• The government wants to improve the effectiveness of the IR35 legislation.
• HMRC have issued a discussion document.
• IR35 is ineffective and the government or HMRC would like to find a solution.
• The rules in fact require the taxpayer to pay more than their fair share of liabilities.
• Why should the secondary National Insurance liability not be paid by the person for whom the work is actually done?
• Many contractors using personal service companies are working for the public sector.
• It is time to scrap IR35 and replace it with something that works.

They may have been easily overlooked by readers of Taxation, but tucked away on page 44 of the summer budget’s Red Book, and followed a few days later by a discussion document with the innocuous title of Intermediaries Legislation (IR35), were the following two paragraphs:

As always, it is helpful to understand who is behind this initiative: HMRC or the government? I do not know the answer for certain, but these paragraphs came as a real surprise because HMRC have already been “having a dialogue” with those parts of business most affected by IR35. This has been going on since 2011, when the IR35 Forum was set up at the suggestion of the Office of Tax Simplification, its aim being to “advise on improvements in the administration of IR35”.

There seems to be a lack of joined-up thinking here. The only obvious results of four years of deliberations were the famous business entity tests. Unfortunately, these had to be withdrawn after a couple of years because the public sector was incapable of operating them properly. As for the private sector, the reality is that hardly anyone pays any attention at all to IR35, even though there is an obligation on personal service companies to self-assess IR35 liabilities if they arise. This cannot be condoned, but it is a fact of life. It is also the case that very few cases are actually investigated. We have ended up in an environment which does not work for anybody.

What is really doing on here? HMRC’s document is revealing. Having discarded several other options which they recognise as ineffective, the document suggests that “those who engage a worker [emphasis added] through a PSC [personal service company] would need to consider whether or not IR35 applies … and, if so, deduct the correct amounts of income tax and National Insurance contributions as they would for direct employees.”

I will consider the implications of this later in this article, but first one very important point needs to be made.

The Red Book referred to above says “IR35 … requires that [those it affects] pay broadly the same tax and National Insurance as other employees.” This is fundamentally wrong. It actually requires quasi-employees to pay very much more than this, because they have to pay the employer’s share of National Insurance as well as their own, while the business that HMRC regard as the employer is for some reason let off this impost. The “employer” may pass as the person paying it in the eyes of some in government, but to purveyors of plain English, and to those in the real world who might be affected by the tax, this is plainly not the case.

The idea presumably is that the “employer” pays some extra money to the contractors which they can use to pay this levy. Like all such utopian ideas, this just does not happen.

The effects are startling as shown in The IR35 Effect. The marginal rate of tax is 48.6% for a higher-rate taxpayer on IR35, as opposed to 42% for an employee and 39.2% as opposed to 32% for a basic-rate taxpayer.

Any analysis which does not take this basic point into account is fundamentally flawed.

Let me put forward a suggestion. Instead of trying to “improve the effectiveness” of IR35, which is broken beyond repair, why not replace it with something simpler? This could be levying a new class of employers’ National Insurance on those that HMRC deem to be employers, when payments are made to companies providing their labour. There will doubtless be squeals from the very same businesses that lobbied the government into backing down from making them operate IR35 when it was introduced in 2000, but this suggestion is a great deal simpler than that original proposal. That would have made the “employer” responsible for the whole of the PAYE and National Insurance contributions, whereas this suggestion leaves the employees’ side as it is.

“Employers’ National Insurance should be paid by employers” is, I think, pretty hard to argue against as a principle for raising tax.

While “employers” will doubtless protest, how many of them would actually be affected by such a proposal? A good many of the contractors that IR35 is aimed at work for the government. Employers’ National Insurance, in this instance, is circular money travelling around various government departments and it should not be difficult to find a solution that satisfies everybody. These comments also apply to local government and large charities, because these are substantially funded by the central government too.

