Taxation logo taxation mission text

Since 1927 the leading authority on tax law, practice and administration

Walking the tightrope

25 August 2009 / Andrew Hubbard
Issue: 4220 / Categories: Comment & Analysis
How good are you at your job, asks ANDREW HUBBARD. And what is good and what is bad?


  • Responding to the Working with tax agents consultation.
  • What constitutes good tax work?
  • Evidence of poor tax work.
  • Tax law is too complex.
  • HMRC staff must also be seen to meet high standards.

What a question! Of course you are good at your job. You are a professional tax practitioner who has spent years honing your skills.

The fact that you are reading this article shows that you are keeping your CPD up to date – unless you are reading this in the year 2019 and have only just got round to working your way through the pile of Taxations on your desk.

So let me refine the question slightly: in your capacity as somebody who prepares or reviews returns for clients (be those income tax, corporation tax, PAYE, VAT or any other type of tax return) how do you know that those returns meet acceptable quality standards?

Later in this article I will reframe the question to address those of you whose job it is to check those returns within HMRC. Whatever side of the fence you sit, there is something for you in this article.

Importance of high standards

The context of these questions is the Working with tax agents consultative document. Underpinning that consultation is the whole issue of the quality of the work of agents.

Readers will be forming their own views as to what the appropriate response to the consultation should be.

In my discussions with tax agents so far I have heard a wide range of views, from support for regulation through to maintenance of the status quo, but I take it as read that everybody is interested in monitoring and enhancing the overall quality of tax work in the UK. Included within that is the work of taxpayers, agents and HMRC.  

It seems quite possible that one of the outcomes of the consultation might be that HMRC will require certain minimum performance standards from agents as a condition of allowing them to submit returns.

I stress that I have no inside knowledge of this, but it does seem a reasonable assumption. Let’s leave aside the issues of whether any system should be voluntary or compulsory, structured as regulation or registration, and whether it should be opt in or opt out.

Those are no doubt battles to be fought later. Instead I want to concentrate on what ‘good’ and ‘bad’ work by an agent might actually look like.

The good and the bad

It is surprising just how little this issue has been addressed in the past. There is a general assumption that we automatically know what good work is, but are those assumptions the same across the profession? All of us have, after all, had the experience of joining a new firm or being joined by somebody from another firm and getting the reaction ‘that’s not the way we do it’.

Let’s get a few things out of the way. First, I am not dealing here with professional standards as set out in documents such as Professional conduct in relation to taxation and Practice rules and ethical guidelines.

These underpin everything we do and I take it for granted that the agent is working to those standards. But they, perhaps understandably, say very little about the way that an agent goes about completing tax returns.

Second, what little debate there has been on the subject has inevitably focused on the extremes. At one level there is what we would all recognise as a lack of the basic skills, i.e. the inability to add up, deducting depreciation rather than adding it back, calculating taper relief on gains made by companies etc.

We all make mistakes but there comes a point where errors become so frequent and pass undetected despite review that we have to put them down to incompetence.

At another level there is fraud, for example, deliberately putting a director’s personal expenses through purchases so as to conceal a benefit in kind. Let’s ignore these extremes. Let’s also ignore questions of tax avoidance or taking extreme tax positions.

Instead I want to look at how to define ‘good’ and ‘bad’ behaviour in relation to the sort of issues which many readers of this magazine will face every day of their working lives. I have chosen three very common issues.

Private motoring expenses

Assume you act for a small partnership. What is ‘good’ in relation to dealing with the add back of private motoring expenses?

Assume that there is no technical problem in distinguishing between private and business motoring: the question here is how far is it reasonable to expect an agent to go in determining the facts?

Here are some of the approaches which could be taken.

  • Make no add back unless the client volunteers one.
  • Ask the client to give you a figure, but do not give him any guidance on the difference between private and business motoring.
  • Ask the client to give you a figure and give him some guidance on that difference.
  • As above, but also ask him to substantiate the figure by means of mileage logs.
  • As above, but also review and carry out your own mileage logs, for example by testing the mileages using a journey planner programme.
  • As above, plus checking the taxpayer’s private diary to test that the mileage logs are consistent with the private mileage declared.
  • And so on.

How should you respond to the following reactions from the client?

