The new umbrella tax legislation: a word of warning
The long-awaited draft legislation on the PAYE debt liabilities of umbrella companies arrived on Monday. The derails are largely uncontroversial, and the Government is to be commended on taking a pragmatic approach to the issue, simply making agencies (or end-clients where there are no agencies) jointly and severally liable with the umbrellas. Considering what they might have done this is welcome news.
However there is a sting in the tail, directed especially at those intermediaries who do not operate PAYE because the people that the are paying are genuinely self-employed. There are quite a few of these intermediaries in the construction industry, who do in fact deduct tax under the Construction Industry Scheme but in a rather lesser quantity than they would do if forced to operate PAYE. They now might have to operate PAYE, and for reasons largely outside their control and left rather vague in explanation. Some of them call themselves ‘CIS Umbrellas’, a moniker that they would be well advised to drop because of the expectations that it raises.
Here’s how. Most of the new law assumes that the umbrella worker has a contract of employment with the umbrella company, and only applies if he does (this is in the definition of the term ‘umbrella company’). However the proposed section 61Z1 – the last clause in the new draft law – introduces the concept of a ‘purported umbrella company’ – one where the individual is not employed by the umbrella but ‘it is reasonable to suppose that one or more participants in the arrangements, other than the purported umbrella company or the individual, would assume that a substantial proportion of amounts provided to the purported umbrella company in respect of the services will be paid to the individual as earnings’.
Let’s unpack this a little. There have to be a reasonable supposition (‘reasonable to suppose’) and an assumption (‘would assume’). The assumption has to be on the part of a party to the ‘arrangements’, which might include the agency (or agencies) and the client in the chain, which seems fair enough. (Be very careful whenever you see the word ‘arrangements’ in a tax statute – it can mean almost anything, and is certainly not confined to things that are legally enforceable.) But the supposition does not. So who can make one? The client? The man on the Clapham Omnibus? The average HMRC inspector? All that the draft law says is that ‘it is reasonable to suppose’ – which can mean anyone. Maybe a tax tribunal judge will restrict this a bit some time in the future, but until that happens we are in the dark.
What this allows ‘anyone’ to do is to claim to read the mind of a participant in the arrangements, to suppose what that participant might assume about the worker’s employment status. Let us take the client as an example – the client is the participant least likely to be actually asked. This client might be a large construction company that has asked a large agency to source a lot of skilled labour for a big project – electricians, joiners, plumbers, bricklayers, forklift drivers: you name them, there they are – and is not all that bothered about their employment status as long as they are compliant: they think that that is the agency’s business. All that they have specified is that there are to be no personal service companies in the chain, as that would give them obligations to issue status determination statements and such-like – they don’t understand things properly and don’t want to be bothered with that. That client’s project manager is just the sort of person who might ‘assume’ something without giving it any real thought.
He might assume that everybody was ‘on the books’; he might equally assume that nobody was as to his knowledge skilled construction workers are generally self-employed; he might also assume that proper tests have been conducted by the agency to determine whether or not the agency rules
apply. Which of these assumptions could an outsider to ‘reasonably suppose’, bearing in mind that an ‘assumption’ is something that by its very nature something that is not the result of any detailed enquiry? Is the answer perhaps ‘all three’?
If it is, then the one that matters for this purpose is the assumption that the individuals are employed, as that one, as long as the supposition behind it is reasonable, triggers the law and the other two do not, even if they are better suppositions.
Now look at this from the point of view of the agency and the paying intermediary. The contract between the paying intermediary and the individual is patently not a contract of employment, and the agency has done due diligence work to satisfy itself that the agency rules do not apply. But this isn’t good enough. If some hypothetical outsider whom they don’t even know reasonably supposes that the client – who doesn’t want to get involved and so they have not actually asked – assumes that the individual is working under a contract of employment, that is enough to trigger the PAYE obligations on both the paying intermediary and the agency. It not only alters the party responsible for paying the individual’s tax; it creates an employer’s National Insurance out of nothing.
