Independent guide. Based on representative IOU commercial tariffs across major US service territories.Verified May 2026
Commercial demand charges 2026: why kW matters more than kWh for many bills
For most US commercial electricity customers above about 25 kW peak demand, the demand charge (billed in dollars per kilowatt of peak draw) is a larger line item than the energy charge (billed in cents per kilowatt-hour of total use). This page covers how demand charges are calculated, the typical $/kW ranges across the major ISO/RTO regions, the ratchet clauses that protect utilities, and the battery peak-shaving business case that has become the highest-leverage cost-reduction move for many commercial facilities.
How a demand charge actually works
On a residential bill, the only variable is total kWh used in the month; the bill is rate per kWh times kWh, plus a small fixed charge. On a commercial bill above the small-business cutoff (typically 25 kW peak demand, varying by utility), the bill splits into two parts. The energy charge is the same rate per kWh times kWh, often at a slightly lower per-kWh rate than residential. The demand charge is a dollar per kilowatt figure ($5 to $35 per kW per month, depending on tariff) multiplied by the facility's peak demand for that month.
Demand is measured as the highest average kW over a defined interval (typically 15 or 30 minutes) anywhere in the billing month. The meter records every 15-minute interval, and the utility's billing system extracts the highest one. A facility that briefly runs at 200 kW for 15 minutes on a single day sets the demand charge for the whole month at 200 kW, even if average load is 60 kW for the other 2,975 fifteen-minute intervals in the month. This is the core reason demand charges are sometimes called a "peak penalty" by facility managers.
Why utilities bill this way
The cost of operating the grid scales with peak load, not total energy. A 200 kW peak requires a 200 kW capacity at the substation transformer, 200 kW worth of conductor, 200 kW worth of upstream transmission, 200 kW worth of generation reserve. Building that capacity is expensive whether the customer uses it for 15 minutes or 15 hours per month. Charging only by kWh would mean two customers using the same monthly energy (one with a flat profile, one with a sharp peak) pay the same bill, even though the second is much more expensive to serve. The demand charge corrects for this.
From a societal standpoint, demand charges create an incentive to flatten load profiles, which reduces the system-wide need for peaker generation, transmission upgrades and battery storage. From an individual facility standpoint, demand charges create a strong reason to invest in load management: a facility that can shave 50 kW off its peak saves real money every month for the life of the facility, not a one-time savings.
Typical demand charge ranges by region
Demand charge rates vary widely by utility and tariff. Representative ranges as of 2026: PG&E B-19 (medium commercial 75 to 499 kW) charges about $24 per kW per month on the summer maximum demand and an additional time-of-use peak demand charge of $24 per kW for any peak-window demand. SCE TOU-GS-3 (medium business 200 to 499 kW) charges about $19 per kW non-coincident plus $15 per kW for time-related peak demand. ConEd SC9 (large general service) charges about $32 per kW per month. Florida's FPL GSDT-1 charges about $11 per kW per month. Northeast IOUs (Eversource, National Grid, PSE&G) generally run $12 to $25 per kW per month with seasonal multipliers in summer.
The pattern: California and the Northeast charge the highest demand rates because their generation and transmission costs are highest. Southeast utilities (Duke, Southern, FPL) charge lower demand rates but make up some of the difference in higher fuel cost recovery components on the energy charge. Midwest utilities (Ameren, ComEd commercial tariffs, Xcel) sit in the middle. Always confirm by reading the actual tariff schedule for the specific commercial rate class your facility falls under; the rates are public but they are buried in 50-page tariff documents.
The ratchet clause: a trap for seasonal businesses
Many utility commercial tariffs include a demand ratchet: the billed demand each month is the higher of the actual demand or a fraction (typically 50, 75 or 80 percent) of the highest demand recorded in the past 11 or 12 months. The ratchet is intended to recover capacity cost from facilities that have a seasonal peak (a ski resort in winter, a HVAC contractor with summer peak) but use the grid lightly the rest of the year.
