A self-researched cross-border structure built on real Cyprus statutory provisions, every condition technically met. Why the firm declined the engagement, and what the case shows about cross-border planning in 2026.
A prospective client came to us with a plan. Two individuals, both nationals of a high-tax EU country, intending to relocate to Cyprus, acquire residency under the 60-day rule, form a Cyprus company in which they would be the only shareholders, take a notional Cyprus salary to anchor the residency tie, then travel to a third country and work for ninety days plus for a foreign-incorporated company they own and direct. The salary from that foreign company would be substantial, in the order of seven figures. They proposed to claim the Cyprus statutory exemption for employment income earned abroad, rendering that salary fully exempt from Cyprus income tax. Return to Cyprus before year-end, complete the 60-day count, file the return.
They were not naive. They had read the law carefully. They came to us not because they were unsure of the technical position, but because the inflow was large and they wanted the implementation done by professionals who could stand behind it. The plan was sophisticated. The plan was specific. The plan was, on its face, exactly what the statutes appeared to permit.
On a checklist reading, every condition could be met. The 60-day rule conditions are clear, and the architecture proposed satisfied them. The foreign-employment exemption is a real Cyprus statutory provision, used by Cyprus residents in genuine cross-border employment situations every year. The clients had drafted around the conditions correctly.
We declined. This article is about why.
The substance question
The Cyprus exemption for foreign employment income is a deliberate statutory relief used for its declared purpose. The Cyprus general anti-abuse rule cannot override it. That much is settled. The vulnerable point lies one layer deeper, on the question of whether the employment is real.
When the recipients own the foreign payer, direct it as members of its board, set their own contracts, and approve their own compensation, the assessor's question is not whether the boxes are ticked. The assessor's question is whether this is employment in any meaningful sense, or whether it is shareholder extraction routed through a self-designed payroll arrangement.
The defence is documentary, but it is documentary at a level the structure had not been designed to produce. A real employment contract is not a template. It is bespoke to the role, signed before the work begins, with deliverables, reporting lines, and a compensation methodology that an external benchmark could test. Real services are independently describable. Real board discipline is minute, by minute, by minute, with substantive deliberation, with dissent where dissent should appear, with formal approval of the compensation paid by the employer to the employees who happen to be its directors. Work product can be examined: contracts negotiated, documents reviewed, decisions taken. The clients had the role and the title. They had not built the file.
The corporate residency question runs alongside it. A foreign-incorporated company owned and directed by two Cyprus-resident individuals invites the central management and control test. The test is long-established and its application is alive and well. Where the directing minds are physically resident in Cyprus and direct the company from Cyprus, even where formal board meetings happen elsewhere, the company can be pulled into the Cyprus tax net as Cyprus tax resident.
If the company becomes Cyprus tax resident, two things follow. Its worldwide profits become taxable at the Cyprus corporate rate of 15% from 1 January 2026. And the salary it pays no longer satisfies the Cyprus exemption, because the exemption requires either a non-Cyprus tax resident employer, or a foreign permanent establishment of a Cyprus tax resident employer. The structure does not just lose; it collapses on itself. The clients had not stress-tested the plan against the consequence of the company being treated as Cyprus tax resident, because they had not considered it likely. We considered it likely.
What sits around the structure
Tax residency does not end on departure. It ends when the home-country revenue accepts the change of position. Until that acceptance is on paper, the home country continues to claim worldwide taxation. Sequencing is therefore not a logistical question. The clients had assumed that landing in Cyprus, taking the Cyprus salary, and obtaining the residency certificate were the same step. They are not. Complete the home-country exit, obtain the Cyprus tax residency certificate, then receive the inflow. Receiving a seven-figure salary from a foreign employer while still classified as a tax resident of the home country sets up a position that is awkward to defend, and impossible to defend retrospectively.
The disclosure layer sits over the architecture. The EU mandatory disclosure regime operates on hallmarks, several of which concern arrangements that convert income into a more favourably taxed category, subject to the main-benefit test. The defensible position is that pure use of a domestic statutory exemption is not 'conversion' within the meaning of the hallmark; the exemption was put there by the legislature with a purpose. But the question is live, intermediaries face their own exposure if they do not report, and defensive reporting puts the structure on the desk of every revenue authority in the network. Including the home country. Where no intermediary is involved at all, the obligation falls back on the taxpayer himself, who frequently does not know that.
The third-country question is the one most often underestimated. The ninety days plus have to be performed somewhere. Most jurisdictions impose either local income tax on the work performed there, or local social security on the employer-employee relationship, or both. Treaty short-stay exemptions in the OECD-model dependent personal services article protect against income tax for stays under 183 days, subject to conditions, but they do not protect against social security.
The social security position is the loud part. For a 90-day-plus stay performing core remunerative work for a foreign employer, most jurisdictions impose local social security or national insurance on the employer-employee relationship. EU social security coordination rules provide for posted-worker exemptions, but those are designed for genuine secondment arrangements, not for self-arranged stays where the employer and the employee are essentially the same individuals.
Some jurisdictions impose neither, and these narrow the candidate set. Even those, however, raise overlays of their own: reputational, banking-AML, and the live question of whether having two senior employees perform core income-generating work in a country for ninety days plus creates a local permanent establishment for the foreign employer there. The choice of country is not a logistics question. It is a binding tax question, and it narrows what was already a narrow universe.
Banks read these patterns. A seven-figure inflow to a recently arrived account holder, from a foreign company the holder owns, in the holder's first year of residency, will trigger enhanced due diligence. Banks have become noticeably more cautious about exactly that profile of inflow than they were five years ago. The funds may sit in suspense for weeks, may be returned, or may be released conditionally on documentation that the structure was not designed to produce.
Composition, and the trade today
Each of these risks is, individually, manageable. A real employment can be designed and documented. The corporate residency question can be addressed, perhaps by accepting Cyprus residency on the company and accepting the consequence. The exit can be sequenced. The disclosure question can be considered with positions taken. The country can be selected with care. The bank can be primed. Each is a problem with a known answer.
The work of advising starts with the unknown. When the file presents one, you study it, you weigh it, and you take a position on whether the risk is acceptable. That is what advice does. When the file presents several unknowns of different kinds, sitting alongside each other, with others almost certainly hiding behind them, the exercise stops being one of evaluation and becomes one of disposition. The arrangement may still look attractive on paper. On the file taken as a whole, the better judgment is to stay away from it. In 2026, the test is no longer only the letter of the law. It is the spirit of the law too.
The exemption is real. The architecture is legal. That is not the same as it being advice.
Two propositions, stated plainly. The structure is legal. Every component is grounded in Cyprus statute. Every exemption is real. Every box can be ticked. Anyone who says otherwise is wrong on the technical law.
The structure is also not advice. The distance between 'legal' and 'advisable' is the entire content of professional judgment. A structure that ticks every box on paper while inviting scrutiny on substance, directorship, exit, disclosure, country selection, banking and documentation simultaneously is not a tax plan. It is a portfolio of correlated risks with a single payoff.
The trade in cross-border tax planning is no longer between aggressive and conservative structures. It is between structures that can be defended cheaply and structures that cannot be defended at all. The market has moved. Anti-abuse rules, mandatory disclosure, exchange of information, and the assessor's increasing willingness to look through form to substance have shifted the equilibrium. What looked clever five years ago looks expensive today.
A good advisor's job is to know the difference. Sometimes that means saying no.