Independent guide. Based on CPUC Decision 22-12-056 and the publicly filed Avoided Cost Calculator inputs.Verified May 2026

NEM 3.0 California 2026: net billing tariff math for new solar

California shifted from retail-rate net metering (NEM 2.0) to the avoided-cost Net Billing Tariff (NEM 3.0) on April 15, 2023. The change reduced export credits by roughly 75 percent, but added time-varying export rates that reward evening grid-stress generation. The result: solar-only payback stretched to 11-plus years, but solar+battery payback compressed to 7 to 10 years for most households.

Avg export rate

~7c

per kWh exported, weighted annual avg

Summer evening peak

$2-3

per kWh exported, Aug 5-9pm

Sunny afternoon trough

~2c

per kWh exported, May 12-3pm

Why CPUC moved to the Net Billing Tariff

The case for moving off retail-rate net metering rested on three observations the CPUC laid out in the 22-12-056 record. First, NEM 2.0 export credits exceeded the actual cost the utility avoided by accepting that kWh, sometimes by a factor of five during sunny midday hours when the grid is already oversupplied. The difference is a subsidy paid by non-solar customers (largely renters and lower-income households) to solar adopters (typically higher-income homeowners). The cost-shift was estimated at $4 billion per year by 2022 and growing.

Second, the duck curve is now an actual operating problem. CAISO sees negative wholesale prices most spring afternoons because of oversupply from utility-scale solar plus distributed rooftop solar; the system then has to ramp gas generation from near-zero to 15+ GW within three hours as solar drops at sunset. Crediting a rooftop solar export at retail rate at noon (when CAISO is selling power at -2 cents) is economically perverse. The new ACC-based export rate gives an accurate price signal: low value at noon, very high value at 7pm.

Third, the policy goal had shifted from getting any solar installed (the original 1996 NEM rationale) to getting the right resource mix (solar combined with storage that can shift the energy to evening hours) installed. The pricing structure under NEM 3.0 is explicitly designed to incentivise battery adoption, which addresses both the cost-shift problem and the duck-curve problem at the same time.

How the avoided-cost export rate works

The export rate is set by the Avoided Cost Calculator (ACC), a CPUC tool that calculates the marginal cost of energy the utility would have purchased absent the rooftop export. The ACC produces a value for every hour of every day, by climate zone, with components for: energy (the wholesale market price), generation capacity (the value of avoiding building new gas peakers), transmission capacity, distribution capacity, ancillary services, losses, GHG emissions, and methane leakage in upstream gas. The published 2024 ACC values show an average export credit around 5 to 8 cents per kWh, with massive variance: under 2 cents during spring middays, over $2 per kWh during August 7pm grid-stress windows.

For a solar-only system, exports are concentrated in the middle of the day, exactly when the ACC value is lowest. That is the core reason solar-only economics broke under NEM 3.0: the system generates the most when the market values exports the least. For a solar+battery system, the battery charges from the panels at noon (avoiding low-value exports) and discharges to the grid at 7pm (capturing high-value exports), which is precisely the economic case the tariff is designed to incentivise.

Worked example: 6 kW + 10 kWh in PG&E territory

Take a typical PG&E customer in Climate Zone 12 (Sacramento area) with 950 kWh per month of usage, installing a 6 kW DC solar array (about 9,300 kWh per year production) plus a 10 kWh battery. The system cost runs about $30,000 turnkey after a competitive bid process (so $5,000 per kW for solar, $1,000 per kWh for battery installed including inverter and electrical work). After the 30 percent federal Investment Tax Credit, net cost is roughly $21,000.

Operationally, the array generates 9,300 kWh per year. The household self-consumes about 4,000 kWh directly during sunlit hours. The battery captures another 3,000 kWh of midday generation and discharges it to the home during the 4-9pm peak window (avoiding PG&E E-TOU-C peak imports at about 50 cents per kWh). The remaining 2,300 kWh is exported at the ACC-weighted average export rate, which (with active battery dispatch to capture the 7pm window) averages closer to 12 cents per kWh than the unweighted 7-cent annual average. Total annual financial benefit: about $2,400 (the avoided peak imports plus the export credit minus the small remaining grid imports during cloudy stretches).

