16 November 2020
The case of Narinder Sambhi (TC7717) was reported in Taxation, 15 October 2020, page 22 and is worthy of more detailed examination as to the light it sheds on the deductibility of travel expenses. But before delving further into that case, it is worth exploring the legislation in ITEPA 2003, s 339: ‘Meaning of “workplace” and “permanent workplace”’. And in doing so, it may help to consider the example of Harry, a leading expert in his field, who is employed
by a firm of tax specialists based in Belfast. The firm has won a new engagement helping a major multinational business to update all its policies and, in particular, advise on tax-efficient
mileage claims. The client is so big that the project will last three years and Harry has agreed to spend the first 12 months working at the client’s Scottish office in Edinburgh, the next 12 months in its Welsh office in Cardiff and the last 12 months in the English office near Heathrow.
Harry flies over from Northern Ireland every Monday and flies back on Friday, lodging in the UK during the week. The flights are with the same airline, they all take about an hour
and cost roughly the same. The lodging costs are also similar.
Let’s first consider ITEPA 2003, s 339(5):
‘A place is not regarded as a temporary workplace if the employee’s attendance is:
(a) in the course of a period of continuous work at that place:
(i) lasting more than 24 months; or
(ii) comprising all or almost all of the period for which
the employee is likely to hold the employment; or
(b) at a time when it is reasonable to assume that it will be in
the course of such a period.’
Note that subsection (ss) 5 makes no reference to the duties performed. Nor does it refer to the contract or assignment length, but rather to a period of continuous work physically
undertaken at a particular place. Under ss 5(a)(i) each physical place, geographical co-ordinate, postcode, factory site, building, business park or construction site where Harry works is deemed a permanent workplace if he actually works at that (singular) place continuously for more than 24 months. Subsection 5(a)(ii) is not in point because this assignment does not comprise his entire employment. Subsection 5(b) can deem a singular place to be permanent
rather than temporary if it is objectively reasonable to assume that the 24-month threshold will be breached. Note there is no reference to either the intent of the employer or employee; it is
an objective or factual test. So if, at the outset, Harry has a piece of paper saying he will be working in Edinburgh for three years then that location is permanent at that point. However if, after the first month, he was told instead that he would be moving to another office at the end of the 12th month, then the last 11 months would be temporary (because he is neither in ss (5)(a) nor ss (5)(b)), but the first month would still be caught and travel to that office during that time would remain private (commuting) travel because of ss (5)(b).
Continuous work at that place
ITEPA 2003, s 339(6) states: ‘For the purposes of subsection (5), a period is a period of continuous work at a place if over the period the duties of the employment are performed to a
significant extent at the place.’ The word ‘significant’ is not defined, and nor does a 40%
figure appear in statute, but it is commonly accepted through custom and practice (and HMRC’s guidance in its Employment Income Manual at EIM32080) that anything less than 40% is
not significant. For both Harry and Mr Sambhi their duties were full time. Note that ss (6) does not put the taxpayer into ss (5) but it can take them out of that subsection by suggesting
that anything less than two days a week on average is ‘not continuous’ and so cannot be in ss (5).
Before looking at s 339(7), which was the crux and central issue of the Sambhi case, let us first look at ss (8):
‘An employee is treated as having a permanent workplace consisting of an area if:
(a) the duties of the employment are defined by reference to an area (whether or not they also require attendance at places outside it);
(b) in the performance of those duties the employee attends different places within the area;
(c) none of the places the employee attends in the performance of those duties is a permanent workplace;
(d) the area would be a permanent workplace if subsections (2), (3), (5), (6) and (7) referred to the area where they refer to a place.’
Because Harry is working on the British mainland for three years and his duties are ‘defined’ by reference to Edinburgh, Cardiff and London then under ss (8)(a) HMRC might suggest
that Britain (the UK excluding Northern Ireland) is his permanent workplace for the next three years. In that case, the first few minutes of the flight from Belfast to English, Scottish or Welsh airspace would be a commute (from home to the edge of the work area) but all costs inside Britain would qualify as business travel. How the extent of any disallowance should be
calculated would be interesting. However, if Harry demonstrated that he worked in Belfast
at the weekends writing up his notes and preparing for next week then arguably the whole of the UK would be his work area and so all travel and subsistence costs for three years would qualify as business. A different – and perhaps more likely – interpretation is that Harry has three separate temporary workplaces in three different countries, none of which breaches 24 months.