The trap: a facility that has one anomalous high-demand month (a brief equipment malfunction, a one-time event, a new chiller running incorrectly during commissioning) gets locked into paying for that demand level for the next 11 months even if the underlying load returns to normal. A 200 kW spike from a chiller malfunction on a single day in July, at a 75 percent ratchet and $20 per kW demand rate, translates to about 150 kW of ratcheted billing demand at $3,000 per month for 11 months, or $33,000 of unnecessary cost. Facility managers responsible for demand-charge bills should review the monthly interval data and identify any anomalies before they ratchet in.
Worked example: 50 kW small business
A retail store in PG&E B-10 territory averages 30 kW load with a 50 kW peak around 2pm on summer afternoons (when both the HVAC and the cooler refrigeration are running hardest). Monthly kWh: about 12,000. Monthly demand charge at $17 per kW: $850. Monthly energy charge at 18 cents per kWh: $2,160. Total: about $3,000 per month, with the demand charge representing about 28 percent of the bill.
Mitigations for this size of facility: stagger HVAC compressor start-ups by 5 to 10 minutes (uses about 75 percent of nameplate during operation but full nameplate during start-up), pre-cool the store before the 4pm peak window if on TOU, ensure the cooler refrigeration runs in defrost mode during off-peak hours rather than reactive mode. A small (20 kWh) battery costs about $20,000 installed and could shave 10 to 15 kW off the peak, saving $170 to $255 per month or $2,000 to $3,000 per year. Payback around 7 to 9 years; not the strongest investment unless paired with other ROI (resilience, EV charging buffer).
Worked example: 500 kW mid-size facility
A small manufacturing facility in SCE TOU-GS-3 territory operates two production shifts, with a 500 kW peak around 11am when both shifts overlap. Monthly kWh: about 200,000. Monthly demand charge at $19 per kW non-coincident plus $15 per kW for the peak-window demand (assume the peak does land in the peak window): $19,000 of demand charges. Monthly energy charge at 14 cents per kWh: $28,000. Total: about $47,000 per month, with the demand charges representing about 40 percent of the bill.
Mitigations: re-stagger the shift overlap so the second shift starts as the first shift finishes its lunch break (rather than during the peak production hour); install a 200 kWh battery sized for 100 kW discharge for 2 hours to shave 100 kW off the peak ($25,000 per month savings or $300,000 per year, against a battery cost of about $200,000 installed before the federal Investment Tax Credit, for a 6 to 12 month payback). The battery payback math for mid-size facilities in high-demand-charge territories is often the strongest dollar-return investment available; demand-charge mitigation is usually the leading category in commercial battery business cases.
EV charging and the demand-charge problem
DC fast chargers (50 to 350 kW per unit) are essentially demand-charge generators. A single 150 kW Tesla Supercharger stall operating during a peak window can add $1,500 to $4,500 per month to the host facility's demand charge. A six-stall DC fast charging site with all stalls active during peak could easily add $20,000 per month in demand charges, often exceeding the per-kWh revenue from the chargers themselves. This is why early DC fast charging deployments often failed financially; the demand charges ate the margin.
The industry has converged on a few solutions. First, behind-the-meter batteries that buffer the demand: the battery discharges to supply the charger during the charging session, then recharges slowly from the grid afterward, so the meter sees a steady 20 to 50 kW draw instead of intermittent 150 kW spikes. Second, demand-charge-aware dispatch software that throttles charger output if the facility is approaching a new peak. Third, ratemaking advocacy: several state PUCs have approved special demand-charge waivers for DC fast charging during the first 3 to 5 years of a site's operation, to allow the load to build before the demand-charge math becomes punishing.
Sources and further reading
- EIA commercial electricity sales and revenue
- PG&E commercial rate plans
- SCE business rates
- ConEd commercial tariffs
- NREL battery storage cost benchmarks
- Business electricity rates overview
- PJM capacity market context
- How we source these numbers