How Tax Authorities Prove Where You Actually Live
← Back to blog

How Tax Authorities Prove Where You Actually Live


Short answer: Tax residence disputes are won and lost on evidence, not on paperwork. Where an authority has reason to look — and Spain is the clearest documented example — it reconstructs a calendar of physical presence from ordinary traces: card transactions, utility consumption, invoices from service providers, forms you signed yourself. Under published Spanish doctrine it can also count the days between two proven presences. But the same standard cuts both ways: in April 2026 a Spanish court annulled an assessment of over EUR 55 million because the authority had built its case on inference rather than proof.

What decides it Practical weight
Physical presence, day by day Reconstructed from transaction and consumption records, not from passport stamps alone
Days between two proven presences In Spain, treated as presumed domestic days under published doctrine unless presence elsewhere is proven. Not a universal rule — and the mechanism that catches people
A home kept available Relevant under treaty rules; a genuine arm's-length letting changes the analysis
Foreign residency certificate Rebuttable presumption, not a conclusion
Who carries the burden The authority must establish its case under domestic law first; once it does, or once a statutory presumption applies, the taxpayer needs contemporaneous evidence to displace it
Documents you signed Frequently the strongest single item against you

What tax residence is — the difference between an immigration permission and a tax status, and which criteria apply — we have covered separately in our pieces on residency versus immigration status and the residency trap. This article is about something narrower and much less discussed: the machinery of proof.

What did two Spanish judgments six days apart actually decide?

In April 2026 two Spanish courts ruled on tax residence within a week of each other, applying the same statutory test, and reached opposite conclusions. Read together, they are the clearest available answer to the question of how these disputes are decided.

On 9 April, the Criminal Chamber of the Spanish Supreme Court upheld convictions for tax offences against a businessman who had spent his career in Venezuela and held Venezuelan residency certificates. The courts had established presence in Spain of at least 223, 252, 262 and 221 days across four consecutive years. The certificates did not save him.

On 15 April, the Spanish National Court annulled income and wealth tax assessments and penalties totalling over EUR 55 million against a performing artist for a single tax year. The authority itself had put her presence at 163 days. It then attempted to add periods spent abroad as "sporadic absences" to cross the statutory threshold, on the basis that she was in a relationship with a Spanish resident. The court rejected that construction: a relationship is not a marriage in law, and the authority had not shown that her economic base or family nucleus was in Spain.

The second judgment is under appeal. The tax authority applied to the Supreme Court for review in July 2026, so it is not settled doctrine.

The decisive passage in the second case is procedural rather than factual. The authority had spent its analysis arguing that the taxpayer did not really live in the jurisdiction whose certificate she held. The court redirected it: what has to be proved is residence in Spain, not the absence of residence elsewhere. That distinction decides more cases than any list of criteria.

What evidence actually builds a day count?

The criminal case is unusually explicit about method, which makes it the most useful public document on this question. The presence figures were not derived from border records. They were assembled from ordinary life.

Evidence category What it produced
Card use and cash withdrawals Activity in the country on 160, 130, 135 and 145 days across four years, from six cards on one account
Gaps between card transactions A further 47, 48, 71 and 53 days with no card use anywhere, immediately following domestic use — treated as presence
Utility consumption Water and electricity at rented properties, at levels consistent with continuous family occupation
Telephone accounts Three mobile lines and four landlines, with annual billing quantified
Medical and dental services Treatment received locally over the period
Physiotherapy invoices Individual invoices with specific dates — seven in one year, eight in the next
Golf club billing Invoices across all four years
Property A house owned since 1973 and renovated during the period, plus three separate leases
Courier records Parcels sent from the country to overseas destinations
Self-signed documents A US withholding form on which he entered a domestic address as his permanent residence, and handwritten address details on Swiss and offshore bank documentation

Two features of that list deserve attention. First, none of it is exotic surveillance. It is bank data, utility bills and invoices, obtained through ordinary inspection powers. Second, the reasoning about the gaps is the part that most people never anticipate: once presence is documented on two dates, the days between them can be treated as domestic days unless presence elsewhere is proven. That is a formal doctrine, set out by the Spanish central tax tribunal in two 2023 resolutions and reproduced in the tax agency's own published manual, which distinguishes certified presence, presumed days, and sporadic absences.