The Sambhi case
In the case of Mr Sambhi, his original contract and subsequent revisions point to his duties potentially covering the whole UK although not necessarily formally defining the UK as his ‘area’. If we agree that his area is the UK, then Mr Sambhi lives inside his area – the UK – and therefore this trumps s 339(5) to (7) and all of his work travel is business and therefore allowable.
If it was the case that the UK was not sufficiently clearly defined as his area in his contract, then we have to ask the question ‘does he actually have another defined area such as Greater London’? For that argument to succeed, HMRC must prove that the conditions in s 339(8)(a) to (d) are all satisfied. Even if HMRC could prove this, Mr Sambhi is nonetheless still allowed to claim some costs for travel once inside that area.
Mr Sambhi most likely failed s 339(8)(a) because there is no document to evidence that all the duties of his employment were ‘defined by reference to an area’ – in other words, Greater
London. If we are not in s 339(8) then we cannot replace the word ‘place’ with the word ‘area’ as regards ss (5). So we can discount the area rule and just look at the separate places where he worked.
Further, s 339(7) does not and cannot impute an area to exist where none does under (8). Subsection (7) is there only to qualify ss (5) and ss (6). It is anti-avoidance legislation to prevent the 24-month clock being artificially reset by effecting a minor change to the workplace. It does, however, protect the employee if the costs or journey change substantially even though the two places are in close proximity.
So what does s 339(7) – with my italics – really say?
‘An actual or contemplated modification of the place at which duties are performed is to be disregarded for the purposes of subsections (5) and (6) if it does not, or would not, have any substantial effect on the employee’s journey, or expenses of travelling, to and from the place where they are performed.’
Throughout the Sambhi case and enquiry, HMRC kept referring to a ‘change’ of temporary workplace thereby misleading and misdirecting themselves, the taxpayer and the court. The legislation talks not of a change but ‘modification’ of a place or the place (singular) at which duties are performed. The distinction is subtle, but absolutely crucial.
Modification is not change
Sadly, in the Sambhi case, no time seems to have been spent exploring the word modification. Indeed the judge herself constantly refers to a ‘change’ of place in her decision rather than looking at modification. All the tribunal’s effort and time went into looking at the meaning of ‘substantial’ which is not relevant in the case of an actual change. The term ‘substantial’ is relevant only if we are first in a situation where there has been a modification. I asked a friend ‘What does modification of the place mean?’ They replied, sensibly, that it is like adding a garage, conservatory or attic room to a house. The man on the Clapham omnibus would say that a modification of a place (or the place) is a minor tweak to it.
The Cambridge and Oxford dictionaries define ‘modification’ as a change to something, not a change of something.
So, for example, if the owner of a Morris Minor upgrades the wheels to alloy and replaces the windscreen with tinted glass then they have modified the car, but it is still a Morris Minor. However, if they were to sell the Morris and buy a Ferrari they have not modified their car, but would instead have changed it.
The man on the Clapham omnibus could translate ss (7) as follows: ‘Any small geographical alteration to a temporary workplace will result in a completely different place if there is a substantial impact or effect on the journey such as the time or cost of getting there. Otherwise we will ignore the alteration and treat it as the same place as before.’
For example, an employee does temporary work at the west side of a business park for 20 months, but then does new and perhaps different temporary work for a new client of their
employer on the east side of the same park for six months. Ordinarily, one may say that they are working at the same place (the same business park) for 26 months. But let’s say that this park is separated by a dual carriageway and the employee has to drive three times further to get to the east side (perhaps they must go to next motorway junction, turn around and come back to leave at the correct exit). In that situation, the employee would arguably have two separate workplaces, each lasting less than 24 months.
Difference in location or costs?
