Why remediation must end with a deliberate choice between an intentional UK PE, a genuinely limited UK subsidiary and a real migration of the business

Angelo Chirulli ADIT FCA BFP IFA TEP CPA  |  September 2026

In my previous article, I considered an international group that had identified and quantified a historic UK permanent establishment of its overseas principal. The group then reconstructed how far back HMRC could assess, analysed the conduct behind the failure to notify and coordinated the UK remediation with transfer pricing, CFC and foreign-relief positions.

The working assumption for the next stage was that the historic settlement could be agreed. The audit committee therefore asked a deceptively simple question:

If the PE exposure has been settled, can the group continue operating in the same way?

The answer was not automatically no.

A permanent establishment is not an unlawful structure. A foreign company may deliberately trade through a UK PE, register, file Corporation Tax returns and attribute the appropriate profit to the United Kingdom.

But the answer was not automatically yes either.

A historic settlement does not convert an unmanaged PE into a sustainable operating model. Nor does HMRC’s acceptance of a figure for earlier periods amount to a clearance for later periods, particularly where functions, people, contractual authority or the applicable law have changed.

The group therefore had to make a forward-looking choice. It could retain the commercial model and operate an intentional UK PE. It could genuinely restrict the UK subsidiary to a service-provider role and move the relevant foreign-enterprise functions outside the United Kingdom. Or it could migrate the customer-facing business, risks and residual return into the UK company.

Each route could be technically defensible. Leaving the facts unchanged while rewriting the agreements was not a fourth route.

The historic answer was not the future answer

The historic exercise had reconstructed when the UK PE first arose and how much profit it earned. That conclusion was fact-specific and period-specific.

The future analysis began with a different question:

What operating model does the group actually want, and what tax result follows from that model if it is executed as designed?

That order matters. The board should not begin by deciding that it wants no PE and then ask the business to manufacture evidence supporting that conclusion. It should decide where customers are to be won, where commercial risks are to be controlled, where valuable assets are to be developed and exploited, and which entity should possess the people and capital required to perform those functions.

Only then can the legal agreements, transfer-pricing policy and compliance architecture be built around the commercial answer.

The group also needed to recognise that the PE threshold had changed. For chargeable periods beginning on or after 1 January 2026, the UK domestic dependent-agent test can be met where a person habitually concludes contracts or habitually plays the principal role leading to contracts that the foreign company routinely concludes without material modification. Formal signing authority is no longer the sole focus.

Source: HMRC International Manual – dependent agent PE from 2026. The wording of the applicable treaty must still be checked separately.

That change did not automatically determine the treaty outcome. Many UK treaties use wording that differs from the 2017 OECD Model, and a treaty may restrict the charge permitted by domestic law. HMRC’s own guidance emphasises that the particular treaty provision must be examined.

But it materially changed the risk analysis. A model built around UK executives negotiating every substantive term before an overseas director applied a routine signature was not an obvious no-PE model before 2026. It was even less credible afterwards.

Option one: retain an intentional UK PE

The first option was to keep the commercial model substantially unchanged and accept that the overseas principal would continue trading through a UK PE.

This was not simply the passive continuation of the historic problem. An intentional PE needed a defined functional perimeter, a compliance owner and a repeatable attribution process.

The group would need to identify which UK personnel acted for the UK subsidiary and which acted for the foreign enterprise; maintain contemporaneous records of the assets and risks attributed to the PE; recognise the internal dealings between the PE and the remainder of the enterprise; attribute appropriate capital; and reconcile the PE result with the arm’s-length remuneration of the UK subsidiary.

The distinction between the subsidiary and the PE remained essential. The UK company was a separate legal enterprise. The PE was part of the foreign company. Paying the UK subsidiary a larger cost-plus return did not necessarily exhaust the foreign company’s UK profit if the subsidiary’s people also performed functions for the foreign enterprise that created and formed part of the PE.

Source: HMRC International Manual – separate-enterprise principle and PE attribution. HMRC expressly notes that an arm’s-length agent commission may not represent the whole profit chargeable on the non-resident enterprise.

The finance system therefore needed a PE profit-and-loss account that followed functions rather than invoice location. Customer revenue did not become UK revenue merely because the customer was British. Equally, foreign-booked revenue could not be excluded where the economically relevant activities were performed through the UK PE.

A deliberate PE also required practical compliance. The non-UK company needed the appropriate Corporation Tax registration and filing process. If it had established a physical place of business in the United Kingdom or usually carried on business from a UK location, separate Companies House registration requirements could arise.

Those two concepts should not be conflated. A tax PE can exist without a registrable UK establishment, particularly in a dependent-agent case. Conversely, Companies House registration does not determine the profit attributable for tax purposes.