The mechanics of the day count are borrowed from an unlikely place. The tax agency's manual, in support of the rule that any day counts in full regardless of hours, cites the OECD Commentary on Article 15 — the article governing the taxation of employment income. That Commentary sets out the "days of physical presence" method: part of a day, day of arrival and day of departure all count, and a day on which the taxpayer was present at all, however briefly, counts as a day of presence. The rule was written for a 183-day exemption on employment income. It now shapes how residence itself is measured.

Why doesn't a residency certificate end the argument?

Because it proves a legal characterisation in one country, not a set of facts about another.

In the criminal case, the Spanish courts did not build the conviction on rejecting the Venezuelan certificates. They built it on proven presence, and only then addressed the certificates as defence evidence. The certificates stated that under domestic Venezuelan law the taxpayer was a resident there; they contained nothing establishing where he had actually been. The appellate court's observation is almost tautological, and all the more damaging for it: if presence in Spain exceeding the threshold is established for each year, he cannot simultaneously have been present in Venezuela for a comparable period.

The taxpayer relied on the Spanish Supreme Court's 2023 administrative doctrine that a certificate issued for treaty purposes must be presumed valid and cannot be brushed aside on the basis of indications. The Criminal Chamber accepted that doctrine and then applied it exactly as written: the presumption is rebuttable, and it was rebutted. Obtaining the certificates was, in the courts' view, part of the arrangement itself.

In the civil case decided six days later, the same instrument fared differently. A certificate covering residence since 2007 was treated as an evidential indication — helpful, not decisive. It did not need to be decisive, because the authority had failed at the prior stage.

If you hold a UAE tax residency certificate, the practical reading is uncomfortable but simple. The certificate is worth having. It is not a defence. What it actually achieves depends on the certificate itself, the treaty relationship between the two states, the domestic law of each, and the facts of the particular tax year.

Does the treaty tie-breaker rescue a day count?

This is where relocation marketing and the actual text diverge most sharply.

Where two states both claim a person under their domestic law and a treaty applies, Article 4(2) of the OECD Model sets a cascade: permanent home available, then centre of vital interests, then habitual abode, then nationality, then agreement between the authorities.

The Commentary on the permanent home limb is more demanding than it is usually reported to be. A home counts as permanent where the individual has arranged to have it available continuously at all times, rather than occasionally for a short stay. And a property owned by the individual is not treated as available during a period in which it has been let. That single sentence is the reason advisers tell clients to let the old home on genuine arm's-length terms rather than leave it furnished for visits — it is not folk wisdom, it is the Commentary.

At the centre of vital interests stage, the Commentary directs that the circumstances be examined as a whole, while noting that considerations based on the individual's own personal acts deserve particular attention. That is worth holding next to the withholding form in the criminal case, completed by the taxpayer in his own hand.

The habitual abode limb contains the passage that most directly contradicts the "183 days and you are free" pitch. The Commentary, as replaced in November 2017, states that the test is not satisfied by simply determining in which of the two states the individual spent more days. What matters is the frequency, duration and regularity of stays forming, in the Commentary's phrase, "the settled routine of an individual's life". The relevant period must be long enough to establish that pattern.

So the day count is decisive at the domestic level and explicitly insufficient at the treaty level. Anyone selling a day-counting strategy is selling half a test.

What do cross-border reporting systems actually see?

Partly what the marketing claims, and mostly not. Precision matters here, because this is where confident assertions are cheapest to make.

The EU does operate a central payments database. CESOP has been live since 1 January 2024 under a Council directive and regulation adopted in February 2020. Payment service providers report cross-border payments to national authorities, which forward them to the central system, where the data is cross-checked and made available to anti-fraud specialists.

But the design is narrower than the headline. The reporting obligation is triggered by the payee receiving more than 25 cross-border payments in a calendar quarter, and the purpose is VAT fraud in cross-border commerce. It is not a live feed on individual consumer spending, and it is not a residence-monitoring tool.

That does not make the underlying risk imaginary — as the Spanish criminal case demonstrates, card data can be obtained and reconstructed into a calendar. But it is obtained through inspection powers directed at a specific person, not harvested automatically from a European database. The distinction determines what you can plan around.

The Common Reporting Standard is the other system routinely misdescribed. Under the OECD standard, banks report account balances, income and holder details to their local authority, which passes them to the authority of the jurisdiction where the holder is tax resident. The UAE participates. What this produces is visibility of accounts, not visibility of movements: it tells a foreign authority that an account exists and what is in it, not where you slept last Tuesday.

