Why historical remediation is not simply a matter of reopening six years
In my previous article, I considered an international group that had identified a UK permanent establishment of its overseas principal.
The group had attributed £3.2 million of profit to the PE. It had then reconciled that amount with the UK subsidiary’s transfer-pricing adjustment, the overseas company’s tax position and the UK CFC computation.
The tax director’s next question was inevitable:
How many historical periods do we need to correct?
The finance team’s immediate answer was six years.
I often hear that answer. It is usually based on a general recollection of HMRC’s assessment powers rather than an analysis of the particular company and the particular failure.
In this case, six years might have been right. Four years might have been right. In some circumstances, HMRC might have been able to assess as far back as twenty years.
The answer depended on what had happened, why the overseas company had not paid UK corporation tax and whether it had complied with its obligation to notify HMRC.
The PE did not begin with a formal event
The group’s UK activity had developed gradually.
At the outset, the UK subsidiary employed three people and provided marketing and administrative support under an intercompany services agreement.
Eight years later, it employed thirty-two people. Some UK executives had regional responsibilities. They negotiated significant customer arrangements, managed commercial risks and approved exceptions to pricing policies formally owned by the overseas principal.
There was no board resolution transferring those functions to the United Kingdom. No agreement recorded the date on which the business model changed.
That is not unusual.
A permanent establishment often emerges from a succession of operational decisions rather than one identifiable legal step. The contracts remain unchanged while the people, authority and decision-making move on.
The first task was therefore to determine when the facts first satisfied the domestic and treaty PE tests.
Only then could the group consider which of those periods remained assessable.
These are separate questions. A factual review may need to cover ten or fifteen years even where the ultimate UK assessment period is shorter. Earlier years may reveal when authority moved to the United Kingdom, how the functions developed and whether the PE existed continuously.
In practice, I would not begin by opening six columns in a spreadsheet. I would begin by reconstructing the business.
Four, six or twenty years?
For UK corporation tax, the normal time limit for a discovery assessment is four years after the end of the relevant accounting period.
The period may extend to six years where the loss of tax was brought about carelessly by the company or a related person.
It may extend to twenty years where the loss was brought about deliberately or was attributable to a failure to comply with the statutory obligation to notify chargeability.
HMRC summarises the corporation tax time limits in its Enquiry Manual.
Those rules cannot be applied simply by looking at the size of the adjustment or the complexity of the technical issue.
A difficult PE question is not automatically evidence that the company took reasonable care. Conversely, reaching the wrong conclusion does not, without more, establish carelessness.
The real issue is the process by which the company arrived at its position.
Who considered the UK tax consequences? What facts were available? Was professional advice obtained? Were the advisers given a complete picture? Most importantly, was the position reconsidered when the business changed?
Historic advice did not necessarily protect the later years
The group had obtained professional advice when the UK subsidiary was incorporated.
On the facts presented at that time, the adviser concluded that the overseas principal should not have a UK PE.
That advice was plainly relevant. It was not a permanent certificate of non-PE status.
The UK operation subsequently became much larger and more commercially significant. The contract approval matrix changed. UK executives acquired wider responsibilities, and the distinction between supporting an overseas decision and making the decision became increasingly difficult to sustain.
No one revisited the original advice.
This is where I would be cautious about relying too heavily on the fact that advice once existed. Reliance on professional advice may support a reasonable-care position where the adviser was properly instructed and received the relevant facts. Its value is far more limited where the facts later changed materially and the business had no process for reconsidering the conclusion.
The question was not whether the group had sought advice eight years earlier.
It was whether a group of that size could reasonably continue relying on it after the operating model had changed.
“We did not know” was not enough
The first draft of the proposed disclosure stated that the overseas company had not notified HMRC because it did not realise that the UK activities created a PE.
That explained very little.
A proper conduct analysis needed to identify who was responsible for considering the company’s UK tax obligations and why no further review took place.
The group therefore examined board papers, advice received, internal correspondence, reporting lines and changes to the delegation-of-authority matrix. It also considered whether concerns had previously been raised by the tax team, auditors or local advisers.
The review could not be limited to the current tax director. Relevant conduct might involve directors, finance personnel, UK management, previous advisers or others acting on behalf of the overseas company.
This did not mean attributing every employee’s knowledge to the company indiscriminately. It meant identifying the person responsible for the act or omission that produced the loss of tax.
For each accounting period, the group recorded the relevant UK activities, the advice available, any material operational changes, whether HMRC had been notified and the preliminary view of the company’s conduct.
