What Depot Charging Actually Costs in California (And Why the Tariff Decides Everything)
Table of Contents
- The Question We Were Asked
- The Depot and the Standard Tariff
- What the Standard Tariff Actually Costs
- Why the Gap Is This Big
- What the Battery Sweep Taught Us
- The Tariff That Flips the Answer
- This Isn't Just Us Saying It
- What About Incentives?
- The Honest Conclusion
- Run Your Own Depot's Numbers
- FAQ
We came across this LinkedIn post: a chart from a researcher in electromobility, plotting the gap between diesel and electricity prices across dozens of European and North American markets, split by season and time of day. The point was straightforward and useful: that gap is roughly the budget a fleet has to spend on charging infrastructure and still come out ahead of diesel. Wide gap, more room to invest. Narrow gap, a much harder case to make.
Someone asked the natural follow-up in the comments: that chart shows the savings side, but what about the cost side? What does charging infrastructure and demand charges actually cost, market by market? The reply, in effect, was that it's the other half of the equation someone else would need to run.
So we ran it. Not a market survey, one real depot, priced against two real tariffs, the way we'd run it for any customer. The result surprised us: the electricity itself was never really the question. The tariff was.
The Question We Were Asked
The diesel-versus-electricity gap tells you the ceiling: how much a fleet can spend on charging infrastructure before the economics stop making sense. It does not tell you what a depot actually pays once you factor in the thing that dominates most commercial electric bills and that a simple energy-price comparison leaves out entirely: the demand charge, a monthly fee based on the single highest 15-minute power draw a depot records, independent of how much energy that draw actually delivered.
We picked California for this first pass, for two reasons. It's one of the highest-diesel, most economically favorable markets on that chart, so if the numbers don't work here, that's worth knowing. And PG&E publishes some of the most detailed, most current commercial tariff data of any utility in the country, which meant we could run this without guessing at a single input.
The Depot and the Standard Tariff
We used a real depot fixture from our digital twin: 15 trucks, 15 × 180 kW DC chargers on a 2.6 MW service, each truck needing 260 kWh per shift on a tight 18:00 to 20:15 turnaround. That's a genuinely realistic distribution depot, not a cherry-picked edge case, and its short, coincident evening turnaround turns out to matter a great deal, more on that below.
Our first pass used PG&E Schedule B-19, the standard general-demand commercial tariff, secondary voltage, effective March 2026. Its demand charges stack seasonally: $94.05 per kW in summer (four months) and $39.68 per kW in winter (eight months), on top of peak-window energy at roughly $0.17-$0.19 per kWh. For the diesel side, we used the current EIA weekly California on-highway diesel price, $6.471 per gallon, converted to a per-kWh equivalent using the same 4 kWh-per-liter method used in that LinkedIn chart. That works out to roughly $0.43 per kWh: the ceiling this depot's electricity cost needs to beat.
What the Standard Tariff Actually Costs
We modeled the same depot four ways on B-19, running the actual simulation engine each time, not an estimate:
| Scenario | Peak demand | Levelized cost | vs. diesel ceiling ($0.43/kWh) |
|---|---|---|---|
| Unmanaged, coincident charging | 2,700 kW | $1.50/kWh | 3.5x over |
| Managed charging, no new hardware | 2,400 kW | $1.35/kWh | 3.2x over |
| Managed + 500 kWh battery | 2,141 kW | $1.23/kWh | 2.9x over |
| Managed + 750 kWh battery | 2,000 kW | $1.16/kWh | 2.7x over |
The unmanaged case isn't just expensive, it's electrically broken: naive simultaneous charging locates 30 separate breaches against the depot's own circuits before you even get to the cost question. Managed charging alone, with no new hardware, brings the peak down 300 kW and the levelized cost down about 10%. Adding a right-sized battery gets another 14% on top of that. All real, all worth having. None of it, on this tariff, gets the depot under the diesel-equivalent ceiling.
Why the Gap Is This Big
This isn't a story about electricity being expensive. PG&E's peak-window energy rate on B-19 is under $0.19 per kWh, cheaper than diesel on a pure energy basis. The gap comes almost entirely from the demand charge, and specifically from how this depot's 1h45m nightly charging window interacts with it.
