Solar for California Retail: Shopping Centers, Strip Malls, and Restaurants

Demand Charges Are the Real Cost Driver, Not the kWh Rate

Every major California utility bills commercial retail accounts on a structure that layers a demand charge — priced on your single highest 15-30 minute spike in kilowatts that month — on top of the per-kWh energy rate. This is the single biggest reason commercial solar economics differ from a home system, and it hits shopping centers, strip malls, and big-box stores squarely because most sit in the 20-500 kW demand range where these charges apply.

PG&E uses tiered schedules B-1 (small accounts, low or no demand charge), B-10, B-19, and B-20, distinguishing secondary, primary, and transmission service voltage — accounts running above roughly 499 kW of demand for three consecutive months move to B-19 or B-20. PG&E's own newsroom confirms a March 2026 "Base Services Charge" restructuring, but that change separates a fixed monthly charge from the per-kWh rate for residential customers only — it does not touch these commercial demand-charge schedules, and as of this writing there's no confirmed change to the B-19/B-20 structure itself. SCE uses TOU-GS-1 for under roughly 20 kW (generally no demand charge), then TOU-GS-2 (roughly 20-200 kW), TOU-GS-3 (roughly 200-500 kW), and TOU-8 (500 kW+), layering a year-round Facilities-Related Demand charge with a Time-Related Demand charge tied to on-peak usage — SCE's commercial rates rose about 7.2% in March 2024 and about 10% more in October 2025, cumulative over 80% in the past decade. SDG&E's nonresidential schedule, AL-TOU, applies once monthly maximum demand hits 20 kW, with a secondary-voltage basic service fee of $213.30/month at or under 500 kW and $766.91/month above that, plus demand charges layered on top, per SDG&E's own rate table effective October 1, 2025.

Exact current dollar-per-kW figures vary by tariff sheet and change with each rate case — don't quote a single statewide number as universal. Pull your own demand-charge line items off a recent bill before anyone sizes a system for you; a proposal built only from your kWh usage is missing the number that usually matters most.

NEM 3.0 and Daytime Load: Why Timing Beats Size

Under the Net Billing Tariff, in effect since April 15, 2023, export credits are set by the CPUC's Avoided Cost Calculator rather than retail rates — a cut of roughly 75% from NEM 2.0's average, down to somewhere around 5 to 8 cents per kWh, and that rate swings by time of day: it's close to zero at midday when solar production peaks, and rises toward the evening on-peak window.

That structure rewards a system built to serve your own load, not one built to sell power back. Retail buildings typically operate roughly 9am to 9pm, sometimes 24 hours for certain formats, which overlaps solar production hours reasonably well on its own. A battery that stores midday solar and discharges it during the 4-9pm on-peak window adds a second layer of value, because that window often coincides with a retailer's highest foot traffic, heaviest lighting and HVAC load, and highest demand-charge exposure. A system sized well past your own daytime consumption is exporting a growing share of its output into a near-worthless midday rate — right-sizing to load, not to roof space, is the retail-specific rule.

Restaurants: The Best Load-Match Retail Format

Restaurants run refrigeration (walk-ins, reach-ins, ice machines), kitchen exhaust hoods and make-up air units, and HVAC essentially continuously through prep-to-close hours, which lines up with solar production hours better than a store that's only busy in the evening. That said, treat any specific savings figure with caution: one vendor-published case study cites annual solar savings in the $25,000-$50,000 range for restaurants, but that's a single source's illustration of its own client results, not an independently verified or typical outcome — don't quote it as an average for your building.

Battery storage has a second, separate case for restaurants beyond bill savings: keeping refrigeration, lighting, and point-of-sale running through an outage protects perishable inventory and avoids a forced closure, which matters more for a kitchen than for most other retail formats. Size the storage conversation around your own demand-charge exposure and outage risk, not around a vendor's marketing number.

Who Owns the Roof: Landlord, Tenant, and Virtual Net Billing

Most shopping centers and strip malls have one owner and several tenants on separate utility meters, which raises a question a single-tenant building doesn't have: whose meter gets credited for the solar? Virtual Net Billing Tariff (VNBT), which replaced the prior Virtual Net Energy Metering framework for new applicants starting February 15, 2024, is the CPUC mechanism built for exactly this — a property owner with one array serving a multi-meter property allocates the generation's bill credits across tenant meters and common areas under an allocation agreement filed with the serving utility (PG&E, SCE, or SDG&E territory). As of this writing there's no stated application deadline.

VNBT solves the metering and credit-allocation mechanics — it does not by itself decide who pays for the system or who keeps the tax benefits. The property owner, as the asset owner, is the one who captures the ITC and depreciation regardless of how credits get allocated to tenants. If you're a landlord considering this, confirm the allocation agreement and application process directly with your utility (or your utility's own VNEM/VNBT program page) before assuming the mechanics work like standard net metering — they don't.