Only then did it consider the statutory time limit.
Failure to notify was a separate issue
The overseas company had never registered for UK corporation tax and had filed no UK corporation tax returns.
If it had traded in the United Kingdom through a PE, the problem was not merely an inaccuracy in a return. The company may also have failed to comply with its obligation to notify chargeability.
That distinction was potentially important because a loss of corporation tax attributable to a qualifying failure to notify can fall within the twenty-year assessment period.
The twenty-year rule should not be applied automatically simply because the company was unregistered. The notification obligation, the relevant deadline and the link between the failure and the loss of tax still need to be established.
The company also needed to consider whether it had a reasonable excuse.
For a sophisticated international group employing UK personnel and operating intercompany arrangements, it would be difficult to rely only on the complexity of the PE rules or the fact that the company was incorporated overseas.
A reasonable-excuse analysis is fact-sensitive. HMRC’s guidance recognises that the circumstances of the particular taxpayer must be considered, but it also expects the failure to be remedied without unreasonable delay once the excuse has ceased. See HMRC’s reasonable-excuse guidance.
A twenty-year assessment period did not automatically mean a deliberate penalty
There was another misconception within the finance team.
It assumed that, if HMRC could assess twenty years, the failure must necessarily be treated as deliberate.
That does not follow.
A twenty-year assessment period may arise because the loss was attributable to a failure to notify. The penalty analysis separately examines whether that failure was non-deliberate, deliberate, or deliberate and concealed.
For an onshore failure to notify, the potential penalty is calculated by reference to the potential lost revenue. The applicable range depends on the behaviour, whether the disclosure is prompted or unprompted and the quality of the disclosure.
HMRC’s published ranges for an unprompted disclosure can extend from:
- 0% to 30% for a non-deliberate failure;
- 20% to 70% for a deliberate failure; and
- 30% to 100% for a deliberate and concealed failure.
The minimum ranges are generally higher where the disclosure is prompted. HMRC sets out the framework in its failure-to-notify guidance.
I would not put those percentages into a tax provision until the underlying conduct had been analysed. Doing so creates an appearance of precision without answering the most important question: why did the failure occur?
Was the disclosure really unprompted?
The group intended to approach HMRC voluntarily and assumed that the disclosure would therefore be unprompted.
Again, the position was not quite so straightforward.
Broadly, a disclosure is unprompted if, when it is made, the company has no reason to believe that HMRC has discovered or is about to discover the failure.
The absence of a formal PE enquiry does not necessarily decide the point.
In this case, the PE issue had emerged during a wider transfer-pricing and CFC review. HMRC had already asked questions about the functions of UK executives and the overseas principal’s intercompany arrangements.
The group therefore needed to consider whether those questions already pointed towards the unregistered foreign company.
I have seen groups weaken an otherwise credible disclosure by describing it as unprompted without first reviewing the history of HMRC contact. The chronology should be established before the disclosure is submitted, not debated after HMRC challenges the proposed penalty treatment.
Correcting the position required more than one procedure
The phrase “correct the historic years” suggested a single administrative exercise.
There was no single exercise.
A return still within its amendment window might be amended by the company. A filed period outside that window might need to be regularised through disclosure and an HMRC assessment. An overseas company that had never registered or filed returns required a coordinated approach to notification, registration, return notices, computations, payment and disclosure.
The group also needed to deal with the consequences elsewhere:
- the UK subsidiary’s transfer-pricing position;
- the relevant UK shareholder’s CFC computation;
- foreign corresponding adjustments;
- residence-state double-tax relief; and
- any applicable treaty or MAP time limits.
This last point was particularly important.
The fact that HMRC can assess an older UK period does not mean that the foreign jurisdiction must still accept a corresponding adjustment or foreign tax credit for that period.
A company can therefore pay historic UK tax and discover that the overseas relief deadline has expired. The UK look-back and the foreign relief analysis must be undertaken together.
I would not calculate every year in full at the outset
The group initially proposed preparing complete PE attribution computations for twelve accounting periods before approaching HMRC.
I did not consider that the best starting point.
First, the historical operating model needed to be divided into meaningful factual phases. The UK activity had not remained constant throughout the period and there was no reason to assume that the PE began, or generated the same level of profit, in every year.
The domestic and treaty PE tests could then be applied to each phase.
Once the likely start date had been identified, the group mapped the assessment position for each accounting period. It considered whether a return had been filed, whether HMRC had received relevant information, whether there had been a failure to notify and what conduct had brought about the loss.
