The big changes wrought by the Net Energy Metering 3.0 approved last year made me wonder if I would install the same solar+storage system I invested in 2019 if I were subjected to the new controversial tariff?

The California Public Utilities Commission’s final NEM 3.0 decision issued Dec. 15, 2022, represented a significant retreat on the Commission’s more drastic proposal released the previous year. Two of the three prongs in the original proposal were removed, which was a major win for public opinion.

Even then, the final decision is far from admirable. Export compensation is being dropped off a cliff, not following a glide path as claimed. The decision uses below market costs for system installations, underestimating payback periods and overestimating likely adoption rates. 

It says the objective is to promote battery installations. Most solar customers are reluctant to install batteries because they are very expensive. Forcing them to buy a battery is akin to a car salesperson forcing a buyer who came into the showroom to buy a basic car to instead buy a luxury car that he or she cannot afford. 

If the CPUC’s intent was to promote battery installations, the easiest way to do that would have been to double or triple the rebates for installing them. 

The CPUC’s first cut at NEM 3.0 issued Dec. 13, 2021, however, would have imposed a grid access charge of $8/kW-month on new solar customers, radically dropping export compensation rates for new solar customers. It also would have applied the new rules retroactively on existing customers 15 years after their installation. 

This three-pronged attack on rooftop solar customers unleashed a firestorm of protests across the state for overreaching. No other state has reformed NEM retroactively. Just about no one has a grid access charge. Arizona eliminated it. Nova Scotia Power applied for it and was rejected. These two provisions were removed in the final decision.

NEM 2.0 and me

I contracted to have my solar+storage system installed in June 2019. It was installed in December 2019 and I was given permission to operate the next month.

Under NEM 2.0, I import power from the grid at the same time-of-use rate at which my exports are compensated. Since I also have an electric vehicle, I have chosen the three-period rate, which is Pacific Gas & Electric’s EV2-A rate.

Currently, the peak period price in the summer is 55.4 cents/kWh while the off-peak price is 24.1 cents/kWh. The mid-peak price is 44.4 cents/kWh. Back in 2019, the peak price was 47 cents/kWh while the off-peak price was 17 cents/kWh. I also pay a non-bypassable charge of roughly $14 a month.

I charge the battery system (LG Chem, rated at 9.8 kWh) in the morning, when the off-peak price is applicable. Once fully charged, it goes into standby mode. The house is now charged by the 25 Silfab solar panels (8 kW-DC), with any difference between consumption and production being met by importing power from the grid until 4 pm, when the peak period begins.

That’s when the battery begins to discharge, if solar power is not sufficient to power the house, which usually is the case. The battery keeps on discharging until it drops down to 30%, and then it goes into standby mode to help power five essential circuits during an outage. I have had eight outages since June 202 during which the solar+storage system kept the five essential circuits going. Interestingly, all of the PG&E outages occurred during normal weather. It’s also worth noting that the power lines in my neighborhood are underground.

The payback period on the net investment of $25,000 was estimated at nine years. If I had not installed the battery, it would have been seven years. I installed the battery partly for resilience and partly to see how it worked as an instrument of arbitrage against the time-of-use rate.

The system works like a charm. In 2022, it produced 10,800 kWh while I consumed 11,400 kWh (lots of EV charging at home). The prior year, it produced 9,670 kWh and consumed 10,200 kWh. In 2020, the first year, it produced 9,270 kWh and consumed 9,680 kWh. My electric bill used to average $203 a month and would probably have risen to $285 a month by now. Over three years, my bill has averaged $58 a month. 

After talking to solar and storage experts, I estimate that about 40% of the bill reduction comes from export compensation. If I had been under NEM 3.0 in 2019 (think of this as a “thought experiment”), export compensation would have dropped by 75-80%. The investment would have lost its charm.

The CPUC’s NEM 3.0 analysis does not seem to have used actual market data on system costs to estimate payback periods. I am not sure whether they reached out to customers who were thinking of installing rooftop solar panels to understand the decision making process.

Ignored all other cost shifts in rate design

The decision was premised on the need to offset the cost shift between solar and non-solar customers. Either willfully or otherwise, the regulators ignored all the other cost shifts inherent in rate design, between large and small customers, between peaky and less peaky customers, between customers living in apartments and in single family homes, between renters and homeowners, and between rural and urban customers. 

It ignored the cost shifts created by spending $1.5 billion on energy efficiency, which causes energy consumption to fall for those customers who participate in the energy efficiency programs. This causes a shortfall in revenue which is recovered by raising rates for all electric customers, causing a cost shift.

The CPUC also ignored the 35% electric discount being provided to low income customers through the CARE program. Finally, it ignored the fact that the state has a progressive income tax structure, which benefits low income customers.

So what is my answer to the Hamletian question I posed?  No, I would not have installed the system.

Would anyone else once NEM 3.0 goes into effect around April? We will find out in the next several months.

For details of my analysis, see the article which appears here.

Edited by Elizabeth McCarthy