Simple payback: $21,000 / $2,400 = 8.75 years. Internal rate of return over the 25-year warranty: roughly 10 to 11 percent, depending on assumptions about future electricity-rate escalation and battery degradation. The same system under NEM 2.0 (without the battery) would have paid back in about 5.5 years; the NEM 3.0 plus battery configuration is slower to pay back but produces a similar lifetime IRR because the battery doubles as backup power and improves the household's resilience to Public Safety Power Shutoffs.

Solar-only under NEM 3.0: why it usually does not pencil

A solar-only 6 kW system in the same Sacramento household costs about $18,000 installed, $12,600 after the ITC. It produces the same 9,300 kWh per year, but without the battery, only 4,000 kWh is self-consumed (the daytime baseload, well pump cycling, fridge, EV charging if scheduled during sunlit hours). The other 5,300 kWh exports during midday, when the ACC value averages around 4 to 6 cents per kWh. The household still imports about 4,000 kWh during evening and early morning at full retail rate.

Annual financial benefit: avoided imports of 4,000 kWh at 41 cents (off-peak E-TOU-C) saves about $1,640, plus the export revenue of 5,300 kWh at 5 cents averages $265, total about $1,900 per year. Payback: $12,600 / $1,900 = 6.6 years. That looks fine in isolation, but the math gets worse the more the household tries to add load. Adding an EV pushes evening imports up, and without a battery there is no way to use surplus daytime solar to charge it economically; the EV charging stays at off-peak rates of 41 cents or worse. Households planning to add an EV, a heat pump or any other major electric load within five years should size for a battery from the start.

SGIP rebates and the Self-Generation Incentive Program

California's Self-Generation Incentive Program (SGIP) offers per-kWh rebates for battery storage that can offset 10 to 20 percent of the battery's installed cost. The rebate level depends on which "step" of program funding is currently active (the program is funded in tranches and the rebate declines as each step's funding is committed). The Equity Resiliency Budget within SGIP offers a much larger rebate (up to 100 percent of battery cost in some cases) for low-income households in High Fire Threat Districts or medically vulnerable customers reliant on grid power.

Application is via the SGIP portal through your utility (PG&E, SCE or SDG&E). The current step's rebate level should be checked at the time of bid because it changes; sizable rebates remain available as of early 2026 but the program is not unlimited. SGIP is in addition to the federal ITC, so a battery's stack of subsidies is meaningful when both are active.

If you are a NEM 2.0 customer: protect your grandfathering

Customers grandfathered onto NEM 2.0 keep that tariff for 20 years from their permission-to-operate date. Several actions can trigger loss of grandfathering: increasing system size by more than 10 percent (or by more than 1 kW, whichever is larger), changing the inverter to a different model in a way that requires a new interconnection application, or moving the system to a new property. Battery additions do not by themselves trigger loss of grandfathering, but if the battery requires upsizing the inverter and the inverter upgrade is processed as a new application, the new application falls under NEM 3.0.

Practical advice: if you are adding a battery to an existing NEM 2.0 system, use a DC-coupled battery that ties in behind the existing inverter, or use an AC-coupled battery wired into a critical-loads panel that does not require interconnection changes. Your installer should confirm in writing that the addition will not trigger a new interconnection application before you sign the contract. Some installers have been less than careful here, and unwary customers have lost NEM 2.0 status worth $5,000 to $15,000 over the remaining grandfathering period.

What NEM 4.0 might look like

The CPUC has not formally opened a NEM 4.0 proceeding, but the regulatory direction is clear. The next iteration will likely tighten the export rate further (the current ACC includes a generous methane-leakage adder that some stakeholders argue should drop now that gas peakers are being retired), require explicit demand-response participation as a condition of receiving the higher evening export credits, and add some form of fixed monthly charge to cover distribution costs that net-metered customers under-pay for. None of this is imminent, but new installations planning for 25-year payback should not assume the current NEM 3.0 export rates will hold throughout.

For households comparing offers from solar installers in 2026, the practical takeaway is: model the payback at the current NEM 3.0 export rates, then run a sensitivity at 50 percent of those rates to see what happens if NEM 4.0 cuts further. If payback under the 50-percent sensitivity is still inside 15 years, the investment is robust. If the 50-percent sensitivity pushes payback past 20 years, the system is over-sized for the new tariff environment and should be re-bid with a smaller array plus a larger battery share.