Triple-Net Leases and C-PACE: Solving the Split Incentive

In a standard triple-net lease, the tenant pays the utility bill directly, which means the tenant — not the landlord who would front the solar cost — captures the operating savings. That mismatch is the reason many landlords never look seriously at solar on leased retail buildings, and it's a real, structural problem, not a sales objection to argue past.

Commercial PACE (C-PACE) financing is the mechanism built to address it: the C-PACE assessment is billed through the property tax bill, which most NNN lease structures already treat as a passable operating expense — so the tenant ends up paying the financing cost as a tax/CAM pass-through while also receiving the utility bill savings, and the landlord keeps the ITC, depreciation, and any renewable energy credits without financing the system out of pocket. This is a real and increasingly common structure, but California's C-PACE availability is opt-in by county and city, administered by different programs in different jurisdictions — confirm which program actually covers your property's location before assuming C-PACE is available there.

Carports vs Rooftop for Parking-Lot Retail

Solar carports cost meaningfully more than rooftop. EnergySage's own 2026 marketplace data puts the national average at $3.62 per watt installed for carports versus $2.60 per watt for rooftop — roughly a 39% premium — driven by engineered steel structures, foundations, wind and seismic load calculations, and the added permitting and interconnection study work commercial carport jobs require. That EnergySage figure is a national marketplace average, not California- or commercial-scale-specific; a steel canopy sized for a shopping-center parking lot, with the added seismic engineering California requires, should be expected to run above that average — get a project-specific quote before budgeting off any single published per-watt number.

The case for paying the premium: a shopping center's rooftop is often already crowded with HVAC units, grease vents and kitchen exhaust, refrigeration lines, and communications equipment, and rooftop space may be split or contested across multiple tenants in ways parking-lot square footage isn't. Carports also pair naturally with EV charging for customers, adding a visible amenity a rooftop system can't offer. If your roof has clean, contiguous, structurally sound space and no near-term membrane replacement scheduled, rooftop is the cheaper starting point — carports earn their premium when the roof genuinely can't take the array, not as a default upgrade.

Roof Realities for Big-Box and Strip-Mall Buildings

No special roof type is required for solar, but the roof has to carry the array for the panel and inverter warranty period, typically 15 to 25 years — and many big-box and strip-mall buildings are built to code-minimum structural strength with roof membranes that get replaced on a 5-10 year cycle. That mismatch means racking and modules may need to come down and go back up mid-system-life, a real cost and logistics factor that a fast rooftop-solar sales pitch tends to skip over. Get a structural assessment and a roof-membrane-remaining-life estimate before signing, and ask specifically who pays for removal and reinstallation if a re-roof falls inside the system's life.

Retail and big-box rooftops also already carry HVAC units, grease vents and kitchen exhaust, refrigeration lines, and communications equipment that any array has to be engineered around — plan the layout around what's already up there, not around an assumption of open roof space.

The Incentive Clock: 48E ITC, Bonus Depreciation, SGIP, and Property Tax

The federal Section 48E investment tax credit phases out for solar placed in service after December 31, 2027 unless construction begins on or before July 4, 2026, per the One Big Beautiful Bill Act (signed July 4, 2025) — projects that start construction by that date can use a continuity safe harbor reaching a December 31, 2029 or 2030 placed-in-service deadline depending on exactly when construction started. Systems over 1.5 MW starting construction on or after September 2, 2025 face a stricter physical-work-only test rather than the easier cost safe-harbor, per Notice 2025-42 — most retail rooftop and carport jobs fall under that 1.5 MW threshold. 100% bonus depreciation is now permanent for property placed in service in the first tax year ending after January 19, 2025, and it can stack with the ITC in year one, though standard MACRS convention requires reducing depreciable basis by half the ITC claimed. Credit transferability (selling the ITC for cash) survives OBBBA too, which matters if your business doesn't have tax appetite to use the credit directly — ask about foreign-entity-of-concern restrictions on the supply chain if that route is being discussed.

SGIP's commercial storage incentive is currently closed: per SelfGenCA's own program-metrics page, the Large-Scale Storage budget (covering non-residential storage of essentially any size) is closed to new funded applicants in every territory as of September 5, 2026, at Step 5, $0.25/Wh, with new applications going to a waitlist funded only by attrition. Separately, California's Active Solar Energy System property-tax exclusion is scheduled to sunset January 1, 2027 absent a legislative extension — a large commercial system adds meaningfully more assessed value than a home system, so completing before that date matters more here. Whether a pending extension bill has actually passed is unresolved as of this writing; don't assume it has.