The part almost nobody discusses is how the system decides which authority receives the file. It relies on you. For a new individual account, the standard requires the bank to obtain a self-certification at opening — a signed statement of your name, residence address, jurisdictions of tax residence, tax identification number and date of birth — and to test it for reasonableness against the documentation gathered in the onboarding process. You must declare every jurisdiction in which you are tax resident, not merely the convenient one.

That signed form is a document you created, and it belongs in the same evidential category as the withholding form that helped convict the taxpayer in the Spanish criminal case. If you certify a low-tax residence to a bank and your former home country later establishes on the facts that residence never moved, the certification does not merely fail to protect you. It may become a highly inconvenient contemporaneous record of the position you took. CRS is worth understanding less as a surveillance channel than as a place where you put your own version of the facts in writing, years before anyone challenges it.

How does the UK handle the same question?

Differently in structure, identically in substance — and the two most recent UK decisions make that unusually clear.

The UK once ran on the same fact-based logic as Spain. In 2011 the Supreme Court, deciding a judicial review of how the tax authority's guidance booklet should be construed, held that the ordinary law required a person to make a distinct break in the pattern of their life in the UK, and that establishing one demanded a multifactorial inquiry rather than a calculation. A government committee had called the resulting uncertainty intolerable as far back as 1936. Two years after that judgment, Parliament finally legislated.

Since April 2013 the UK has applied a Statutory Residence Test set out in Schedule 45 to the Finance Act 2013. Automatic tests are applied in sequence; if they do not resolve the year, residence turns on defined ties — family, accommodation, work, a 90-day look-back, and for recent leavers a country tie — measured against days spent in the UK. The more ties, the fewer days permitted. Presence is measured at midnight, and specific days can be disregarded: transit days, and days where exceptional circumstances beyond the person's control prevented departure.

The design goal was certainty. What actually happened is that the argument moved from "where is your life centred" to "what happened on the fourth Tuesday in February" — and got no easier to prove.

Six days, £3.1 million, and no records

In February 2025 the Court of Appeal decided what it was told was the first case in which the Statutory Residence Test had come before the courts at all.

The taxpayer moved to Ireland on 4 April 2015, immediately before the start of the tax year in which she received roughly £8 million in dividends on shares her husband had transferred to her. The judgment records the obvious inference about timing, and then sets it aside: the test applies identically whatever the motivation for moving. That point is worth pausing on, because relocation commentary tends to assume that visible tax motivation is itself fatal. In a rules-based system it is not.

She had three UK ties — family, accommodation and the 90-day look-back — which under the statutory table set her limit at 45 days. She was present at midnight on 50. The entire dispute, worth £3,142,550.58 in tax, therefore came down to six days: two in December 2015 and four in February 2016, when she was in England dealing with a twin sister in crisis from alcoholism and that sister's two young children.

She had taken professional advice before moving. She knew the day limit, knew the exceptional-circumstances exemption existed, and knew she needed to record where she was each day. She did not do it. The tribunal found her defensive and vague under cross-examination. She produced no day-by-day account, no phone records despite having itemised billing, no retained messages. Her credit card statements — which were in evidence — showed a restaurant meal within two hours of landing on the day she says she arrived to a crisis, £239 at an optician the same day, and a £400 cash withdrawal at a children's hospital she could not remember visiting. She described the February week as a blur.

She won anyway, but the route matters. The First-tier Tribunal found for her; the Upper Tribunal reversed and re-made the decision against her; the Court of Appeal restored the first decision. It did so largely on the limits of appellate review: whether circumstances are "exceptional" is a question of fact, and an appeal court can only disturb it where no tribunal could reasonably have reached that conclusion. The Court of Appeal also rejected the narrower reading below, holding that a sufficiently compelling moral obligation can form part of exceptional circumstances and can prevent departure, provided it is objectively compelling rather than idiosyncratic — the court's illustration being that a taxpayer who stays for a pet tortoise or a cup final is not prevented from leaving.

The outcome is often reported simply as a win for the taxpayer. The judgment reads differently: she won despite the evidential gaps, on the narrow ground that an appeal court cannot substitute its own view of the facts. Read as guidance rather than as a result, the case says something uncomfortable. She was two appellate stages and one judicial value-judgement away from a £3.1 million liability, on facts she could have documented in a notebook and did not.

Four days, £65,000, and a complete file

The contrast arrived in May 2026, when the First-tier Tribunal decided the first case to consider the transit exception.

A chartered engineer working in Iraq on rotation was present in the UK at midnight on 100 days in 2019/20. Seven were conceded as pandemic-related. He needed four more disregarded to come in under 91 days and pass the third automatic overseas test. The authority put him at 93 and assessed £64,945.65.

