Standard Mileage Rate vs Actual Expenses
One method needs a mileage log and a calculator. The other needs every receipt for the car, all year. The choice is mostly made for you in the first year you use the vehicle.
For most self-employed people driving an ordinary car, the standard mileage rate produces a bigger deduction with far less work — you multiply business miles by a published cents-per-mile figure and you are done. Actual expenses wins when the vehicle is expensive to own relative to how far you drive it: a new truck, a heavy depreciation year, high insurance, or a low annual mileage spread over big fixed costs. The catch is that you do not get to make this choice fresh every April. What you elect in the first year the vehicle is used for business decides which doors stay open afterwards.
The two methods, briefly
Standard mileage rate. Business miles × the IRS rate for that year. The rate is designed to cover fuel, oil, maintenance, tyres, insurance, registration and depreciation in one number. You do not deduct any of those separately.
Actual expenses. Add up everything the vehicle actually cost for the year, then deduct the business-use share of it. Business-use share is business miles divided by total miles — so you still need a mileage log either way.
Parking and tolls incurred for business are deductible on top of either method. So, for a self-employed filer, is the business-use share of car loan interest and any state personal property tax on the vehicle.
The first-year election is the whole decision
This is the part that surprises people, and it is worth reading twice.
- Owned vehicle. If you want the option to use the standard rate at all, you must choose it in the first year the car is available for business use. Take actual expenses in year one and the standard rate is closed for that vehicle, permanently. Choose the standard rate in year one and you may switch to actual expenses later — though depreciation from then on has to be straight-line.
- Leased vehicle. Whichever method you pick in the first year, you keep for the entire lease term, renewals included.
So the safe default for a vehicle you own is to elect the standard rate in year one. It costs nothing, it keeps both methods available, and you can run the comparison each year with a log you were keeping anyway.
What each method asks you to keep
| Standard mileage rate | Actual expenses | |
|---|---|---|
| Mileage log | Required | Required (for business-use %) |
| Fuel receipts | Not needed | Every one |
| Service, tyres, repairs | Not needed | Every one |
| Insurance, registration | Not needed | Annual totals |
| Depreciation schedule | Not needed | Required |
| Time cost per year | Minutes | Hours |
| Parking and tolls | Deductible on top | Deductible on top |
A worked comparison
A self-employed contractor drove 18,500 total miles in 2026, of which 12,400 were business — a business-use share of 67.0%. The mileage split across the year was 5,800 miles January–June and 6,600 miles July–December, because 2026 had two different rates: 72.5 cents for miles driven 1 January through 30 June and 76 cents from 1 July onwards. (Rates as published on our cleaning business mileage guide, checked against the IRS announcements in August 2026 — always confirm the figure for the year you are filing, because mid-year changes are unusual and easy to miss.)
Standard mileage rate
| Period | Business miles | Rate | Deduction |
|---|---|---|---|
| 1 Jan – 30 Jun | 5,800 | $0.725 | $4,205.00 |
| 1 Jul – 31 Dec | 6,600 | $0.760 | $5,016.00 |
| Total | 12,400 | $9,221.00 |
Actual expenses
| Cost | Amount |
|---|---|
| Fuel | $3,100 |
| Insurance | $1,680 |
| Maintenance and repairs | $940 |
| Tyres | $620 |
| Registration | $210 |
| Depreciation | $3,800 |
| Total vehicle cost | $10,350 |
| × 67.0% business use | $6,934.50 |
The standard rate wins by $2,286.50, and it wins without a shoebox. That is the ordinary result for an ordinary car driven a lot for work.
Flip one input and the answer flips with it. Drop business mileage to 4,000 and the standard-rate deduction falls to roughly $2,980, while a vehicle with $10,350 of annual cost and, say, a 40% business-use share still returns $4,140. Low miles and high fixed costs is the shape where actual expenses pays.
The log you need either way
Both methods stand on the same evidence, and it is the evidence — not the arithmetic — that fails audits. Each business trip needs four things recorded: the date, the business purpose, the destination, and the miles. Written at the time, or close enough to it that you are not reconstructing from memory in April.
You also need total miles for the year, not just business miles. Without that you cannot compute a business-use percentage, which means actual expenses is unavailable to you and the standard-rate figure has nothing supporting it. One odometer reading on 1 January and one on 31 December is enough.
Common questions
Can I use the standard rate for one car and actual expenses for another? Yes. The election is per vehicle, not per business. Many people run the standard rate on a daily driver and actual expenses on a heavy vehicle bought outright.
Does the standard rate include depreciation? Yes — a depreciation component is baked into the rate, and it reduces the vehicle’s basis over time. That matters when you sell: a car deducted at the standard rate for six years has a lower basis than its purchase price suggests.
What if I forgot to keep a log for part of the year? A partial log covering a representative period can support an estimate for the rest of the year if your driving pattern is genuinely consistent — but “I drive about the same every week” is a much weaker position than a contemporaneous record, and it is the first thing an examiner pulls on. Start logging today rather than reconstructing January.
Get the next template
One short email when a new spreadsheet or template ships — plus the occasional tip for getting more out of the ones you already have. Nothing else.
We send the list and nothing more. Your address is never sold or shared, and every email has a one-click unsubscribe.