  • I’ve not got a clue: what percentage do most people use?
  • What is the minimum which is likely to be acceptable to HMRC?
  • Just use the same as last year. (But what do you do when you know the client has moved house and now lives further away from his office?)
  • Just use the same as for my other partners. (But what if you know that this particular partner lives close to the office and has a deskbound job whereas the others live further away and spend much of their time out of the office visiting clients?)
  • My previous adviser never asked me any of these questions and always included only a nominal add back: it was never questioned by HMRC. Why are you making such a fuss about this? Am I ending up paying you higher fees for the privilege of you increasing the tax I have to pay?
  • Its all business mileage – I’m not giving you a figure.
  • Let’s not put anything down and see if HMRC asks.

I could go on, but you get the point. Where is it reasonable to draw the line? More important in this context, where does HMRC think that it is reasonable to draw the line?

Suppose that there is an enquiry into this particular return and that it is established that the true level of private motoring is much higher than was disclosed.

At what point does HMRC take this as evidence of poor work by the agent? It is after all the client’s return and the agent is only acting on the client’s instructions.

But there must be a point when the agent has to take some responsibility, so when is a failure to do so regarded as ‘bad’ work?

I suspect that most of us would take the view that if the agent was in the first of the possible approaches listed above, HMRC could legitimately say that he had failed to do a professional job.

A firm whose computations consistently failed to show any private motoring adjustments across a large number of clients would probably be seen by most of us as not meeting minimum professional standards.

However, once we move away from that first approach, I would imagine that there might be less certainty. My gut feeling is that most agents would think that the third approach would be about the right level and that anything more was over the top.

When it comes to HMRC, I suspect that they would expect that the right level of professional care would be in the fourth or even fifth approach. If that is indeed the expectation then arriving at an agreed position on what constitutes acceptable professional behaviour by agents is going to be difficult.

Small company rate

What about an area which is a mixture of fact and law? Let’s ask ourselves similar questions about determining whether or not the small company rate of tax is due. I think that we would all accept that this can be a very complex technical area.

I have no problem in admitting that I find it impossible to keep all of the rules in my head and need to consult my reference books in cases of trusts and associates. I would hazard a guess that I am not alone in this.

It is also my suspicion that there may be more associated companies than are acknowledged in computations. That is not because of deliberate concealment, it is simply a consequence of the wide scope of the legislation and the sheer difficulty of applying the rules in practice.

No client can possibly be expected to know the rules about associated companies and this is an area where it is reasonable to depend on the agent to deal with the technicalities. But the agent can only do this with knowledge of the facts.

How far then should an agent who is attempting to do a proper professional job go in determining the number of associated companies?

In theory an agent could send the client a form listing all of the possible ways in which another company could be associated with his company and ask him to fill it in.

But that form would run to many pages and some of the questions would be extremely complex. In practice almost no clients would be prepared to fill this in. The reaction of most clients would be ‘you do my books – if you don’t know the answer how on earth can I be expected to know?’

Most agents are likely to tackle this problem either by making the best effort they can using the information which is readily available, or by asking the clients to confirm that they do not control any other companies, but without going into an exhaustive definition of what control is.

Is that good enough? Would an agent who prepared a return on this basis, only to find out that there were in fact other associated companies, be open to the challenge that he had not acted competently?

Agents would, quite rightly, be aggrieved by the suggestion that they were at fault in such cases. Dealing 100% accurately with associated companies is almost impossible and in the end most agents do the best that they can within the cost constraints under which they are operating.

The problem here is ultimately not agent behaviour but the sheer complexity of the underlying tax legislation.

So what is it reasonable for an agent to do in order to determine the number of associated companies? If we don’t know that, it will be impossible to have an objective assessment of whether an agent has done an acceptable job.

Capital versus revenue

My third and final example is the capital versus revenue divide. Let’s look at something which is on the borderline between a repair and an improvement. Assume also that the agent has the full facts and that we are therefore dealing purely with a matter of categorisation.

This is a matter for the agent – the client is unlikely to be involved in the decision making.

How should agents deal with this issue, assuming for the sake of argument that the categorisation will affect the amount of tax which is due? It is worth noting that the tax legislation imposes no specific disclosure requirements here – a person is simply required to submit a return which is correct to the best of his knowledge and belief.

So an agent could simply make the decision that the item is a repair and complete the return on this basis with no additional disclosure. Would he be right to do this?

Does it depend on how confident he is in his judgment? Should the approach be different in a case where he rates the argument as 90-10 in favour of it being a repair to that where he rates it as 50-50? What happens if he rates the chances of being only 10% in favour of it being a repair?

He can convince himself that there is a ‘filing position’ to treat is as a repair but would he have much confidence that he would win the argument if challenged?