Three of the four days were airport stopovers between Iraq, Naples, Tokyo and Dublin. The authority's argument was that because he had bought separate tickets rather than a single through-ticket, he had not arrived "as a passenger" — the same journey, differently ticketed, would have qualified. The tribunal called that distinction arbitrary and unsupported by the legislation. It found he had done nothing in the UK beyond sleeping, eating at the hotel or terminal, and travelling between them.

The fourth day was 29 February 2020. He had boarded a flight to Dublin that was cancelled on the tarmac after Dublin Airport closed during a named storm. He submitted airport data: no aircraft could land at Dublin between 14:20 and 15:53, twenty flights were scheduled to land in that window, at least two cancellations, two diversions and twenty-three delays. He also ran a six-year analysis of cancellation rates across twenty-seven UK airports to show that cancellation is uncommon rather than routine. The tribunal accepted that the day was exceptional, and rejected the argument that he should have chased speculative alternative travel or abandoned checked luggage to prove he intended to leave promptly.

He won on both points and left the UK count at 89. What he had that the earlier taxpayer did not was boarding passes, cancellation notices, hotel invoices, bank statements and credit card records — assembled during an enquiry that ran from December 2020 to a closure notice in October 2022, and a hearing bundle of 897 pages.

What the pair actually demonstrates

Ireland case (2025) Iraq case (2026)
Amount in dispute £3,142,550.58 £64,945.65
Days in issue 6 4
Documentation Advised to keep records; none kept Boarding passes, invoices, statements, airport data
Path Won, lost, won — three instances over roughly three years Won at first instance
What decided it Appellate restraint over a tribunal's value-judgement The file

Two systems, one lesson. Spain reconstructs a hundred days from card data because it has to; the UK argues over four days because the statute says so. In both, the person who can evidence their own movements wins, and the person who cannot is relying on someone else's benefit of the doubt.

Facts-based model (Spain) Rules-based model (UK, post-2013)
Core test Days, economic base, or family presumption — any one suffices Sequenced automatic tests, then ties measured against a day table
Day counting Reconstructed, including presumed days between proven presences Statutory; midnight presence, with defined exceptions
Predictability Low; heavily fact-sensitive Higher; the thresholds are published
Where it bites Proof, inference and burden Ties you did not think counted, and single days you cannot account for
Motivation for moving Relevant to how the file is read Expressly irrelevant to the test

Fact versus interpretation

Claim Assessment
More than 183 days abroad ends residence in the home country False. Domestic rules commonly reach further, and treaty tie-breakers apply only where both states claim you
A residency certificate overrides a competing claim False. Rebuttable presumption; displaced by evidence of actual presence
Days between two documented presences count against you Correct in Spain, as published administrative doctrine. Not a general European rule — check the position in the specific jurisdiction
The EU automatically monitors your card spending for residency purposes False. CESOP is payee-focused and aimed at VAT fraud
The authority must prove residence in its own country before adding absences So held by the National Court in April 2026. That judgment is under appeal and is not final — treat it as a strong argument, not settled doctrine
Visible tax motivation for moving undermines the position Not under a rules-based test. The UK Court of Appeal applied the statutory test identically regardless of why the taxpayer moved
A rules-based system removes the evidential problem False. It shrinks the argument to individual days, which still have to be proved
CRS tells a foreign authority where you physically live False. It reports account data, routed on the basis of the tax residence you yourself certified
A property left available but unoccupied is neutral False. Availability, not use, is what the OECD Commentary tests

What does any of this mean for a move to the UAE?

Everything above is European. That is where the documented case law is, and it is also where the risk sits: a UAE relocation is challenged by the country you left, under its rules, using its evidence-gathering powers. The Emirates side of the arrangement is rarely the contested part.

Translated into the decisions a European entrepreneur actually makes on the way out, the four judgments say this.

The residency certificate is documentation, not protection. A certificate issued by the UAE Federal Tax Authority establishes UAE residence for treaty purposes. In both Spanish cases the foreign certificate was treated as a rebuttable indication and nothing more — decisive in neither direction. Obtain it, keep it current, and do not build the plan around it.

Your ordinary spending is the day count. The Spanish criminal case reconstructed four years of presence largely from card use, gaps between transactions, utility consumption and dated service invoices. That machinery works identically in reverse. A relocation where daily card use, utilities, medical care and professional services are concentrated in the Emirates produces the record; one where they remain concentrated in the country of departure produces the opposite record, and nobody has to fabricate anything.