A very similar example (again with my italics) is found in HMRC’s Booklet 490: Tax and National Insurance contributions for employee travel (tinyurl.com/yy8cgmnm) in example 4.8:
‘Justin is employed on a major bridge construction project. To begin with he works on the north shore but he is then transferred to work on the south shore. Crossing the river is inconvenient (which is why a new bridge is needed), and it takes Justin longer to travel to the south shore and costs much more than it did to travel to the north shore. The north and south shores could be described as a single construction site and, as the crow flies, they are not far apart. However, Justin’s move from the north to the south shore has had a significant effect on his journey to work (and, in particular, the cost of that journey) so his workplace has changed for tax purposes.’
HMRC therefore agrees that a difference in costs means that Justin has two separate workplaces. But we cannot and should not infer that if the cost of travel is the same then the
two workplaces must be at the same location. That is not what s 339(7) says or implies. Nor was it the intent of that legislation.
So, for example, returning to the example of Harry, let us assume he does have three temporary workplaces of 12 months each in duration. We do not believe that s 339(7) can apply to deem Edinburgh, Cardiff and London as being the same place just because the costs and time involved in travelling to each is the same or similar.
Separate and distinct
We believe that a similar position applied for Mr Sambhi. The tax years in question covered only 2015-16 and 2016-17, but the court looked at his entire work pattern from September 2013 to September 2019, in other words for the seven tax years during which he worked at 11 different sites throughout Greater London, of about six months each in duration.
Each project was completely distinct from and unconnected with the previous project. The first six years of his employment were spent working on projects in the Midlands and the next six working on projects in London, though he could equally be moved to a project in Scotland next week under his contract.
From September 2013 to 2019, Mr Sambhi rented three different properties in London. There was no finding of fact that Mr Sambhi had a second home in London and this was not considered in depth by the court. It was also irrelevant for the purposes of the judgment that he chose somewhere in north-east London even though for some of the time he was working on projects in south-west London. He had a limited budget and presumably the former was significantly more affordable and therefore more within his budget compared with the latter area.
The temporary sites where he worked in 2015-16 were significantly far apart; for example, Putney to Guildhall and Guildhall to Croydon. Travelling from any one to the other takes one hour on a good day by car and takes three to three and a half hours on foot. They are all clearly separate and distinct places and one is not a modification of the others. The logic then is that if we have no modification of a single and distinct place then we cannot be in ss (7) and his appeal
against a refusal to allow his travel costs should have stood.
It may also help in understanding the intent of the use of the word ‘modification’ to look at the predecessor legislation in FA1997, s 62 (inserting s 198 and s 198A into TA 1988), which was later repealed by ITEPA 2003 (tinyurl.com/y3cq9v2l).
In those days, one had to deduct the cost of ordinary commuting to a place (or area) in arriving at an allowable claim. Otherwise the current and previous legislation are similar. The old TA 1988, s 198 simply required the cost be different to evidence that the 24-month clock had stopped and been reset.
The meaning of modification
A clue as to the meaning of modification is given by TA 1988, s 198A, namely the phrase ‘travelling between two places in a case where, because of the proximity of one place to another, the journey in question is, for practical purposes, the same as a journey…’. In other words, it looks at the proximity of two places as a measure of whether there has been a modification.
The current legislation has simply reversed the order of the previous sentence by trying to say that even if two places are in close proximity, they are to be treated as different places if the
cost or journey there is significantly different.
This notion of ‘proximity’ still flows through HMRC’s manuals and Booklet 490 which, at examples 4.6 and 4.7 and with my italics, gives examples (tinyurl.com/yyzcrujt) of when we have to look at differences in journey or cost in the context of a modification:
‘But the position is different if an employee moves to a new office in the next door building. In these circumstances there is no significant effect on the employee’s journey to work and under the tax rules there is no change of workplace.
‘Josh is employed on a construction site. His employer buys an adjacent plot to extend the site.
‘Yinka works for an employer who has several offices close to each other in London.’
