental income tax rate of 9.3% was used for California taxes, and an incremental tax rate of 25% was assumed for Federal taxes. The effect of the lower CA tax from the 15% rebate was included in calculating how much the Federal tax deduction would change.
Update: Because of the Great Recession, California increased its income tax rates by 0.25% for 2009 and 2010. The spreadsheets from Year 6 (2009) on, have been adjusted to take into account the 9.55% tax rate for those years. The effect is pretty small, especially compared with the effect of the PG&E rate changes.
In the PV-only scenario, there is negative cash flow in the first two years. In the first year this is caused in part by having to make loan payments during the planning and installation phase before the PV system started generating electricity. In the second year, it is caused by the Federal income tax paid on money we got from the California Solar Tax Credit. It is assumed that these extra costs are paid for using a Home Equity Line of Credit (HELOC) with the same 5.75% interest rate as the mortgage, and that in the third year it was wrapped into the refinance to 5% interest. In reality our HELOC had a floating interest rate that was lower than 5.75%. (In fact it was no more than 4.25%.)
The bill for the electricity used for the entire year is due at the end of the billing year. It is assumed that after we pay the bill, we take the amount we saved that year and use it to pay down the mortgage principal faster. This assumes no penalties for early payment of the mortgage.Future year savings are assumed to be the same as the current year savings.
When the PV system was installed, a new electric meter was put in that was rated to be able to accurately measure power flowing both in and out of our house to the grid. When the new meter was put in, the new meter read a usage 46% lower than the old meter. We suspect that the old meter read Volt-Amperes and the new meter reads Watts, and that the difference between the two is caused by the Power Factor from the inductive loads of the computers and fluorescent lights.
Whatever the true cause, I estimate that if we still had the old meter, we would probably still be showing an apparent measured usage of 80 KWH/day, so the new meter alone is saving us money even if the PV array wasn't on.
With our measured usage lower with the new meter, our savings from our solar electricity generated is lower then we had originally projected because we were expecting the solar to offset power bought at the expensive Tier 4 and Tier 5 rates. The lower usage means in some months the solar is offsetting cheaper Tier 3 rates and sometimes even rates with no surcharge. Because of this, I think it is proper to include the savings from the meter change in calculating the payback for the PV array.
PG&E is planning to install smart meters at all residences in our area in 2012. I expect the new smart meter would correct any reading errors that our old meter had. Current estimates is that the PV including the meter will pay for itself by then. But once the smart meters are installed in our area, I should stop counting any additional savings that were based on comparison with the old meter.