A demand charge bills once a month on the single highest 15-minute draw, no matter how much energy that peak actually delivered. A depot that spreads its charging across a wide overnight window pays that charge against a peak that represents a large share of its total energy use. A depot that squeezes the same energy into a narrow, coincident evening window, like this one's 18:15 to 20:00 turnaround, pays the same monthly demand charge against a peak that represents a much smaller share of total energy delivered. Same tariff, same trucks, very different cost per kWh, purely because of how tight the turnaround is.
That's the finding worth taking away from this section, independent of the specific dollar figures: short-dwell, high-coincidence depots are structurally the hardest case for demand-charge tariffs. It only shows up once you model the actual depot, which is exactly why we kept going instead of stopping here.
What the Battery Sweep Taught Us
We swept the battery from 250 kWh up to 2,000 kWh, at a fixed 400 kW power rating, to see where the returns stopped. Below 500 kWh, the battery was too small to shift the peak at all. From 500 kWh up to 750 kWh, it kept delivering real, growing peak reduction. Past 750 kWh, adding more capacity did nothing, the peak reduction flatlined completely, all the way out to 2,000 kWh.
The reason is straightforward once you see it: past 750 kWh, the battery wasn't running out of energy anymore, it was running into its own 400 kW power rating. A battery can only push out power as fast as its rating allows, no matter how much energy is sitting in it. Buying a bigger battery at that point doesn't buy you a lower peak, it buys you unused capacity. The lever that would have actually moved the peak further was a higher-power battery, not a bigger one, and that's a distinction a back-of-envelope estimate would have missed entirely.
The Tariff That Flips the Answer
PG&E doesn't only offer B-19 to a depot like this. It also offers Schedule BEV-2, a rate built specifically for commercial EV charging, and its structure is fundamentally different: it replaces the demand charge and the customer charge entirely with a flat monthly subscription, sold in 50 kW blocks at $95.56 per block. A depot pays for the capacity it reserves, once, every month, not for the single worst 15 minutes it happened to record. The tradeoff is a higher per-kWh energy rate, about $0.37 per kWh at peak, to fund that swap.
We ran the same three peaks, the same fleet, the same annual energy, through BEV-2 instead:
| Scenario (same depot, same peaks) | Subscription | Levelized cost | vs. diesel ceiling ($0.43/kWh) |
|---|---|---|---|
| Unmanaged, coincident charging | 54 × 50 kW blocks | $0.41/kWh | under |
| Managed charging, no new hardware | 48 × 50 kW blocks | $0.41/kWh | under |
| Managed + 750 kWh battery | 40 × 50 kW blocks | $0.40/kWh | under |
Every single scenario clears diesel parity, including the unmanaged one that was electrically broken and 3.5x over diesel on B-19. The subscription model is simply so much cheaper per kW, roughly $1.91 per kW-month versus B-19's blended $58 per kW-month, that even the worst-case peak barely moves the annual bill. Managed charging and a battery still help, they take the number from $0.41 to $0.40, but on this tariff they're a refinement, not a rescue.
Same depot. Same trucks. Same energy. The only thing that changed between a 3.5x loss and a genuine win was which tariff schedule the account was on.
This Isn't Just Us Saying It
Demand charges as a barrier to EV charging economics isn't a finding unique to this one depot, or to us. A 2024 rate design study by Dunsky Energy + Climate Advisors, prepared for New Brunswick Power, examined this exact problem for fast-charging operators and reached the same structural conclusion: low-utilization, high-peak charging sites get hit hardest by demand charges precisely because so little of their energy use lines up with their single worst 15-minute draw. One fast-charging operator told Dunsky that at typical utilization, the effective electricity cost on a standard demand-charge tariff can run around a dollar per kWh, in the same range we found for this depot on B-19.
The study surveyed rate designs across North America built to fix this, and PG&E's Business EV rate, the same BEV-2 schedule we ran above, is one of its case studies, cited with the identical $12.41 per 10 kW and $95.56 per 50 kW block figures we pulled straight from the tariff sheet. It's not the only one: Eversource in Connecticut and Massachusetts, ConEdison in New York, Hydro-Québec, and BC Hydro have each built or proposed their own version of a demand-charge-mitigation rate, several after regulators explicitly directed them to address the barrier. The load-factor-based and subscription-style structures behind these rates are becoming the recognized fix across the industry, not a one-off PG&E quirk, and not something we're the first to point out.
Read the Dunsky / NB Power rate design report.
What About Incentives?