Every example given in the HMRC manuals and in Booklet 490 implies that modifications are minor tweaks; say a building next door so that there is no significant change in the journey or expense of getting there.
It therefore appears that the arguments made by HMRC in the Sambhi case contradict its own manuals and booklets simply because someone at the department misread the modification and assumed it meant change and used this against Mr Sambhi.
Perhaps we can summarise by saying that a different cost means a different place, but the same cost does not necessarily mean same place.
A few final words
Unfortunately, HMRC’s view in Sambhi now means that the operation of s 339(7) is highly subjective. In this case, the judge looked at whether each new workplace was a change (rather
than a modification) of a workplace.
To do this they looked at the whole journey from Birmingham to the locations in London. This in itself is odd because the journey from Birmingham was only undertaken on Mondays and Fridays rather than every weekday. If the employee had instead lived in, say, High Wycombe, then the fact that the Putney is not near to Croydon would have had a bigger proportionate effect on the overall journey so as not to fall foul of s 339(7). Would we also have had a different result if we had started at Oxford or Banbury? Even if HMRC is right in contending that s 339(7) applies to changes as well as modifications, where exactly on the M40 is that line in the road where s 339(7) is triggered?
Had the place of work been moved from one building in Putney to a neighbouring building then, no matter where we start on the M40, the journey and cost are likely to always be
substantially the same – which is what the law intended.
Further light is also shed by example 4.9 in Booklet 490.
‘Anna lives near Ludlow. Each day she used to drive 25 miles north to her workplace in Shrewsbury. Anna’s job is moved and each day she now drives 25 miles south to her new
workplace in Hereford.
‘The mode of transport is the same, the time taken is the same, but the journey is in an entirely different direction. The fact that the costs of travelling to and from work do not change is a chance consequence of where she lives.
‘Had Anna lived elsewhere, the change in workplace would have had a significant effect on the cost of her commuting.
‘As such, the change in Anna’s journey would be regarded as a fundamental change to the place where her duties are carried out, rather than a modification. We regard the new location as a new and quite separate workplace.’
So, what if Mr Sambhi had lived in Putney rather than Birmingham, or in Dulwich midway between Guildhall and Croydon?
Example 4.9 says that we cannot look at where the employee lives by chance, but have to look at the journey objectively. So HMRC is contradicting its own manuals to measure the journey
cost from Birmingham to Guildhall and Birmingham to Croydon. The department should arguably, according to its own guidance, look at it equally from the point of view of a person
living in Dulwich. From the perspective of starting in Putney or Dulwich, all the journeys Mr Sambhi undertook for six years would be completely – in other words substantially – very
different. And that is the test which should have been applied.
Looking at a few paragraphs of the judgment our views are as follows.
First, at paragraph 23 we would reach a different conclusion. We would argue that the Guildhall in the City of London is not a modification of Putney. Croydon is not a modification of the Guildhall and Romford is not a modification of Croydon. These are fundamental changes in
the workplace rather than a modification of the workplace and so, in our view, s 339(7) does not apply.
Second, from paragraphs 40 to 47 there is an acknowledgement that the journeys are different and the length of travel time is different and the cost is different by as much as £14 a week, but because the difference in cost is not significant then all the places were assumed to be one place, in other words an area. But this contradicts the facts and the legislation in s 339(8) because the duties of the employment were not defined by reference to an area.
Finally, all the tribunal’s energies went into looking at the word substantial without ever analysing whether there had actually been a modification of a single location or place to see
whether s 339(7) even applied.
In conclusion, our sympathies go to Mr Sambhi and we are disappointed that although the law has remained roughly unchanged for many years, we still have tax cases where taxpayers seem to be being penalised, particularly when there is no profit element or bounty for the taxpayer.
Also, with major construction projects (on the same site) now taking so much longer than before – either by their very nature or due to delays caused by Covid-19 (HS2, terminal 5, Olympic parks, offshore windfarms, nuclear power stations and the like) perhaps it is time that the 24-month rule be extended to, say, 48 months or scrapped altogether?
Also, with more people working at home during the pandemic, does the whole concept of a commute being private mileage require an overhaul?