Sources and further reading

CPUC Decision 22-12-056 (the foundational NBT order)
CPUC NEM portal (current rules, ACC export values)
SGIP battery rebate portal
IRS Form 5695 (federal Residential Clean Energy Credit, the 30 percent ITC)
PG&E rate plans (essential context for self-consumption math)
California state electricity cost page
EV TOU rate plans (the EV that pairs with the solar+battery)
How we source these numbers
For solar-system cost benchmarks: see the sister site solarpanelinstallcost.com

FAQ

What is NEM 3.0 in plain English?

NEM 3.0, formally the Net Billing Tariff (NBT), is California's third-generation rule for compensating residential solar owners for the electricity they export to the grid. Under NEM 1.0 and 2.0, exported kWh were credited at the full retail rate (about 35 to 40 cents per kWh on PG&E, SCE and SDG&E). Under NEM 3.0, exported kWh are credited at a much lower 'avoided cost' rate (typically 5 to 8 cents per kWh on average), with higher credits during summer evening grid-stress hours and lower credits during sunny afternoons when the grid is oversupplied. The effect is to flip the economics so that a battery (which lets you self-consume your own solar during the high-credit evening hours) is now essential for a competitive payback.

When did NEM 3.0 take effect?

CPUC Decision 22-12-056 was issued in December 2022. New residential solar interconnection applications submitted to PG&E, SCE and SDG&E on or after April 15, 2023 fell under NEM 3.0. Applications submitted before that date were grandfathered onto NEM 2.0 for 20 years from the interconnection date. The grandfathering deadline triggered a rush of applications in early 2023 (industry sources reported a 4-5x year-over-year spike in March and April 2023).

Are NEM 2.0 customers grandfathered forever?

Grandfathered for 20 years from their permission-to-operate (PTO) date. After 20 years, those customers transition to whatever tariff is in effect at that time (could be NEM 4.0 or beyond). A customer who installed under NEM 1.0 in 2007 would have transitioned at the 20-year mark in 2027 unless the CPUC extended the grandfathering window; the 2022 decision did not extend NEM 1.0 grandfathering, so those early adopters are likely to roll onto NEM 3.0 over the next 1 to 5 years.

What is the typical payback for a NEM 3.0 system in 2026?

For a 6 kW solar + 10 kWh battery system in PG&E territory: turnkey cost roughly $30,000 before the 30 percent federal Investment Tax Credit, $21,000 after the ITC. Annual bill savings about $2,400 (avoiding most peak-window grid imports plus capturing the evening export-credit window). Simple payback about 8.7 years. NEM 2.0 systems without batteries paid back in roughly 5 to 6 years; the additional cost of the battery and the lower export credits stretch payback by 3 to 4 years.

What is the avoided-cost export rate based on?

The export rate uses the Avoided Cost Calculator (ACC), a CPUC tool that models the marginal cost the utility avoids when one more kWh of solar is generated at a given location and hour. Inputs include the wholesale market price (CAISO LMP), the generation capacity value (how much the kWh helps avoid building new fossil generation), the transmission and distribution capacity value, and a small environmental adder. The ACC produces an hourly value for every location-and-hour combination; it averages around 5 to 8 cents per kWh annually but spikes to 30 to 50 cents during the summer 7-9pm grid-stress window.

Does NEM 3.0 apply to public power and CCAs?

NEM 3.0 applies to the three big investor-owned utilities (PG&E, SCE, SDG&E). Municipal utilities (LADWP, SMUD, Imperial Irrigation District, Roseville Electric) set their own net-metering policies and most still offer something closer to NEM 2.0 retail-rate net metering. CCAs use the underlying IOU's distribution tariff so CCA customers are on NEM 3.0 for the export-rate portion; the CCA generation credit replaces the IOU generation portion.

Is solar still worth installing in California under NEM 3.0?

Yes, but only with a battery. Solar-only systems under NEM 3.0 have payback of 11 to 15 years for most households, which is borderline given the 25 to 30 year panel warranty. Solar+battery systems pay back in 7 to 10 years for most households, well within the warranty, and the battery doubles as backup power during the increasingly frequent Public Safety Power Shutoff events. Bare solar without storage is now a poor choice for most California residential customers; bundle a battery or wait until the math improves.

Disclaimer. NEM 3.0 export rates vary by location and hour using the CPUC Avoided Cost Calculator; figures cited are illustrative annual-average values current to May 2026. Confirm exact rates for your installation address with your installer or the CPUC ACC table before signing a solar contract. Independent resource; not a solar installer and not affiliated with any solar company.