Detailed profit-attribution computations were then prioritised for the periods that were both assessable and material. Sensitivity ranges were used initially for the remaining years.
That approach produced a more reliable disclosure and avoided spending substantial time calculating years that might ultimately fall outside the legal or factual perimeter.
The disclosure had to survive comparison with the existing documents
The group’s historic transfer-pricing documentation described the UK subsidiary as performing routine support services.
Internal board papers suggested that UK executives were involved in important commercial decisions.
The CFC analysis attributed certain significant people functions to the United Kingdom.
Customer correspondence showed UK personnel negotiating terms more extensively than the written authority matrix suggested.
None of those documents was necessarily conclusive. They had been prepared for different purposes and applied different legal tests.
But they could not all be ignored.
The disclosure needed to explain how the functions had changed, in what capacity the UK executives acted and why the PE attribution did not duplicate the remuneration already earned by the UK subsidiary.
It also needed to acknowledge weaknesses in the historic documentation rather than constructing an entirely new version of events.
The strongest disclosure is rarely the one that presents a perfect history. It is the one that explains an imperfect history credibly and supports its conclusions with evidence.
Paying the historic tax did not complete the remediation
HMRC would also want to understand what had changed.
The group therefore reviewed its authority matrix, contracting procedures, intercompany agreements, board terms of reference and the capacity in which UK executives acted.
It introduced annual reviews covering PE, corporate residence and changes in significant people functions. It also established escalation points for changes to senior personnel, approval authority and business models.
Those actions did not alter the historic facts. Nor did they prove that the earlier position had been correct.
They showed that the group understood how the exposure had arisen and had put in place controls intended to prevent it from continuing unnoticed.
A disclosure that calculates the past but leaves the same uncontrolled operating model in place is only half a remediation.
The board needed more than one exposure figure
The audit committee initially asked for one provision.
The analysis produced three more useful figures:
- the gross UK cash exposure, including corporation tax, interest and potential penalties;
- the expected net global tax cost after available treaty, corresponding and CFC relief; and
- the downside exposure if HMRC disputed the PE start date, the conduct analysis or the profit attribution.
The gross UK liability and the ultimate global cost were not the same.
Some UK tax might be relieved overseas. Other amounts, particularly interest and penalties, might remain as permanent costs. Relief could also be delayed, creating material temporary double taxation.
One precise number would have concealed those differences.
The question international groups should ask
When a group identifies a historic UK PE, it should not start by assuming that four, six or twenty years must be corrected.
It should determine when the PE first existed, whether it continued throughout the period and what the relevant people knew at the time.
It should establish whether the overseas company was required to notify HMRC, why it failed to do so and whether the resulting loss was brought about carelessly or deliberately.
It should also determine whether any disclosure would genuinely be unprompted and whether double-tax relief remains available in the other jurisdictions.
The correct question is not:
How many years of computations should we prepare?
It is:
What happened in each period, what should the company have done, what can HMRC still assess and how should the position now be put right?
The group had now identified the PE, attributed its profit, mapped the double-tax relief and determined the historical remediation period.
The audit committee then asked:
If HMRC accepts the settlement, can the group continue operating through the same structure?
That depends on whether the PE is intended, registered and properly managed—or whether the group’s functions, contracts and decision-making need to be redesigned.
That operating-model question will be the next stage of the analysis.
How Vectigalis AC Tax can assist
Vectigalis AC Tax advises UK and international groups on transfer pricing, permanent establishments, CFCs, corporate residence, treaty relief and historical remediation.
Our work includes PE risk reviews, historical fact and conduct investigations, authorised OECD approach profit-attribution modelling, failure-to-notify and penalty analysis, voluntary disclosures, CFC and treaty-relief computations, MAP strategy and the redesign of cross-border operating models.
The objective is not simply to calculate historic tax.
It is to determine the defensible period, quantify the correct profit, preserve available relief, approach HMRC coherently and ensure that the same exposure does not continue into the next accounting period.
Angelo Chirulli, FCA, ADIT, TEP
Dual-qualified UK Chartered Accountant and Italian Dottore Commercialista
Vectigalis AC Tax
www.vectigalistax.co.uk
Mal: angelo@vectigalistax.co.uk
This article is for general information and does not constitute tax, legal or accounting advice. The outcome depends on the specific facts, the applicable treaty, the conduct of the relevant persons and the procedural position in each jurisdiction.