California Cut What It Pays for Rooftop Solar. Did It Fix the Right Problem?

California changed how it pays rooftop-solar owners in 2023.

Under the state’s Net Billing Tariff — commonly called NEM 3.0, adopted by the California Public Utilities Commission in December 2022 and applied to customers who applied to connect on or after April 15, 2023 — exported electricity is no longer credited at roughly the same retail rate customers pay to buy electricity. Instead, compensation is based on an estimate of what that electricity is worth to the grid at the hour it arrives.

And that is, in large part, the economically sensible thing to do.

Same electricity, different product

Electricity is unusual because its value changes dramatically depending on when it is produced.

Imagine two solar homes. One exports electricity at 1 p.m. on a mild spring day, when thousands of other panels are doing exactly the same thing. The other sends stored electricity onto the grid at 7 p.m. in September, when solar generation has faded and demand is still high.

Those two kilowatt-hours are physically identical. Economically, they aren’t. Here is what PG&E’s 2026 export schedule actually pays on a weekday:

The September evening export is worth about 150 times the April afternoon one. On August evenings, the credit is roughly double the 55-cent peak price PG&E customers pay to buy power. See all 24 hours in the interactive chart.

California’s current system recognizes this. Export credits are usually below retail rates but can rise above retail on late-summer evenings. The state explicitly encourages pairing solar with batteries for this reason — and people responded: nearly 70% of new solar customers had added storage by the end of 2024, compared with about 10% under the old rules.

Why California changed it

There were two stated motives. One was the price signal: midday solar was flooding the grid while evening demand went unmet. The other was cost. When solar owners were credited at the full retail rate, the fixed costs of running the grid were shifted onto customers without solar. The commission’s ratepayer advocate estimated that shift at $8.5 billion in 2024. The solar industry disputes the method, arguing it undervalues rooftop solar’s benefits. Both arguments deserve a hearing; the size of the cost shift depends on accounting choices people genuinely disagree about.

The part that isn’t fixed yet

The more interesting question is what California hasn’t fixed.

Retail electricity prices contain far more than the short-term cost of producing another kilowatt-hour. Customers pay for generation, transmission, distribution, wildfire-related spending, public programs and other system costs. California’s Legislative Analyst estimates that two-thirds to three-quarters of what the big utilities recover through rates are fixed costs, and that residential rates at the three large investor-owned utilities rose 48–67% between 2019 and 2023, far faster than inflation.

So California has moved rooftop-solar compensation toward the grid’s avoided cost while consumers still face an expensive bundled retail price. That creates a strange situation. The state correctly tells a homeowner:

Your noon electricity isn’t worth the full retail price.

But the same system may still charge that homeowner 55 cents a kilowatt-hour at peak when electricity flows in the other direction.

To be fair, California has started on the other half. In May 2024, the commission approved a monthly fixed charge — $24.15 for most customers, less for low-income households — designed to move some fixed grid costs out of per-kilowatt-hour prices and lower them by an estimated 5–7 cents. It took effect in late 2025 and early 2026. Whether it meaningfully lowers the price of electricity at the margin is one of the things to watch.

The long-term opportunity: batteries

Thousands of home batteries can be coordinated as a virtual power plant. Instead of building one large facility that runs only during shortages, software can call on thousands of distributed batteries at once.

This is no longer hypothetical. On July 29, 2025, more than 100,000 California home batteries, aggregated by two companies, delivered an average of 535 megawatts to the grid from 7 to 9 p.m. in a statewide test. A Brattle Group study estimated California’s virtual power plants could exceed 15% of peak demand by 2035 — a modeled potential, not a result.

The right question

So the right question isn’t simply: How much should California pay solar owners?

It is: How much is electricity actually worth at the moment it is delivered?

• Pay low-value electricity less.
• Pay genuinely scarce electricity more.
• Price transmission and distribution transparently.
• And if a neighborhood full of batteries can replace part of the need for expensive peak generation or grid upgrades, compensate it for the value it actually provides.

Part of that already exists: the export credit includes a component for avoided transmission and distribution costs, and batteries can earn payments through state demand-response programs. How far it goes is the open question.

That is a price signal. And price signals work best when they tell the truth — on both sides of the meter.