The only remedy for this that I can see is for the paying intermediary and the agency to make it abundantly clear to all concerned – particularly the client – that they are paying under a self-employed arrangement with no deduction of tax (or CIS only if that is the case), every time that they take on someone new. Even that does not strike me as foolproof but it ought to go a long way.
read moreHMRC get it right for once
Just had a look at HMRC’s consultation on off-payroll offsets https://lnkd.in/edPrZWGD. It’s actually rather good, allowing tax due back to taxpayers to be offset against tax paid by them when they both arise from the same mistake, and allows interest to be calculated correctly automatically.
Let’s see HMRC do something similar in the case of the old IR35 rules and managed service companies. It would remedy a rank injustice in their charging interest on balances that do not exist in any meaningful economic sense
read moreA Tax Tribunal Win
Very pleased to have helped the NWM Group win an important case on subsistence expenses in the tax tribunal – judgment at https://lnkd.in/eCg8EftU.
This goes back to the pre-2016 rules when employers needed dispensations. NWM had been paying the Revenue’s long-approved £5 and £10 scale rates for meals, but had not been auditing them after the event. HMRC expected there always to be some form of receipt to support the fact of taking a meal at work, even though it almost invariably would not say that the amount paid was £5 or £10. For that kind of money you would usually buy your lunch from a burger van or a sandwich shop – not the easiest places to get receipts from.
Anyway, the judges said that they found themselves in agreement with our counsel’s submission that ‘if receipts were necessary irrespective of the subsistence scale rates applying, then one wonders what benefit would be derived from having the scale rates at all. Those rates were supposed to be an administrative convenience for employers, employees and HMRC. But the processes apparently demanded by HMRC might arguably be more onerous than claiming deductions against tax for expenses in the amounts substantiated by the receipts. That cannot have been what Parliament intended.’
In other words, this was a bonkers regime imposed by HMRC all along. Congratulations to David Ewart KC and Laura Ruxandu of Pump Court Tax Chambers for shedding the light of day on it.
The autograss regime and CEST
This post is for big business readers. That is to say, for those whose group has a UK turnover of £200m or more or a balance sheet total of £2bn or more. If you’re not working for or advising one of these, you may still like to see how big business is going to tie itself in knots before it has to surrender itself to the Revenue and tell them all it knows. This may fill you with delight when you see that your competitors have to do something that you don’t, or possibly with horror at the thought of this regime coming to your company as well at some later date.
As from next April, big business is going to have to tell HMRC whenever it has adopted an ‘uncertain’ tax treatment that might have saved the company £5m or more (a drop in the ocean for companies this size). This is all described here, and ‘uncertain’ means one of three things:
- A provision has been made in the accounts, to reflect the probability that a different tax treatment will be applied to a transaction.
- Reliance has been placed on an interpretation or application of the law that is different from HMRC’s known interpretation or application.
- It is reasonable to anticipate that, if a tribunal or court were to consider the way in which the amount was arrived at, there is a substantial possibility that the treatment would be found to be incorrect.
Now I’m not going to go through the whole gamut of uncertain tax treatment that those clever creative types in the Big Four might be advising their clients to take, as in most areas of tax I am not familiar with them. However it is I think worth taking a look at what the implications of this might be for the off-payroll regime, known to old-fashioned people as IR35. Employment status is an inherently uncertain matter and one in which there are acknowledged grey areas. (Even HMRC will agree that much if you press them hard enough, although I have rarely seen them give any leeway to taxpayers because of it.) Basically, it means that big companies are going to have to choose the most conservative treatment possible (i.e. employed status, or inside IR35) if they are going to avoid certain trouble. Some might say that they are already doing that, which I suppose is true of many large companies, but it is not true of all of them.
Criterion no. 1 above should not present too much of a problem. My observation in the days when I did company accounts – admittedly some time ago – was that clients would go to any length to avoid having to make a provision of this sort, as doing so was an advertisement to the Inland Revenue (as they then were) that the company did not think that it had a very good case – something that generally speaking the adviser would be no keener to say than the client company would be. A reason as usually found for not putting a provision in, with varying degrees of intellectual contortion required to justify this stance.
