Does a Battery Actually Pay Back Under NEM 3.0 in California?

The core math: export credit versus peak retail

Every battery-payback question under NEM 3.0 reduces to one comparison: what you're credited to export a kilowatt-hour versus what you'd otherwise pay to buy one back in the same hour. The export side is set hourly from CPUC-approved avoided-cost values, which the CPUC says are usually lower than the retail rate. The buy-back side is whatever your rate plan charges at that hour, highest in the evening peak window.

A battery captures the difference by storing midday solar you'd otherwise export and using it that evening instead of buying grid power. It does not capture value by exporting more. NEM 3.0 vintage also matters: the CPUC says residential PG&E and SCE customers who apply to interconnect before the end of 2027 receive slightly higher-than-normal bill credits for exported energy for nine years, and SDG&E customers are excluded. The CPUC also says the original customer is guaranteed use of the tariff for nine years. None of this changes the fundamental shape of the trade.

PG&E: the E-ELEC rate

The CPUC lists E-ELEC as PG&E's approved rate for Net Billing Tariff customers. PG&E's residential average rate was 33.7 cents per kWh as of June 2026, according to the CPUC Public Advocates Office, but an average hides the hourly prices a battery works against.

Pull the E-ELEC price for each time-of-use period from PG&E's current tariff sheet and the hourly export values from PG&E's published export-credit tables. Compare the evening price with the midday export value. That difference, times the kilowatt-hours your battery actually shifts each day, is the yearly value. A proposal that skips those two inputs has not done the PG&E math.

SCE: check which time-of-use plan you are on

The CPUC lists TOU-D-PRIME as SCE's approved rate for Net Billing Tariff customers. SCE's residential average rate was 34.4 cents per kWh as of June 1, 2026 (CPUC Public Advocates Office). SCE's other time-of-use plans, such as TOU-D-4-9PM and TOU-D-5-8PM, have different peak windows and prices, so check which plan your bill names before you run the numbers, and take the prices from SCE's current rate pages rather than from a proposal's summary.

The early-adopter export adder the CPUC describes applies to residential SCE customers who apply to interconnect before the end of 2027. Ask the installer whether its export values include it.

SDG&E: highest average rate, no export adder

The CPUC lists EV-TOU-5 as SDG&E's approved rate for Net Billing Tariff customers. SDG&E had the highest residential average rate of the three utilities, 45.5 cents per kWh as of June 1, 2026 (CPUC Public Advocates Office), and its customers are excluded from the early-adopter export adder that PG&E and SCE customers receive.

A high retail rate widens the gap a battery can capture, but the evening price on your own plan is what matters, so read it from SDG&E's current rate tables and compare it with the hourly export values SDG&E publishes.

SGIP won't help you pay for it

The state rebate that used to soften battery costs is not available to a typical homeowner now. The program administrators' tracker (selfgenca.com, as of September 22, 2026) shows the Small Residential Storage category closed at Step 7 ($0.15 per Wh) for CSE, SCE, SoCalGas and PG&E customers. The ratepayer-funded Residential Solar and Storage Equity category ($1.10 per Wh) and Equity Resiliency ($1.00 per Wh) also show closed. The AB 209 Residential Solar and Storage Equity category ($1.10 per Wh) shows a waitlist for CSE, SoCalGas and LADWP, and it is limited by income and other eligibility rules.

At $0.15 per Wh, the closed general category would have paid about $2,025 on a 13.5 kWh battery. If an installer's proposal still counts an SGIP rebate for your project, ask it to show you, in writing, which category you qualify for and its current status. A waitlist does not promise a rebate.

The federal credit is gone for owned batteries — a PPA is a different calculation

The federal residential clean energy credit (Section 25D) — which covered both solar panels and owned battery storage — does not apply to expenditures made after December 31, 2025, per the IRS. If you buy and own a battery outright now, there is no federal tax credit offsetting its price.

A third-party-owned system — a lease or PPA where the provider keeps title — is a different calculation. The owner, not you, may be able to claim the commercial Section 48E credit, and whether any of that value reaches your payment is a contract question, so ask for it in writing. The tradeoff is counterparty risk — Freedom Forever converted to Chapter 7 liquidation in mid-2026 after filing Chapter 11 that spring, Sunnova filed Chapter 11 in mid-2025 and is now administered by a legacy-portfolio servicer rather than writing new contracts, and SunPower filed Chapter 11 in 2024. If you sign a 20-year PPA, the company on the other end of it needs to still exist in year 15.

How to run the payback math yourself

No primary source publishes a payback period for a NEM 3.0 battery, and this page does not invent one. The arithmetic is simple enough to run on your own numbers.

First, kilowatt-hours shifted per day: the midday solar your battery stores that you would otherwise export, capped by the battery's usable capacity and by your evening use. Second, value per kilowatt-hour shifted: your plan's price in the hours you would otherwise buy from the grid, minus the export credit you give up for that kilowatt-hour. Third, yearly value: kilowatt-hours shifted per day, times value per kilowatt-hour, times 365. Fourth, simple payback: the installed price, less any rebate you have in writing, divided by the yearly value.

For scale, LBNL's 2023 storage premium of $750 to $1,000 per kWh puts a 13.5 kWh battery at roughly $10,000 to $13,500 on top of the solar price. Ask each installer to show those four lines with your own twelve months of usage. A proposal that shows only a payback year is incomplete.

What actually moves your number

Four things swing these numbers more than the utility does. First, how much excess solar your system produces relative to your household load — a battery can only shift power you're actually generating and not using, so an undersized solar array starves the arbitrage. Second, whether you charge the battery from solar only or also from off-peak grid power, which changes which spread applies. Third, financing cost — interest on a battery loan eats into or erases the yearly value; a cash purchase or a very low-rate loan preserves it. Fourth, EV charging: if you're charging a vehicle at home, that load competes with the battery for the same stored solar, and the math has to account for which use gets priority.

None of these four are the same for any two households, which is exactly why a generic 'batteries pay back in X years' claim from an installer should be treated as a sales pitch to verify against your own bill, not a fact to accept.