Tax/Expenses Allowances on Car Mileage

sabice

Now, I don't want to set the world on fire, but ...

In my work I do a lot of mileage with work travelling around different sites. In the old days I would take a note of my mileage from the car's mileometer, in other words, the actual mileage done.

In recent years, with the advent of web based journey planners, I've switched to stating my 'from' and 'to' locations and noting down the computer measured mileage.

Now and again, however, when for whatever reason I've used both, I've noticed that the actual mileage always seems to come out slightly more than the computer generated mileage, say around 5-10% more.

Initially, I put this down to 'one-off' factors such as making slight diversions due to traffic or taking a wrong turn.

Then, this morning, it struck me that the maps used for journey planning are essentially two-dimensional, length and breadth only; no variations in heights or gradients are taken account of, particularly when travelling across hilly areas. Now, going back to school days of pythagoras, we know that these hypotenuses are 'longer' than the straight or level line.

Reviewing all of the above then, it seems that I've been underclaiming my tax/expenses allowances for mileage by the order of 5-10%. I'll be taking a few test measurements over the coming weeks to double-check this.

Does anyone know if this has been highlighted before as a tax saving matter, and how many years back this can be retrospectively claimed?

rs65

Interesting. The tax relief claim is on the difference per mile.

How far back will your employer let you retrospectively adjust your mileage claims?

sabice

I've mostly been freelance/self-employed over the last few years, so this may vary with each client. I'll have to take a view on this as I don't necessarily want to upset any apple carts here.

In respect of HMRC, however, where say a client did not pay expenses direct but I rolled this into the overall fee invoice, then I need to look at the 40-45 pence allowance figures I've been putting on the self-assessment.

So for example, if I've done 5000 miles at 45p tax allowance per mile, resulting in a £2,250 allowance, and the 'actual' mileage was say 5500 (10% more), then this allowance should have been £2,475 - a £225 difference in allowance. Accumulated over a number of years, the corresponding rebate amounts could result in quite a notable sum worth raising as a query/correction with HMRC (£45/year at basic rate).

jennifernil

I think you are clutching at straws here, I find it difficult to think that the difference could be 5-10% on the exact same route.

sabice

There's only one way to find out!

Yesterday's journeys amounted to around 5%. Doing the sums now for today's journeys. Tomorrow it's Hull, so likely to be less given the 'flatness' of the terrain. If you live in Lincolnshire then you're probably right.

sabice

Today's journeys:

1st leg: Online Map=10.2 miles, Actual=10.9 miles, Difference=6.9%
2nd leg: Online Map=35.9 miles, Actual=37.7 miles, Difference=5.0%

This is also taking worst case, where I've modified the online map to include any diversions I took for whatever reason (traffic, wrong turn, etc). In other words, it's the exact same route as far as I can tell - if I took the face value output of the online maps, the mileage would be lower and the percentage difference greater.

Tomorrow's journey is much longer, so will be interested to see the outputs for this one.

Whilst I'm looking at this from the tax allowance perspective, these discrepancies will have a much greater impact for anyone who claims mileage as a direct expense from an employer.

zygurat789

sabice wrote: »

Today's journeys:

1st leg: Online Map=10.2 miles, Actual=10.9 miles, Difference=6.9%
2nd leg: Online Map=35.9 miles, Actual=37.7 miles, Difference=5.0%

This is also taking worst case, where I've modified the online map to include any diversions I took for whatever reason (traffic, wrong turn, etc). In other words, it's the exact same route as far as I can tell - if I took the face value output of the online maps, the mileage would be lower and the percentage difference greater.

Tomorrow's journey is much longer, so will be interested to see the outputs for this one.

Whilst I'm looking at this from the tax allowance perspective, these discrepancies will have a much greater impact for anyone who claims mileage as a direct expense from an employer.

You are assuming that your car is giving you an accurate mileage, we all know that the speedometer is overoptimistic, by design, so why shouldn't the on line map be the more accurate?

martinsurrey

The uk is a VERY flat place in the grand scheme of things

http://www.bikehike.co.uk/mapview.php

(select elevation data)
I tried a 100mile trip in Scotland from Aberdeen to Inverness in the Scottish highlands.

A total of 8000ft vertical assent/descent combined, which is a huge amount for the UK.

Now that is 1.5miles

To maximise the total distance travelled that would be 1.5m vertical + the 100m horizontal (as you know, the sum of the length of the 2 sides of a right angle triangle is longer than the hypotenuse, and is infact the longest it can be)

So that 100 mile map drive is at worst a 101.5 travelled or 1.5% out (and that is if Scotland is flat with 2mile cliffs you fall off!.

In a more realistic case, if we assume the gradient was constant (so you drive the hypotenuse of a right angled triangle with sides 100/1.5) you get a distance of 100.012 miles driven, or 0.012% out.

In short, the gradient has NO meaningful effect on distance in the UK, if your commute was up Everest, then it would be worth it.

Your speedo is out, as everyones is, always over estimating speed, so overestimating distance.

The difference depends on the speed you travel, so different journeys will be out by different amounts (%).

sabice

I think there are several variables at work here, but all told the lesson I have learnt is to revert back to noting my mileage down from the mileometer/odometer, as a straightforward output from an online map is misleading because it is based a set of assumptions which do not always relate to actual travelling. Whether it be for reasons of avoiding traffic, wrong turns, minor diversions, gradient or the like, these otherwise 'unrecognised' factors would seem to generate a 5-10% discrepancy.

Given that mileage rates have 'evolved' naturally from times before electronic mapping, and as you say cars' odometers/speedometers are by design a certain percentage out, then this confirms that to use your car's mileage as the claimed mileage is the correct way to go.

Regarding the two posts above, and I appreciate that the gradient is not the only variable at work here, it strikes me that you're both still basing your calculations on just two dimensions, and not taking into account a third dimension, the z-axis to your already accounted for x and y axes.

I have travelled several hundred miles around Scotland and yes I remember the car going up and down, and forwards and backwards, but I also remember it going side to side.

Goldenyears

Odometers don't have to be "out", but yours probably is. The regulations on limits of error only appy to speedometers. Odometers in the last 4 cars I have owned have only differed from web based distances by no more than 1%.

[Deleted User]

Looking for some help or opinions on a mileage matter.

I work in an office and make regular trips to client premises -

Not a problem, mileage claimed from office to client and back

When I travel to the clients from home however (not going via the office), I deduct the mileage from home to the office (being my usual comute to my place of work). Eg,

home to client = 10
Less home to office = 5
Total claim to employer = 5 miles @ 35p pm


Now the question is this:

I already put in a claim to HMRC for the tax relief on the 10p (45p-35p) on the claim to my employer but can I also claim the tax relief on the E.G. 5 miles I deduct (45P *20%)?

The thought being that as I am carrying out work at the clients it is a temporary place of work and that 100% of the mileage to that place is allowable.