Does CPUC’s PD (NEM 3.0) Stall Rooftop Solar Adoption?
California’s proposed decision (PD) of replacing the current NEM 2.0 policy to its successor NEM 3.0, has attracted enormous public attention. The proposed tariff changes are primarily driven by the issue of cost-shifts from solar adopters to non-adopters, which is compounded by the unparalleled huge price mark-up between the retail and wholesale prices in California (see here). Moreover, the “almost” retail-rate compensation of injected generation under NEM 2.0 with the heavily volumetric charge snowballed the subsidies between adopters and non-adopters.
The CPUC’s proposed NEM 3.0 policy would compensate excess solar at the avoided cost rate in addition to levying a solar capacity based charge CBC (solar tax) of around $8/kW.
What should the compensation rate for excess generation be?
To answer this question, we must understand how rooftop solar achieves bill savings. Generally, the onsite generation is valued at the retail rate if it is self-consumed and at the sell-rate if it is exported back to the grid. As a special case, under NEM 1.0 and NEM 2.0, these two rates are equivalent. Therefore, for PV adopters under those policies, the value of self-consumption is equivalent to the value of exporting back to the grid. While this is creating considerable bill savings for adopters (see here), purchasing solar customers’ excess generation at the retail rate is a clear overcompensation, especially when the rate design is heavily volumetric.
Remark 1: A retail rate based sell-rate is an overcompensation, the sell-rate should be at the value of DER rate.
The arguably high bill savings of PV adopters in California result in a reduced utilities revenue from energy sales forcing them to increase their prices again, which makes it even worse, as PV adopters will achieve bill savings at higher rates and non-adopters (usually less affluent customers) will be unavoidably paying the higher rates. Pricing at the value of solar VoS (or value of DER) reduces utility revenue shortfalls and therefore the need to increase the retail rate, which also reduces cost-shifts. More importantly, valuing the net excess generation at the VoS encourages more self-consumption and discourages net exportation, which necessitates the installation of energy storage devices (batteries).
The Switcheroo In The Retail Electricity Market
Questioning Net Metering Regressiveness
Remark 2: VoS compensation reduces cost-shifts and encourages the adoption of PV+battery packages. VoS is usually higher than the avoided cost rate.
What should the capacity-based charge (solar tax) be?
It should be ZERO. California can adopt a less volumetric-charge-dependent tariff by considering uniform two-part tariffs (e.g. through delivery charges), rather than discriminatively as in CBC. CPUC’s PD seems to be solving a problem (i.e. volumetric cost recovery) by imposing another one (i.e. partial cost recovery via CBC). The consideration of a uniform fixed fee will reduce the blatantly high retail rate in an equitable and fair manner, resulting in a less severe subsidy issue.
Remark 3: The Californian proposed solar tax is a form of discriminatory two-part tariffs, where PV adopters are discriminated just for adopting the green technology.
Although solar tax is extremely effective in recovering utility costs, and lowering rooftop solar bill savings, which prolongs payback times and reduces subsidies, it does so in a very regressive way. There aren’t clear justifications for levying such a high charge on adopters. If the utility designed a rate that creates cost-shifts whenever a customer reduces his/her consumption, why should PV owners pay for it?
Remark 4: Solar tax is a pill for relieving the utility’s struggle in cost recovery, sugar-coated by the cost-shifts argument.
The CBC would add a minimum of $40/month to the bill of a PV adopter increasing the payabck time to more than 20 years. This will definitely impact the diffusion of solar in California, because the discriminatory two-part tariff would lower the retail rate, which even further prolongs the payback time.
Remark 5: Solar tax has a dual effect on the bill savings achieved by rooftop solar: a) it prolongs the payback time by reducing the monthly bill savings via adding the $8/kW/month, and b) it prolongs the payback time by reducing the retail rate, which devalues bill reductions through self-consumption.
Moreover, unlike reducing the sell-rate which encourages batteries’ adoption, levying solar tax does not push customers toward considering PV+battery pairs. Part of the regressiveness of solar tax has to do with the fact that it cannot be avoided nor minimizedq by installing a battery. In fact, they can only be avoided if you don’t install a DER at all.
Remark 6: The regressiveness of solar tax is induced by its unavoidability. Installing a battery does not enable the solar adopter to avoid solar tax.
What should the successor policy be?
Valuing exported generation at VoS is a move in the right direction, but CBC is not. Although VoS would partially relieve the utility cost recovery issue, a more differentiated time of use (ToU) rate (higher peak to off peak ratio) with a later peak period (16:00 and beyond) applied to all customers would bring the retail rate closer to the marginal cost of electricity, without negatively impacting non-adopters.
Remark 7: A proper and just NEM 3.0 policy should include:
- VoS compensation for net excess generation.
- Highly differentiated ToU with a later peak period, applied to all customers.
- Uniform and reasonable fixed charges (delivery charges).
- Shorter billing periods (faster netting)
A later peak period ToU rate implies that most excess generation is compensated at the off-peak price, which in turn encourages PV+battery adoption to arbitrage the peak/off-peak ratio. The implementation of uniform fixed charges might also be necessary if such rate modifications does not reduce the price markup. Uniform fixed charges are almost applied everywhere in the US. They are easy to implement and understand, effective, equitable, and able to reduce the volumetric cost recovery, which fails miserablly under net-metering. Lastly, the policymaker should consider shortening the billing period, which reduces credit banking opportunities and make VoS compensation more effective.