So what about the private sector? Banks have, traditionally, been wholescale engagers of labour via personal service companies and would be affected by my proposal, but in the current political climate they are not in a strong place to lobby against additional tax liabilities. What about the construction industry? There may well be protests from that sector, but it is not noted for the widespread use of personal service companies.

The reality is that if the government were to put this charge on to deemed employers, it would be easier to collect and more likely to be paid, without any theoretical change in the circumstances giving rise to the tax. Facing down the lobbying cries would be a price well worth paying for that.

It is when the attempt is made to force the deemed employers to deduct employees’ National Insurance and PAYE that one can expect trouble. With the new dividend tax coming in next year this ought not to be necessary anyway – there will be little incentive to incorporate if employers’ National Insurance is paid by employers.

The PAYE system is not geared up towards collecting money from the end-users of labour when there is a string of companies in the way, and trying to do so is bound to be messy. If the users of labour were to pay the secondary National Insurance contributions, there would obviously need to be some system for identifying what has been paid, so that if the PSC pays a salary above the threshold this does not have to be paid twice. That, though, would be a minor piece of bureaucracy compared with many and a small price to pay for getting rid of some of the more intrusive and complicated requirements of operating PAYE. That there is very little incentive to incorporate now is shown by Dividend v Salary.

Dividend v Salary compares an employee on a salary with a taxpayer with his own company who has the employers’ National Insurance paid by the “employer”, after the new dividend regime comes in in 2016. Under the “dividend model”, the person takes a salary equivalent to the personal allowance; I have also assumed that his company pays £1,500 in accountancy fees.

On gross pay of £20,000, he is now better off without a company. After that, the advantage moves towards incorporation, but starts coming back down again, because – for higher rate taxpayers – the marginal rate is 46% for those on the company model against 42% for those on the salary model. The point of maximum advantage for companies is about £58,500, giving a difference of a little over £3,000. At that level there might be some tax-motivated incorporations but, bearing in mind that it is barely worth it at both lower and higher levels, I suspect that it is not something that accountants will suggest as a matter of routine. Many people do not like the hassle of having to run a company and require a bigger incentive than is shown in the Dividend v Salary table to feel that it is worth the bother. Further, there is always the danger that these tax rates will be tinkered with in the future so as to reduce the differential still further. Perhaps the government could take a deep breath and abolish IR35 altogether, just collecting the employers’ National Insurance as suggested above?

It is clear beyond any doubt now that IR35 is not just an unenforceable and punitive tax provision: it is deeply flawed in its design in a number of ways. One of the most controversial things about it is that, when investigating and as a matter of routine, HMRC must go to the contractors’ end-clients to get information about their contracts and the way that they work. In the private sector this causes huge resentment. Contractors feel, in some cases with justification, that their clients will think that they are causing trouble and wasting their time and will want to get shot of them. HMRC compliance activity ought not to lead to the loss of work.

It is not as if this kind of risk is something that ought to go with the turf. In certain industries, a contractor will probably have heard about IR35, but what does he find when he wants to know whether it applies? Remember that there is legal obligation to self-assess each and every contract to determine whether or not IR35 applies.

Consider a contractor who wants to be compliant. The HMRC’s Employment Status Manual is 370 pages long and virtually all of it is relevant to the subject. What happens if the contractor asks his adviser to deal with it? In reality, accountants are struggling with this, never mind their clients. Even if they can wade through the department’s manual, they will inevitably have to make judgments for which they are ill-equipped and which they have little incentive to research. Even if they were minded to add a course on this to their continuing professional development for the year, remember that we are talking about general practice accountants who coping with a number of other changes being thrown at them at the same time.

It will be truly awful when the new regime for travel-to-work expenses comes in. The normal employment status tests for deciding whether the company is within IR35 will still be necessary, but completely different tests (purely based on supervision, direction or control as to how the work is done) will be needed to determine whether travel-to-work expenses can be claimed. Who on earth thought of that one? I think I can say with complete confidence that few, if any, advising accountants or inspecting HMRC officers will get this right.

