EV vs. Gas: What a Mile Really Costs, and the Four Ways the Comparison Goes Wrong

Comparing the running cost of an electric car against a gasoline one is a single division done twice. There is no modelling in it, no forecast, nothing that requires an opinion. Fuel is sold by the gallon and rated in miles per gallon; electricity is sold by the kilowatt-hour and rated in miles per kilowatt-hour. Divide the price by the rating and you have the cost of one mile.

gasoline     $ per gallon ÷ miles per gallon = $ per mile
electricity  $ per kWh    ÷ miles per kWh    = $ per mile

Which is why the interesting question is not the arithmetic. It is the four numbers you feed into it. Every one of them has a plausible-looking wrong version that is easier to find than the right one, and three of the four wrong versions push the answer the same direction — they make electricity look cheaper than your bill will show. This guide is about those four inputs.

A worked example, so there is something to move

Take a 27 MPG gasoline car at $3.19 a gallon, and an electric car rated at 3.5 miles per kilowatt-hour on electricity costing 17 cents. These are illustrative figures, not national averages — the whole point of what follows is that you should substitute your own.

gasoline     $3.19 ÷ 27  = $0.1181 per mile
electricity  $0.17 ÷ 3.5 = $0.0486 per mile

over 12,000 miles a year
  gasoline     $1,418
  electricity  $  583
  difference   $  835

An $835-a-year gap. Now watch how far each of the four common input errors moves it.

1. The dashboard readout is not the number your utility bills

This is the big one, and it is the one almost nobody catches, because the wrong number is displayed on the screen in front of the driver and the right one is on a government website.

EPA measures an electric car's efficiency at the wall outlet. Its ratings, in EPA's own words, “assume level 2, alternating current (AC) charging and account for losses from the charging cable… and the on-board vehicle charger. This moves the measurement from the vehicle to the outlet in the wall to better represent how much users would pay to refuel their car.”

Most cars' own efficiency displays report energy that left the battery. The energy that entered the house never appears. The difference is real electricity that your meter counted and your utility charged you for — heat in the cable, heat in the onboard charger — and it is always a loss, never a gain. So the dashboard figure is always optimistic, and by a margin you cannot see.

Enter 3.9 mi/kWh from the dash instead of 3.5 from the sticker and the electric side of the example falls from $583 a year to $523. The gap widens by $60 and none of it is real. You did not save that money; you just stopped counting some of the electricity.

The fix is either to use the EPA rating for your model year from fueleconomy.gov, or — better, if you already own the car — to compute your own from what actually happened: miles driven over a billing period, divided by the kilowatt-hours the utility billed for charging. That figure includes every loss by construction, because it is measured on the meter.

2. The supply rate is not what a kilowatt-hour costs you

Many utility bills separate the price of the electricity itself from the price of delivering it. Depending on the state and the utility, the delivery line can be a substantial fraction of the total. Quote only the supply rate and the electric side of the comparison is understated by exactly that fraction.

There is no lookup table for this, and any calculator that offers you a national average is guessing at the single number that most changes your answer. The reliable method takes thirty seconds with a recent bill:

your delivered rate = total amount billed ÷ kWh used

Total. Not the supply subtotal. That division rolls in delivery, riders, and every fixed charge, spread across the kilowatt-hours you actually used — which is the honest way to price a marginal kilowatt-hour for a household that already has electricity service.

If your utility offers a time-of-use rate and you charge overnight, use the off-peak price, and use it only for the share of charging you genuinely do at home overnight. Which brings up the fourth error, but first, a subtler one on the gasoline side.

3. “Combined” MPG is not the average of city and highway

EPA's combined rating blends its city and highway tests, weighted 55 percent city and 45 percent highway. People reasonably assume that means 0.55 × city + 0.45 × highway. It does not. Under 40 CFR § 600.210-12, manufacturers “shall harmonically average the unrounded city and highway fuel economy values” — that is:

combined = 1 ÷ (0.55/city + 0.45/highway)

The reason is that fuel economy is a rate, not a quantity. What you want to weight is the fuel consumed, not the miles per gallon. Averaging the MPG values directly always produces a number that is too high — a car spends more of its gallons in the condition where it burns them faster.

For a car rated 20 city and 30 highway, the arithmetic average gives 24.5 MPG and the harmonic average — the legal one, the one on the sticker — gives 23.5. At $3.19 a gallon over 12,000 miles that is $1,562 against $1,629: a $66-a-year error, from an averaging convention. Small, but it is free to get right, and several consumer calculator sites publish the wrong form.

You will rarely need to do this yourself, because the combined figure is printed on the window sticker and listed on fueleconomy.gov. Do it only when you are blending published city and highway numbers by hand — and if your driving really is 90 percent highway, skip the blend entirely and enter the highway rating, which models your situation more closely than any weighting will.

4. Pricing home charging, then driving to another state

The fourth error is a scope error rather than an arithmetic one. Public DC fast charging is priced by the charging network, not by your utility, and it is routinely a multiple of a residential rate. An electric car that costs five cents a mile in the driveway can cost considerably more on a highway corridor — and, unlike a gasoline car, it is also least efficient at exactly the sustained high speed where you are most likely to be paying those prices.

This does not make the home figure wrong. It makes it a figure about home charging. If nearly all your miles are commuting and errands from a driveway, the residential rate is the correct input and the answer stands. If you take four long trips a year, blend the two: work out roughly what share of your annual kilowatt-hours come from public charging and weight the rate accordingly. For a single trip, price it separately with the Trip Fuel Cost Calculator, using the network price you will actually pay rather than the one on your bill at home.

What the per-mile figure deliberately leaves out

Energy, and nothing else. Not depreciation, insurance, maintenance, tires, registration or the purchase price. That is a limit rather than an oversight: those costs vary so widely by vehicle, driver and state that folding an assumed figure for any of them into a per-mile number would bury the one thing this calculation can tell you exactly.

The other two thirds of the question have their own pages. Whether cheaper miles ever repay a higher sticker price is break-even mileage — and if you want the honest version of that question, read Does a Hybrid or EV Pay for Itself? first, because the standard version of the calculation is systematically too optimistic. What a car costs in total, per mile, across every category of spending including depreciation, is the True Cost per Mile Calculator.

One last caution that applies to both sides equally. EPA is explicit that its label ratings “may not accurately predict the average MPG you will get” — they are a standardized comparison between vehicles, not a forecast for your commute. If you have owned the gasoline car for a while, measure it instead: two odometer readings and one fill-up gives you a real number, which is what the True MPG Calculator does. A comparison run on two measured figures beats one run on two laboratory figures, every time.