It's worth being precise about what does and doesn't belong in a number like this. Two categories of programs exist, and they don't do the same job.
Capital incentives lower the upfront cost of building the depot. PG&E's EV Fleet program will cover a substantial share of make-ready infrastructure and offers charger equipment rebates (up to $42,000 per port for 150 kW+ chargers, capped at 50% of equipment cost, for up to 25 vehicles per site) for exactly the kind of depot we modeled here. Those are real, meaningful dollars, and they change a project's payback period. They do not change the $/kWh figures in the tables above, which are operating cost, not capital cost. We didn't include them here because doing so honestly would mean modeling a full capital stack, a separate analysis from the operating-cost question this post answers.
Public charging incentives, like CALeVIP, don't apply to this depot at all. CALeVIP funds publicly accessible charging; a private fleet depot serving its own trucks doesn't qualify, regardless of how the funding is described elsewhere.
The tariff switch above isn't technically an incentive, it's a rate design choice PG&E already offers, and it moved this depot's economics by roughly 3x, more than every capital incentive we found put together would move a payback period. That's the actual headline of this analysis: the rate schedule mattered more than any rebate program.
The Honest Conclusion
For this depot, on the standard commercial tariff, the math doesn't clear diesel parity, not with managed charging alone, and not with a reasonably sized battery on top. On the tariff PG&E built specifically for this use case, it clears diesel parity comfortably, in every scenario we ran, including the electrically broken unmanaged one. Both of those are true statements about the same depot, and neither one is the whole picture without the other.
The practical takeaway isn't "California works" or "California doesn't work." It's that the tariff a depot ends up on, often decided early and rarely revisited, can be the single biggest lever in the entire economics, bigger than the battery, bigger than the charging software, bigger than the incentive programs. Getting that decision right, before the capital moves, is exactly the kind of thing worth modeling rather than assuming.
Run Your Own Depot's Numbers
This is one depot on two tariffs. Every depot's charging window, fleet size, and local rate options change the answer, sometimes by more than a battery or a rebate ever could, which is exactly why a market-level comparison can only tell you the ceiling, never the actual number. The digital twin that produced every figure in this post runs the same analysis for any depot, any tariff, any market.
Try the BEV Ready depot digital twin: model your own depot's real levelized cost.
FAQ
Is EV charging cheaper than diesel for fleet depots in California? It depends almost entirely on which tariff the depot is on, more than on the electricity itself. Under PG&E's standard B-19 demand-charge tariff, a depot with a short, coincident charging window can land well above diesel-equivalent cost. Under PG&E's Business EV rate, which replaces the demand charge with a flat subscription, the same depot, same trucks, same energy, comes in under diesel-equivalent cost.
What is a demand charge and why does it matter this much? A demand charge bills on the single highest 15-minute power draw in a billing period, in dollars per kW, separate from the per-kWh energy charge. It recurs every month regardless of how much energy that peak actually delivered. A depot that charges a lot of energy in a short, concentrated window pays a large demand charge relative to the energy it actually used, which is exactly the shape of a tight-turnaround depot.
Does a battery fix the demand charge problem? Partially, and only up to its power rating, not its capacity. In our modeled case under the standard demand-charge tariff, a battery sized between 250 and 750 kWh kept delivering peak reduction as its capacity grew, then flattened out once the battery's 400 kW power rating became the binding constraint. But switching the tariff moved the number far more than the battery did.
Do EV charging incentives change this analysis? Capital incentives, like PG&E's EV Fleet program covering make-ready infrastructure and charger equipment, lower the upfront cost of building the depot, but they don't change the per-kWh operating cost this analysis covers. Publicly accessible charging incentive programs like CALeVIP don't apply at all to a private fleet depot. The single biggest lever in this analysis wasn't an incentive; it was picking the tariff actually designed for EV fleet charging instead of the standard commercial one.
How was this analysis done? Using BEV Ready's depot digital twin, a real 15-truck fixture, and two current PG&E tariffs (Schedule B-19, Cal. P.U.C. Sheet 61071-E, and Schedule BEV-2-S, Cal. P.U.C. Sheet 61092-E, both effective March 2026), we modeled the depot's naive unmanaged peak, its managed peak, and a battery sweep, then priced each against both tariffs to get a levelized cost per kWh. The diesel comparison used the current EIA California on-highway diesel price, converted with the same kWh-per-liter method used in the LinkedIn post this article responds to.