Sources: HMRC guidance for non-UK companies trading through a DAPE; Companies House guidance on registering an overseas company.

The wider consequences also had to be mapped. VAT, payroll, employment, regulatory and statutory-accounting obligations do not all use the PE definition. The group could not assume that one Corporation Tax registration resolved every UK requirement.

The attraction of this route was commercial honesty. The business could keep the people and authority where they were operationally needed. The cost was an ongoing compliance burden, an annual profit-attribution exercise and greater visibility of the foreign enterprise within the UK tax net.

Option two: remove the PE by changing the business, not the minutes

The second option was to preserve the overseas principal but remove the factual conditions creating the UK PE.

This route was available only if the group was prepared to make genuine operational changes.

The first draft proposal was largely documentary. The intercompany agreement would describe the UK company as providing marketing support. UK executives would lose formal signing authority. All contracts would be signed by an overseas director. Board minutes would record that the foreign company retained every material risk.

That did not address the actual problem.

The historic PE had not arisen because the UK executives possessed the wrong job titles. It had arisen because they led negotiations, determined the commercial parameters, managed important risks and used the UK platform to carry on part of the foreign company’s business.

If those activities continued, an overseas signature could become evidence of routine ratification rather than substantive foreign decision-making. Under the post-2026 domestic DAPE wording, the analysis expressly reaches the principal role leading to contracts routinely concluded without material modification.

A credible no-PE model therefore required the foreign principal to possess real capability outside the United Kingdom. Relevant overseas personnel needed the experience, information and authority to evaluate proposals, determine pricing and contractual parameters, accept or reject risk, and manage that risk after the contract was concluded.

The UK team’s activities also needed a sustainable perimeter. It could undertake genuine market research, technical support, lead generation or other services. It could participate in negotiations without every participation automatically creating a PE. But the detailed facts had to show that its activities did not habitually amount to concluding contracts or playing the principal role caught by the applicable domestic and treaty tests.

The fixed-place analysis had to be addressed separately. A subsidiary does not automatically create a PE of its parent. However, HMRC notes that a parent may have a PE where part of the subsidiary’s premises is at its disposal and the parent carries on its own business there. Removing contracting authority alone would not solve that problem.

Source: HMRC International Manual – PE definition and use of subsidiary premises.

The personnel analysis was equally important. An employee’s legal employer is evidence, but not the conclusion. HMRC’s guidance distinguishes employees carrying on the business of the entity to which they have been seconded from personnel who continue carrying on the business of their formal employer while providing services from another enterprise’s premises.

Source: HMRC International Manual – fixed-place PE and personnel.

The group would therefore need more than revised agreements. It would need revised role descriptions, approval workflows, system permissions, CRM records, customer communications, reporting lines, performance objectives and actual overseas decision-making. If those items did not change together, the new policy would be contradicted by the evidence generated in ordinary business.

Option three: migrate the business to the UK subsidiary

The third option was to stop asking the UK subsidiary to behave like a support company when the business expected it to act like a principal.

Under that route, the relevant customer-facing business would be transferred to the UK company. Customer contracts might be novated. Commercial risks and the resources required to control them could move to the United Kingdom. Rights to exploit relevant intangibles might be licensed or transferred. The UK company would receive the arm’s-length return associated with the functions, assets and risks it actually assumed.

This could simplify the forward UK position. It could also align the legal model more closely with commercial reality.

But it was not a free reset.

Moving functions, assets, risks, contracts or profit potential between group companies can have transfer-pricing and restructuring consequences. The foreign jurisdiction may impose exit charges or regard value as having been transferred. Contract novations may require customer consent. VAT, stamp taxes, payroll, employment law, regulatory permissions and accounting treatment may also need to be addressed.

The board therefore needed a global after-tax and execution analysis, not a comparison of UK headline Corporation Tax rates. A model that reduced PE controversy but created an unrecoverable foreign exit charge or commercially unworkable customer migration could be the wrong answer.

For chargeable periods beginning on or after 1 January 2026, the UK reform package also repealed Diverted Profits Tax and introduced the Unassessed Transfer Pricing Profits charge within Corporation Tax. The forward design therefore had to be tested coherently across transfer pricing, PE attribution and the new charging framework rather than treating each as an isolated workstream.

Source: HMRC policy paper – reform of transfer pricing, permanent establishments and Diverted Profits Tax.

The effective date had to follow execution

Whichever model the board chose, the effective date could not be selected merely for accounting convenience.

A resolution dated 1 January did not eliminate a PE on 1 January if UK executives continued negotiating and controlling risks until March. Equally, a contract novated to the UK subsidiary in March did not necessarily transfer the underlying functions and risk control if the foreign team continued performing them.