Once HMRC have done their work and concluded that IR35 does apply, the department will send the client a bill which can be truly mind bending. If HMRC have gone back six years, it may well be three times the company’s actual turnover. But surely the client will not have to pay as much as HMRC’s figures suggest, because they have already paid higher-rate tax on the dividends? This is where time limits can be a real problem. The time limit under ITEPA 2003, s 58 is five years and 10 months from the end of the tax year, so if the investigation has gone on for a long time, and in my experience it often does, the early years do not count. Protective claims need to be put in as soon as an investigation starts.

Then there is corporation tax. As readers will know, the effect of IR35 applying is that there is a “deemed payment” of salary, with accompanying employers’ National Insurance. In principle, that gives rise to additional corporation tax relief. I say “in principle” because time limits are a problem here as well. The limit here is four years from the financial year end. So the taxpayer will probably lose for the earliest years even if a claim is made as soon as IR35 becomes an issue. There is the possibility of special relief under TMA 1970, Sch 1AB, but who wants to have to rely on that?

If this were not bad enough, there is a further problem. Having just explained the counterclaims that can be made on behalf of the taxpayer, the adviser will find that the department has an “interesting” way of further increasing the client’s bill. If HMRC go back six years, interest can be charged on the IR35 claim at 3%. The higher rate and corporation tax counterclaims come back with interest at 0.5%. Advisers might think that HMRC would do what is obviously fair and offset the latter against the former – but no. They will find (in my experience at least) that there is no offset until the matter is finally settled, so the department can make a 2.5% turn on a balance that exists only on paper and is not actually owed. In this way, the interest will very likely add 50% to the overall bill. Whatever the strict legal position, clients in this position feel that the result here is very unfair – and I agree with them.

The HMRC IR35 helpline also merits a mention. This has been a consistently good source of advice throughout the unhappy history of IR35, and users are promised that “any information you give won’t be shared with HMRC compliance teams”. However, I gather that I am not the only person to have found somebody working on both sides of this particular Chinese wall. I am not the only person to have discovered the same inspectors working on the helpline as in compliance teams. While I can accept that information will not be deliberately shared with compliance teams, what procedures are in place to see that this does not happen accidentally? This does not look right.

If IR35 appears to have been designed with the needs of big business in mind, there is a little corner of industry where HMRC practice increases its advantage. In 2008, HMRC agreed a set of guidelines to determine employment status for the radio industry. However, although the big fish in this business know all about them, it seems that nobody thought about letting IR35 advisers into this. Should an adviser be lucky enough to know of their existence, a copy can be obtained very easily by phoning HMRC’s TV and Broadcasting Unit on 0161 261 3254; the guidelines will arrive by return. But can they be found on the HMRC website? No, and I am told that there are no plans to put them there. For the benefit of practitioners needing to give IR35 advice to contractors in the radio industry, they can be found here.

This is not fair. Let’s recap. We have a tax that allows big business taking advantage of this labour to say “nothing to do with us”; is incapable of proper enforcement by HMRC; those potentially affected cannot be expected to understand; their advisers cannot be expected to understand either; requires clairvoyance to put in the correct counterclaims against HMRC to stop them getting away with more than they are entitled to; and allows the department to contact a taxpayer’s client and disrupt relations between the two. Finally, this is a tax where inadvertently getting it wrong risks leaving the taxpayer in a state of financial ruin. To make it worse, this oppressive and unjust tax is causing all this havoc for the sake of raising £550m a year or, so HMRC say, enough to keep the government going for a mere six-and-a-half hours.

It is time for IR35 to go. I have mentioned one possible solution to it in this article – there are doubtless others and some were proposed in the Office of Tax Simplification’s report on employment status last March. When the time comes for HMRC to have a dialogue with Taxation readers, I think that there should be a robust response: “We are not interested in making IR35 work. We’ve been having a dialogue about it for years. It can’t work any way you try it. It is time to scrap it completely and replace it with something that collects the tax from those who ought to be paying it. Employers’ National Insurance should be paid by employers.”