The property you keep is tested on availability, not use. Under the OECD Commentary a home is not treated as available to you during a period in which it has been let. A property retained empty for visits is a different fact from a property let on arm's-length terms. This is one of the few points where a decision taken at the outset changes the analysis permanently.

The forms you sign travel with you. Bank onboarding, CRS self-certification, withholding forms, registry filings. In the criminal case a withholding form completed in the taxpayer's own hand became part of the finding against him. Every one of these should state the same thing, and that thing should be true.

Keep a contemporaneous record from day one. The UK cases make this concrete: one taxpayer had been advised to log her days and did not, and came within one judicial value-judgement of a GBP 3.1 million liability; the other produced boarding passes, cancellation notices, hotel invoices and bank statements, and won on four disputed days. Reconstructing a calendar years later from memory is the position nobody wants to be in.

None of this is a substitute for advice on your own facts, and none of it works retrospectively. That is the point.

What this means if you are relocating

The practical conclusion is not a document checklist. It is that the record has to match the story, over time, without curation.

Every category that convicted the taxpayer in the criminal case would, on different facts, have supported him. Card use concentrated in the new country, utilities consumed where the family lives, medical and professional services engaged locally, addresses given consistently on every form — the same evidence, pointing the other way. A relocation that has genuinely happened generates that record without effort. One that exists on paper generates the opposite, and cannot be repaired retrospectively.

This is the same underlying question as the zero-tax proposition and as exit tax and controlled foreign company exposure: none of them turns on the structure being permissible in the abstract. They turn on whether the facts support it. Where a company is concerned, the same logic runs through the question of substance.

In short: Tax residence is a question of proof. Authorities reconstruct presence from card data, utilities, invoices and forms you signed, then count the days in between. But they have to prove residence in their own country first — and in April 2026 a Spanish court annulled EUR 55 million because one did not. The strongest position is not a well-argued file. It is an ordinary life that happens where you say it does.


Frequently Asked Questions

Can a tax authority really reconstruct my calendar from card transactions? Yes, and it has been done. In a Spanish Supreme Court case decided in April 2026, presence was established partly from card use and cash withdrawals on 160, 130, 135 and 145 days across four years, with further days inferred from the gaps between them.

Does a Tax Residency Certificate settle the question? No. A certificate issued for treaty purposes carries a rebuttable presumption of validity. It can be displaced by evidence of actual presence, and it does not by itself prove where a person lived.

Who has to prove what? The authority must first establish residence under its own domestic law. Once it does, the burden shifts to the taxpayer. A Spanish court annulled a EUR 55 million assessment in April 2026 precisely because the authority had not cleared that first hurdle.

Does the EU track my card spending across borders? Not in the way relocation marketing implies. The CESOP system, live since January 2024, collects data on payees receiving more than 25 cross-border payments per quarter, for VAT fraud purposes. It is not a location feed on the payer.

Can a UAE Tax Residency Certificate protect me from a tax audit in Europe? It does not prevent an audit and it does not decide the outcome. A certificate issued for treaty purposes carries a rebuttable presumption of validity, but a European authority can still assess residence under its own domestic law and displace the certificate with evidence of actual presence. What the certificate achieves depends on the treaty, both domestic laws and the facts of the year.

Is the CRS self-certification I signed at my bank a risk? It can be. The standard requires banks to obtain a signed statement of every jurisdiction in which you are tax resident. If a tax authority later establishes on the facts that residence never moved, that signed statement becomes evidence you produced yourself.

What single piece of evidence causes the most damage? Something you signed yourself. In the Spanish criminal case, the taxpayer had entered a home-country address in his own hand on a US withholding form and on bank documentation.

What records should I keep after relocating to the UAE? Enough to reconstruct where you were, without relying on memory: travel documents and boarding passes, a contemporaneous day log, evidence that daily spending and utility consumption happen where you now live, and consistent address details on every form you sign. A UK tribunal decided a 2026 case on exactly that kind of file.


Sources

Everything above is drawn from the judgments and official texts themselves, not from summaries or professional commentary on them — which, on more than one of these cases, describes the outcome differently. Spanish decisions are identified primarily by ECLI, which is designed for durable citation across Europe; United Kingdom judgments use neutral citation, which serves the same function. Links are a convenience; the identifiers are the source.

Official sources accessed on 24 August 2026.


This article is for educational purposes only and should not be treated as legal or tax advice. Every situation requires individual analysis.

Want to discuss your situation? The first conversation is complimentary.

Book a consultation →