The group therefore built an implementation chronology around observable events:

• when decision-making roles and reporting lines changed;
• when system permissions and approval thresholds were updated;
• when relevant contracts, rights or assets were transferred;
• when the foreign or UK decision-makers obtained the capability to perform their new roles;
• when customers and employees were informed; and
• when the old process stopped operating in practice.

The PE analysis was then performed separately for the transition period. There was no assumption that the historic model ended simply because the future model had been approved.

The redesign could not create a residence problem elsewhere

The group also had to preserve the distinction between permanent establishment and corporate residence.

Retaining an intentional UK PE did not require the foreign company to become UK resident. But if the wider redesign resulted in the foreign company’s central management and control being exercised in the United Kingdom, the analysis would move beyond the attribution of a branch profit to the residence of the company itself, subject to the applicable treaty.

Conversely, moving operational decision-making overseas solely to remove the PE was not enough if the overseas board lacked the people, information or practical ability to make those decisions. Substance was not the number of board meetings. It was the capability and conduct behind them.

For a UK-headed group, changes to significant people functions and control over risk also needed to reconcile with the CFC position. A governance change designed to support one conclusion could undermine another if the group described the same activity differently in its PE, transfer-pricing, residence and CFC files.

The monitoring framework mattered as much as the design

The group’s original problem had developed gradually. A three-person support function became a thirty-two-person regional operation without a formal moment at which anyone revisited the PE conclusion.

The redesigned model therefore needed a trigger-based review process. The tax team would not wait for the next enquiry or the next five-year transfer-pricing refresh.

The review would be reopened when, for example, a UK executive obtained broader pricing authority, a foreign decision-maker left without replacement, a new product line was led from the United Kingdom, the UK team began using premises for the foreign enterprise, customer contracts changed, or internal evidence showed that overseas approvals had become routine.

Annual certification could support that framework. It could not replace it. A signed questionnaire stating that authority remained overseas was of limited value if CRM records, emails and customer correspondence showed the opposite.

The most useful control was therefore a live function-and-authority map tied to named roles, systems and material profit streams. The map was reconciled annually with the transfer-pricing policy, PE attribution, CFC analysis and corporate-residence file.

The group chose commercial reality

In this case, the board initially preferred to eliminate the PE. It assumed that the historic issue could be solved by removing UK signature authority and strengthening the overseas approval language.

The operational review showed that this would not be credible. The UK executives were commercially central. Moving their real authority overseas would either disrupt the business or require a substantial new overseas team. The UK subsidiary was also not intended to become the full entrepreneurial principal for the relevant markets.

The group therefore retained the overseas principal and accepted an intentional UK PE.

It registered and operated the PE prospectively, defined the functions undertaken through it, built a shadow accounting perimeter around those functions, documented the dealings with the remainder of the foreign enterprise and reconciled the result with the UK subsidiary’s remuneration. It also clarified responsibilities that genuinely belonged to the subsidiary so that the PE perimeter did not expand by default.

That answer produced more ongoing UK compliance than the board had originally wanted.

It produced less risk than a paper-only redesign.

The next question for international groups

When a historic PE has been identified and remediated, the next step is not to declare the structure repaired.

The group should decide whether the PE is commercially intended. If it is, the PE should be registered, attributed and governed as part of the forward tax model. If it is not, the functions, authority, premises and evidence that created it must genuinely change. If the UK business already performs entrepreneurial functions, the group should consider whether the relevant business and return belong in the UK subsidiary instead.

The correct question is not:

How do we draft the agreements so that the PE disappears?

It is:

Where will the business actually be carried on, who will control the relevant risks, and which entity or PE should earn the resulting profit?

A historic settlement closes a period.

It does not design the future.

That requires an operating-model decision – and evidence that the business implemented it.

How Vectigalis AC Tax can assist

Vectigalis AC Tax advises UK and international groups on transfer pricing, permanent establishments, CFCs, corporate residence, treaty relief, historical remediation and cross-border operating-model redesign.

Our work includes PE risk reviews, post-disclosure operating-model analysis, dependent-agent and fixed-place assessments, authorised OECD approach profit-attribution modelling, functional and authority mapping, reconciliation of subsidiary and PE remuneration, CFC and residence reviews, implementation support and the design of ongoing tax-control frameworks.

The objective is not to remove a PE at any cost. It is to ensure that the legal structure, commercial operations, allocation of profit and contemporaneous evidence describe the same business.

Angelo Chirulli ADIT FCA BFP IFA TEP CPA

Vectigalis AC Tax

Mail: 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 jurisdictions involved and the way in which the operating model is implemented in practice.