Criterion no. 2 raises some interesting questions about the role of CEST, as you can find out HMRC’s application of the law by using this well-publicised tool of theirs. Do you have to put each case through CEST in order to find this out? I don’t think that the draft legislation goes quite that far: it talks of ‘the way in which it is known that HMRC would interpret or apply the law’ (my emphasis). We know that HMRC use CEST, but we do not know the result in any individual case until we use it ourselves. So, well then, would it perhaps be better not to know? Bearing in mind how controversial CEST is, and the difficulties that some public sector bodies seem to be having in using it properly, I could hardly think of a better reason not to check out one’s own methods against CEST, which one might otherwise want to do in order to establish the risks that one is running.
HMRC’s draft guidance does in fact mention CEST as a possible trigger for this ‘uncertain treatment’ treatment, saying that it will have to be reported in a case where, to quote:
‘A large business uses “Check Employment Status for Tax” (CEST), which gives an “employed” outcome. However, additional factors mean the business considers it is appropriate not to treat their contractors as employees for tax purposes.’
It is interesting that HMRC consider that there might be additional factors to consider here, bearing in mind that they have an ‘unable to determine’ response that absolves them from giving a view in about 20% of cases. But I suspect that where a business differs from CEST on a status test, it will more often be because those in the business simply believe CEST to be wrong, rather than because of any ‘additional factors’.
However it is the third criterion that opens up a real horror show. What is a ‘substantial possibility’ that a court or tribunal might disagree? A 20% chance? Or 30%? Perhaps 40%? It is certainly below 50%. The use of the word ‘substantial’ that tax practitioners will be most familiar with is that in s. 165A of the Taxation of Chargeable Gains Act 1992, where the definition of a trading company is said to be ‘a company carrying on trading activities whose activities do not include to a substantial extent activities other than trading activities’. HMRC’s Capital Gains Tax Manual says with supreme self-assurance: ‘Substantial in this context means more than 20%’. So are we to take a 20% chance here too? Or is this different? It strikes me that it is not a counsel of excessive caution to suggest that 20% might be a reasonable hurdle to have to cross here too.
Now go and ask a barrister when was the last time he took on a case that had a lower-than-20% chance of losing. It doesn’t exist: all litigation is inherently uncertain. One barrister I once asked for odds on winning an employment status case that I thought looked pretty good came back with 55%. I said that I understood that she thought the case was pretty good too. She said that she did: 55% was the maximum odds that she would ever give in a status case.
Remember too that the criterion is that there is a substantial possibility that ‘a court or tribunal’ would find the treatment to be incorrect. That does not just refer to a judgment considered by a majority of three well-known and experienced judges in the Court of Appeal; it means any old judge in the first-tier as well. My own view is that the First-tier Tax Tribunal judges are on the whole pretty impressive, but the odd eccentric judgment does sometimes come out as we know from the number that get overturned on appeal. Basically, if taken at face value this means that any self-employed stances will need to be reported if the tax at stake comes to the stated threshold £5m or more. It is perhaps just as well that we do not have to consider what might happen in the Employment Tribunal or the County Court.
Obviously one way that big companies might decide to avoid this is to engage agencies with a view to making them responsible for operating PAYE. As agencies are usually not that big (although one or two are), this might well work, but is this really desirable? And for agencies that do exceed the thresholds, what are the implications for them? They are unlikely to have any notion as to whether the tax treatment is uncertain or not, as they are not generally aware of how the status criteria apply to individual cases – that is the whole point of making clients produce status determination statements, so that the agencies can be told what to do. Again it is the third criterion that risks tripping them up.
So it’s time to prepare for this grass-yourself-to-the-cops regime. When does it start? Too bad – you may well be already in it. It is going to apply to the first ‘relevant return’ due after 1st April 2022. PAYE returns are of course submitted monthly, and the relevant return is the last one in your accounting period. So if you have a June year end, you will need to file your report along with your PAYE full Payment Summary for the period ending June 5th 2022, and will need to include any uncertain treatments adopted in any of the twelve monthly returns leading up to that point, i.e. from 6th June 2021.
I mentioned above that some public sector bodies appeared to be having difficulty using CEST properly, so are they included in this regime too? No such luck – this only applies to companies and partnerships. Maybe HMRC feel that they can monitor them anyway.
David Kirk.
3rd September 2021
David Kirk discusses the IR35 reforms with Daniel Barnett
Watch David Kirk in discussion with the famous employment barrister Daniel Barnett. IR35 has suddenly become interesting to employment lawyers.
read more