NEM 3.0: solar-killer or grid savior?

In April 2023 NEM 3.0 went into effect, transforming the economics of residential solar and energy storage in California virtually overnight. Six months later we are still seeing a backlog of systems grandfathered into NEM 2.0 getting installed, keeping overall solar volumes roughly flat year-over-year, but installations are expected to fall 30–50% as soon as next year.

We took a deep look into how NEM 3.0 works, and how it will impact the US’s largest residential solar market.

What is NEM 3.0 and why was it implemented?

• NEM 3.0 is California’s updated tariff schedule for residential solar and storage. It replaced NEM 2.0, which had been in effect from July 2017 to April 2023
• While NEM 2.0 was designed to accelerate solar adoption, NEM 3.0 is designed to mitigate the impact of the “duck curve” and better match the dispatch of distributed resources with the needs of the grid
• The “duck curve” refers to the over-generation of power in the middle of the day in markets with high solar penetration, when the sun is up and grid demand is low. When the workday ends demand on the grid goes up, but solar resources are no longer available, leading to an abrupt ramp in net energy generation requirements

• Unlike NEM 2.0, under which consumers are compensated 1:1 at retail rates for power exported to the grid, NEM 3.0 uses “net billing,” which compensates consumers at a lower (wholesale) rate. These export rates are designed to reflect the avoided cost of supplying additional energy to the grid in a particular hour, and have a large range from 3c/kwh during the peaks of solar production to >$2.6/kwh during late evening hours in September (when demand ramps as everyone turns on air conditioners, but supply dips as the sun goes down)
• NEM 3.0 also uses a time of use rate structure, with a higher differential between peak and off-peak pricing than rates used under NEM 2.0; this should help better match grid demand with the supply of renewables, for example, encouraging people to charge their EVs in the middle of the day at work rather than overnight
• Over the course of a year there are only 3 hourly blocks (7–8p in August, 6–7p and 7–8p in September) where export rates exceed import rates, which implies only ~91 hours where it would make sense for a residential user to export power to the grid rather than just use it for self-consumption or to charge a home battery

What do these changes mean in practice?

• We modeled the impact of NEM 3.0 and NEM 3.0 on a household’s net energy use and bill over the course of a year to quantify the impact of these tariff changes at the household level. The homeowner has an EV and solar panels, and energy costs are shown with and without energy storage attached
• Below you can see examples of household net energy consumption on a typical day in June, as well as the implied energy charges on an hourly basis under NEM 2.0 vs 3.0. The spike in net electricity use in the early hours is from charging the EV overnight, and the dip midday is due to solar generation offsetting any power use. Given that the export rates in June are lower throughout the day than import rates, the rational thing for the household to do would be to either use stored power for self generation or elect not to cycle the battery at all. The example below shows the latter, where the homeowner does not use their battery, and the significantly lower cost savings under NEM 3.0 relative to NEM 2.0

• Running the same analysis on a September day, however, shows a very different picture because of the high export rates. Here, a homeowner would choose to discharge their battery completely between 6–8p PT. However, most energy storage systems available in the market today have discharge limits well below the battery capacity — for a Tesla Powerwall 2 there is a 5kW discharge limit relative to the 13kWh capacity, for example. These limits significantly cap the value a homeowner is able to capture, and the examples below are shown with and without those limits on discharge

• In response to the new market rules, OEMs are launching products in the market with higher discharge limits — the Tesla Powerwall 3 (which, unfortunately, is not available to order today… or anytime soon) has an on-grid power discharge limit of 11.5kW
• Over the course of a year, the savings differential between NEM 2.0 and NEM 3.0 add up considerably, nearly doubling the payback period for a solar-only installation. However, assuming no (or higher) discharge limits, a solar + storage installation has similar or faster payback than a solar-only installation under NEM 3.0. It’s important to note that this analysis does not factor in (1) future rate hikes, (2) capex cost declines, (3) annual reductions in export adders, or (4) resilience / backup benefits of energy storage

Putting it all together

• NEM 3.0 will reduce residential solar demand in California by 30–50%, virtually overnight, once the existing backlog of NEM 2.0 systems is cleared. Wood MacKenzie estimates annual volumes in 2024 to be half of 2021 volumes, the lowest it’s been since 2014
• Storage attach rate is estimated to go up from 10–15% to 70–80%+, increasing AOVs for installers
• However, the analysis shown here does not factor in the reduction in payback period with future rate increases. PG&E has increased rates ~6% YoY for the past 15 years, with the annual increase accelerating over the past 4 years to an 8% CAGR. It also doesn’t account for the resilience benefit of solar & storage; given the increased unreliability of the grid consumers are often making this decision for resilience purposes rather than time-of-use savings

On balance, it’s clear that the next couple of years will be tumultuous for the US residential solar sector, with additional headwinds from interest rates and long timelines for implementation of IRA incentives. Longer-term, the goal of market design changes like NEM 3.0 is to make solar adoption more sustainable for the grid overall, and help make DERs part of a more balanced and flexible future energy system.

If you’re building in the home electrification space, do reach out to us at G2 Venture Partners! We have been investing in the space for over a decade, and are current or former investors in companies like Enphase, OPower, Nest, Arcadia, and 1komma5.