New York solar reshapes hourly load curve as billing fights spread from Nevada to California
Small-scale solar is flattening midday demand and sharpening evening peaks, forcing utilities to rethink rate structures that bill for peak usage.
Sunrun said its California originations have yet to regain pre-2023 levels following the state's transition to its Net Billing Tariff, known as NEM 3.0, and the company now faces further pressure from impending changes to the state's tax code, according to a filing reported on August 10. The disclosure puts a number on what utilities have spent years arguing: the way rooftop solar customers get billed shapes whether the industry grows or contracts.5
The underlying load-curve problem is sharpest in New York. EIA data published on June 26 show that small-scale solar now cuts metered demand by an average of 923 MW between 8:00 a.m. and 11:00 a.m. during March and April — a reversal from 2018, when demand in that window rose by an average of 850 MW over the same period.2
The evening swing is more striking. Early evening demand between 4:00 p.m. and 7:00 p.m. grew from an average increase of 681 MW in March and April 2018 to 2,221 MW by 2026, the EIA data show. The gap between the solar-driven midday trough and the post-sunset spike determines how much peaking capacity utilities must procure and how they recover fixed costs from customers.2
Roughly half of New York's 5.6 GW of PV capacity additions have been small-scale solar, the EIA data show, meaning the shift is coming from rooftops and parking-lot arrays rather than utility-controlled plants. That puts the load-shape problem outside direct utility planning authority, which is why rate design has become the main arena.2
Nevada offers the clearest recent test case. A state judge on May 27 rejected the attorney general's bid to block NV Energy's new daily demand charge, which bills southern Nevada customers based on their single highest 15-minute usage period each day. The ruling lets the utility shift cost recovery away from volumetric energy charges and toward peak capacity — a structure that hits solar prosumers who export during the day but draw heavily in the evening.1
California shows what happens when that shift goes further. The state's 2023 move to NEM 3.0 cut export compensation for rooftop solar customers. Sunrun's origination figures in California have not recovered since, and the company is now navigating potential additional pressure from state tax code changes flagged in its August 10 filing.5
The billing systems themselves are part of the constraint. GridX chief commercial officer Scott Engstrom argued in Utility Dive on August 7 that legacy utility billing platforms cannot model dynamic rates across entire service territories quickly enough to support innovation. A utility might design a new dynamic rate on August 3 but lack the software to assess its customer-by-customer impact the following day, slowing the pilots that could smooth the load curve.4
The market response has been to pair solar with storage rather than fight rate design directly. More than 90% of operational solar capacity globally remains standalone, but that share is falling as lithium-ion costs decline and AI-driven power demand accelerates, according to analysis published on July 1. McKinsey projects battery energy storage capacity will grow 50% annually to roughly 680 GWh by 2030.3
Sunrun is already positioning distributed batteries as a grid resource. In June, the company announced a nonbinding agreement with Renew Home and Tesla Energy to make more than 16 GW of distributed energy resources available to hyperscalers, aggregating home batteries into a product that can compete with utility-scale storage.5
Wall Street is drawing the same conclusion. Brookfield Asset Management, with more than $1 trillion under management, has been pivoting its solar exposure toward projects paired with storage, reflecting a view that standalone solar faces growing revenue risk as net metering weakens and peak pricing rewards dispatchable capacity.3
The arithmetic in New York is plain: small-scale solar has converted the morning ramp into a demand decline while nearly tripling the evening ramp. Utilities that cannot reprice that evening peak will either overbuild peakers or see load factors deteriorate as fixed costs spread over fewer volumetric kilowatt-hours. California's NEM 3.0 experience shows the transition is painful for installers. Nevada's May 27 ruling shows courts are not blocking the utilities' path.2,1
NYISO load data through next spring will show whether storage deployment is flattening the evening peak or merely shifting it later. Sunrun's California origination numbers in coming quarters will indicate whether the tax code changes accelerate the direct-sales model the company has adopted. Storage costs are the swing factor, and they are still falling.5,3