
Feature: A Million Water Heaters for Two Percent of the Smog
On July 2, a divided Ninth Circuit — 2 to 1, Judge Lucy Koh writing — upheld South Coast AQMD Rule 1146.2, the June 2024 rule requiring large water heaters, small boilers, and process heaters across Los Angeles, Orange, Riverside, and San Bernardino counties to go zero-emission — in practice, electric — on a schedule that runs from 2026 to 2033. The court rejected the industry's argument that the federal Energy Policy and Conservation Act preempts the rule, reasoning that process heaters aren't covered by EPCA and the Department of Energy has set no standard for them. Legally, it is a real ruling, and it will stand unless the Supreme Court takes it. The question this brief cares about is the one the ruling doesn't ask: what does the rule actually buy the air, and at what cost?
Start by being fair to it, because the honest version of this argument is the strong one. This is a NOx rule, not a climate rule. SCAQMD is not claiming to cool the planet. It is targeting nitrogen oxides — the precursor to the ozone smog that gives the South Coast Basin the worst air in the country. NOx is a genuine public-health problem, and cutting it is a legitimate goal. So judge the rule on its own terms — smog — and not on a target it never set for itself.
On those terms, here is the headline. The rule's backers say full implementation cuts NOx by 5.6 tons a day, which they describe as addressing "10% of NOx from stationary sources." The wire coverage compressed that into "nearly 10% of emissions in the basin." Those are two very different sentences, and the distance between them is the whole story.
Because stationary sources are not the basin's NOx problem. By the district's own inventory, roughly 80% of the basin's NOx comes from mobile sources — heavy-duty trucks, the ships idling at the ports, locomotives, aircraft, construction equipment. Every stationary source combined — power plants, refineries, factories, and yes, water heaters — is only about 15 to 20%. So "10% of stationary" is ten percent of one-fifth of the problem: under 2% of the basin's total NOx. The rule's own figure says the same thing a second way — 5.6 tons a day, set against a basin measured in the hundreds of tons a day. And even that 5.6 is not 2026's number. It is the fully-phased-in 2033 number, seven years out.
Watch how the comparison gets dressed up. Backers like to say 5.6 tons is "about half the NOx from all the cars in the region." Notice which cars: passenger vehicles — the cleanest, smallest slice of that 80% mobile share, not the heavy-duty trucks and port ships that actually dominate it. Measure the rule against the real source of the smog and it doesn't cover half of anything. It barely registers.
Now the price of that under-2%. The rule reaches about one million units across the four counties — the water heaters and boilers in apartment buildings, restaurants, hotels, laundries, hospitals, and factories. Every one of them eventually gets swapped for electric equipment, on the owner's dime, plus the electrical upgrades the buildings need to run it. That is why the challengers were not oil companies. They were the National Association of Home Builders and California's lodging, restaurant, and manufacturing groups — the people who have to buy the hardware and rewire the buildings. And the electricity to run a million new electric heaters has to come from a grid California is straining to keep lit — while the same policy climate discourages the in-state gas generation that still backs that grid up.
Then there is the part that should trouble anyone who actually cares about the basin's air. The largest single fixed NOx source in the region isn't a water heater — it's the twin ports of Los Angeles and Long Beach, which by themselves throw off about a fifth of the region's NOx, roughly ten times what this rule addresses, with ocean-going vessels alone making up 36% of the port total. And California is steadily feeding that source. Having cut its own oil production by roughly 20% and pushed refineries toward closure, the state now imports about 75% of its crude and record volumes of finished gasoline — a "fuel island" resupplied by a growing line of tankers, each arriving on the dirtiest NOx category at the dock. So one arm of environmental policy manufactures marine emissions at the coast by suppressing domestic supply, while another arm chases a 2% stationary slice inland. The two arms pull against each other — and the ships are winning.
This is the pattern this brief keeps documenting, and it is worth naming. Bonding rules that read as accountability but freeze investment. An LCFS that reads as climate policy but lands as a surcharge. And now a smog rule that reads as decisive — a million appliances, a federal-court win, "10% of emissions" — that delivers under 2% of the basin's NOx by 2033, while the ports, the freeways, and the rail yards that produce four-fifths of it run untouched. Not because anyone forgot about them, but because those sources are federal, mobile, and hard, and a water heater is local, stationary, and easy. You regulate what you can reach, then claim credit against the whole.
The smog is real. So is the cost. What's missing from the ruling — and from the coverage — is the line of the ledger where the two are set side by side. So here it is: cleaner air in the South Coast Basin runs through the 80%, at the ports and on the freeways. It does not run through a million water heaters, and a court upholding the rule does not change the arithmetic.