Does the way that this is disclosed make any difference? Is the agent’s position better in the eyes of HMRC if the return clearly shows the expenditure on the disputed item and perhaps gives some narrative?

Or is there some protection in putting disclosure in the white space? I suspect most of us would probably say that there was.

But if the agent submits a return which just treats the expenditure as allowable, has he met appropriate professional standards?

These are deep waters and, in my experience, the issues are almost never discussed; we are all supposed to understand what to do by some process of osmosis. But that would not be an acceptable starting point if agents were personally accountable for ‘bad’ behaviour.

Could an inspector turn round and say ‘you didn’t tell me that there was a disputed item in the computations – that was unprofessional conduct and therefore that gives you a black mark in the book’?

One can all too easily see this happening – think about Langham v Veltema [2004] STC 544 for a start. But what is the legislative basis for this?

The Keith Committee, when looking at these issues a generation ago, came up with a recommendation that taxpayers (and by inference their agents) should be obliged to disclose to HMRC circumstances where they had given themselves the benefit of the doubt in a tax return.

That idea was rejected at the time and has never been implemented. But might it return by the back door as part of evaluating the standard of work of an agent?

That would be wrong. I think that a proper understanding of if and when additional disclosure is appropriate would be a very welcome development: indeed it would be one which will be essential if the work of agents is to be evaluated.

The necessary framework would have to be developed openly and in partnership between agents and HMRC. Let’s hope that this is one of the conclusions of the current consultation.

Good work by HMRC

It is also worth asking the question about what constitutes ‘good’ or ‘bad’ work by HMRC. Any move to improve standards and define professionalism in compliance matters must be capable of a read across into HMRC.

That is a massive topic and one on which I am sure readers (whatever side of the fence they sit) would have entrenched views. I do not propose to go into this here in any detail but in view of the space available I would raise just one matter among the many which I could discuss.

What is the appropriate professional conduct by HMRC to a disclosure made in computations? Suppose a computation contains a specific disclosure in respect of a one-off item.

Let’s use the same example as above and say that it covers the categorisation of an item as a repair rather than an improvement – what does that disclosure trigger in the mind of the inspector? Is it that:

  • The agent has clearly thought about this and has come to a considered conclusion. I trust his professionalism and therefore am happy to accept his view? or
  • Why has he told me this? If he were confident of the position then he would not have needed to make any additional disclosure. So he must be worried about this and therefore I need to make further enquiries so that I can find the truth.

I would like to think that it is usually the former, but I suspect that more often it is in fact the latter. So a stance which is intended by the agent as being open with HMRC by showing that the point has been properly considered works in reality against the client and leads to an enquiry.

Had the additional information not been given, there is probably a greater chance that the return would have been accepted without enquiry.

If we are to define standards of what constitutes acceptable work by agents we need to explore the idea of agreeing similar standards for the work of HMRC. Without this bipartite agreement, it is difficult to see how we could every get anything meaningful off the ground.

Where does this leave us?

My purpose in writing this article is not to give an answer but to try and frame the question. I support the idea of driving up standards of work within the tax system, but if we are to do that we have to understand what we mean by ‘good’ and ‘bad’ work.

It is only by looking at the sorts of practical issues which I address here (and I could have given many others) that we will begin to drill down into the detail of what acceptable standards might mean.

The role of the agent is to represent his client and act on his behalf; it is not to act as an honest broker between the client and HMRC. Agents are in business to make money from their skills, probably with the motivation of helping clients as they do so.

They want to exercise those skills in a way which gives them an income and which enables them to operate to high professional standards. They do not operate in a vacuum or on a purely theoretical basis – they make judgment calls every day about almost every aspect of a tax computation.

They have to deal with clients who may be poor at record keeping, who have little knowledge or interest in tax matters and who leave things to the last minute. They do not have the luxury of infinite time and money to create and implement processes to give a perfect answer.

Any attempt to define what constitutes acceptable and unacceptable standards of work has to be rooted in this commercial reality. If we do not do this then the enterprise is doomed before it even starts.

But if we put the issue in context I am confident that it will be in the interests of taxpayers, HMRC and agents to work towards developing, for the first time, some form of code of practice (let’s not get hung up on the name) which defines quality standards in all aspects of tax compliance.

And then you will really be able to answer the question that I posed at the beginning of this article.

Andrew Hubbard is president of the Chartered Institute of Taxation and tax partner at Tenon

Issue: 4220 / Categories: Comment & Analysis
back to top icon