Sources: Reuters, Ninth Circuit opinion (25-5129), Earthjustice — 5.6 tons/day, 10% of stationary NOx, SCAQMD Rule 1146.2, SCAQMD 2022 AQMP — mobile source share, Port of L.A. — ocean-going vessel NOx, CEC — crude import sources
The Load Nobody Priced
The rule electrifies about a million units, and SCAQMD's own rulemaking concedes the grid "would likely not be able to handle the large increase." It then declined to say how large. So here is an estimate. Converting the covered water heaters, boilers, and process heaters adds on the order of 4 to 35 terawatt-hours a year to the region's grid — the low end if every unit becomes an efficient heat pump, the high end if they become electric resistance. Even the optimistic case is roughly 5% of everything Southern California Edison delivers in a year; the pessimistic case approaches a third of it. For scale, that is the output of three to nine large power plants, built to serve one air district.
Put a price on that. A large power plant runs roughly $1 to $2.5 billion to build in today's escalated market, so three to nine of them lands somewhere between $5 billion and $20 billion — and that is generation alone, before the transmission lines, substations, and neighborhood distribution upgrades needed to move the power, which in an electrification build-out routinely cost as much again. Call it what it is: a multi-billion-dollar capital program, set in motion by an air-district rule that carried no appropriation and no vote.
And here is who pays for it. Not SCAQMD, which wrote the rule and builds nothing. The bill lands in the rate base of Southern California Edison — and utilities don't merely recover the cost of what they build; they earn an authorized return on it, on the order of 10% a year, for decades. So every dollar of that build-out comes back on SCE customers' monthly bills, with a guaranteed profit stacked on top. Those customers already pay among the highest power rates in the country: SCE raised its average residential rate to 35.3 cents a kilowatt-hour in late 2025, up from 31.2, and its approved 2025–2028 revenue plan already runs to $41.6 billion. This rule adds to that pile. The households, restaurants, and apartment renters across the four counties — none of whom voted on Rule 1146.2 — pay for the new capacity, pay the utility's return on that capacity, and pay it on a bill that was already climbing. The air district gets the credit. The ratepayer gets the invoice.
And whether the swap even lowers total energy use is not guaranteed. Heat pumps genuinely cut it — three units of heat per unit of electricity. But the boilers and high-temperature process heaters this rule also covers frequently can't use a heat pump; they convert to electric resistance, which — drawn from California's marginal gas-fired generation (~45% efficient, plus line losses) — burns more primary energy than the gas appliance it replaced. That doesn't eliminate the combustion. It relocates it — NOx and all — from a rooftop in the basin to a power plant, sometimes a gas peaker firing to run a "zero-emission" boiler. Zero at the meter is not zero on the grid.
Load and cost estimates by CA4ES from SCAQMD unit counts and standard efficiency and capital-cost assumptions; SCAQMD published no grid-load figure. Sources: CEC — California Energy Demand Forecast, CAISO — Loads & Resources, EIA — power-plant heat rates, EIA — generator construction costs, CPUC — SCE 2025 General Rate Case, SCAQMD PAR 1146.2 staff report
In the News
Ninth Circuit upholds L.A.-area gas-appliance ban in a 2–1 split — Reuters
The appeals court upheld SCAQMD's zero-emission rule for large water heaters, small boilers, and process heaters, ruling that the federal Energy Policy and Conservation Act doesn't block states from regulating appliance emissions. The dissent is the part to watch: Judge Kenneth Lee called the case "strikingly similar" to the 2024 decision that struck down Berkeley's gas ban on preemption grounds — a split that could carry the fight toward the Supreme Court. (See this week's feature for what the rule actually does to basin emissions.)
Energy Realism: The Foundation and Future of Human Progress — Dr. Scott Tinker, The Epoch Times
The University of Texas energy scientist makes the case for "energy realism" — rejecting the false binary between fossil fuels and renewables, and judging policy by whether it delivers affordable, reliable energy at scale. "Energy is a means, not an end," Tinker writes; "affordable and reliable energy lifts people from poverty." It's the frame underneath everything this brief tracks: policy should be measured by what it delivers to real people, not by the intentions it signals.
California's gas tax takes effect — and this time Sacramento defends it — Fox News
The excise tax rose to 63.4 cents a gallon on July 1, and the Governor's office answered its critics for the first time — calling the increase "an automatic, annual mechanism enacted in 2017 and upheld by voters in 2018," and arguing that repeal "would hand oil companies a massive tax break." The response is most notable for its concession: to make the argument, the state had to confirm that roughly $1.15 of every California gallon is Sacramento's own doing.
Ahead of July 4th, drivers get some relief at the pump — AAA
The national average slid to about $3.80 for the holiday weekend — the cheapest Fourth of July gas in years for most of the country — as crude fell on the Strait of Hormuz recovery. California, near $5.40, stayed about $1.60 above the national average, its own July 1 increase pulling the wrong way exactly as national relief peaked.
Brent holds near pre-war levels as Hormuz shipping recovers — Trading Economics / Reuters
Brent traded around $72 in thin holiday-week trading, near levels last seen before the Middle East conflict, as tanker traffic through the Strait of Hormuz continued to rebuild; the EIA expects shipments to keep ramping through the third quarter. The largest single input to California's pump price has now fully unwound its war premium — leaving the state-built surcharge in unusually plain view.
Sable fight seen heading toward the state's highest courts — Santa Barbara Independent / BOE
With the Coastal Commission injunction upheld on appeal and the federal-preemption question unresolved, court watchers increasingly expect the Santa Ynez Unit dispute to reach the California Supreme Court — likely after the Ninth Circuit rules on whether federal pipeline authority preempts the state. The fight over roughly 60,000 barrels a day of domestic California crude grinds on across every venue at once.
At the Pump
Week of July 7, 2026 — Source: AAA, EIA
California | National Avg | Gap | |
|---|---|---|---|
Regular Gasoline | $5.40 | $3.80 | +$1.60 (+42%) |
Diesel (approx.) | ~$6.45 | ~$4.95 | ~+$1.50 (+30%) |
California's average for regular slipped a few cents to about $5.40, while the national average fell harder — down roughly 11 cents to about $3.80, the cheapest Independence Day gas in years for most of the country. Both moved on the same input, and again it wasn't California: Brent held near $72, roughly pre-war levels, as Strait of Hormuz shipping continued to recover. Two things stand out this week. First, because the national number fell faster than California's, the gap actually widened to about $1.60 a gallon — California now pays roughly 42% more than the national average, even though at the pump it has slipped just behind Hawaii ($5.48) for the most expensive market. Second, the July 1 increase is now in this number: the 63.4-cent excise tax and the LCFS change took effect, and the state's own accounting puts total California taxes and fees near $1.15 a gallon — the built-in premium, landing the same week the rest of the country got its cheapest fill-up in years. (National diesel is an EIA-directional estimate; CA diesel of ~$6.45 is AAA's California average, easing with crude.)
Data: AAA — California, AAA — Some Relief at the Pump, EIA Weekly Retail
Import Watch
The relief is holding, and it is still entirely borrowed. Brent near $72 and a recovering Strait of Hormuz have unwound the war premium the barrel carried all spring — but the EIA is explicit that shipping is still ramping, not restored, and expects it to take months to reach pre-conflict traffic. None of that recovery added a single domestic barrel to California's supply. It simply made the foreign barrel the state already depends on cheaper for a while.
The structure underneath is unchanged. California still refines about 75% imported crude. It still leans on the federal Jones Act waiver harder than any state — absorbing more than 60% of the gasoline cargoes moved under it, roughly 3 million barrels — and that waiver still expires mid-August, now about six weeks out, with no permanent replacement behind it. The one new domestic source off its own coast, Sable's roughly 60,000 barrels a day, remains frozen by an injunction the state is defending toward the Supreme Court. So the picture is the same one we've flagged for weeks: when the borrowed relief fades — a ramping strait and an expiring waiver are both temporary — California is left exactly where it started, at the end of the longest supply lines in the country, having spent the calm week adding cost and rules at home instead of supply.
Sources: Trading Economics, EIA STEO, CEC Oil Supply Sources, MARAD Waiver Report
Calendar
Date | Event | Why It Matters |
|---|---|---|
July 2, 2026 | Ninth Circuit upholds SCAQMD Rule 1146.2 | The gas-appliance ban survives federal-preemption challenge 2–1. Compliance runs 2026–2033 across ~1 million units; a dissent citing the Berkeley precedent leaves the door open to a Supreme Court appeal. |
Q3 2026 | Strait of Hormuz shipping ramp | The EIA expects tanker traffic to keep recovering through the third quarter, but not to reach pre-conflict levels for months. The barrel California imports stays cheap only as long as this holds. |
Ongoing | Sable litigation | With the injunction upheld on appeal, watchers expect the Santa Ynez Unit fight to reach the California Supreme Court, likely after a Ninth Circuit preemption ruling. ~60,000 b/d of domestic crude hangs on it. |
Mid-August 2026 | Jones Act waiver expiration | The DHS extension expires in roughly six weeks. California absorbs the largest share of fuel moved under it; without renewal, the state's supply chain tightens just as the Hormuz relief is still ramping. |
The CA4ES Energy Brief is published weekly by Californians for Energy & Science, a 501(c)(3) nonprofit. Data-driven energy intelligence for Californians.
energyandscience.com | [Unsubscribe]