How does the tax authority examine the substance of foreign companies?
On July 3rd of this year, Poland’s Ministry of Finance published a thirty-seven-page document that tax advisers across Warsaw have taken to calling, with grim humor, “the end of an era.” The document—officially titled “Tax Explanations Concerning the Application of the So-Called Beneficial Owner Clause for Withholding Tax Purposes“—reads less like bureaucratic guidance than a field manual for fiscal warfare. It is, in essence, a detailed instruction set for how Polish tax inspectors should interrogate the reality of foreign entities receiving dividends, interest, and licensing fees from Polish companies. And what it reveals about the gap between corporate structure and corporate substance has sent a chill through the thousands of Polish entrepreneurs who, over the past two decades, have built international business empires on foundations that may turn out to be sand.
For years, the formula was simple enough to seem almost elemental: establish a holding company in Cyprus or Luxembourg, obtain a tax residency certificate, meet the formal capital requirements, and enjoy either exemption from withholding tax or preferential rates under double taxation treaties. The structure was legal, the savings substantial, the logic unassailable. A Cypriot company, after all, was a European company—subject to E.U. directives, entitled to treaty benefits, as legitimate as any entity domiciled in Frankfurt or Paris.
That era, the Ministry of Finance now makes clear, is over.
The Three Pillars of Scrutiny
The new guidance establishes three conditions that must be satisfied before a foreign entity can claim tax preferences—conditions that, in practice, disqualify a vast swath of the holding companies that have served as the architecture of cross-border tax planning. The first is straightforward enough: the foreign company must actually receive the payment “for its own benefit.” It’s not sufficient to be the formal recipient; the company must possess what the Ministry calls “economic dominion”—the power to decide how the money will be used and the obligation to bear the economic risk associated with it.
The red flags are specific. Does the company earn only a minimal margin on payments it passes through—a few percentage points, say? Are payments forwarded within days of receipt? Does the company report no taxable income from the payments it ostensibly receives? Is there no reinvestment of funds, just money flowing through like water through a pipe? Does the company lack other income sources that might enable it to meet its obligations? Is there no foreign-exchange risk, no credit risk, no market risk?
If the answer to these questions is yes, the Ministry suggests, then what you have is not a holding company but a conduit—a entity that exists on paper but not in economic reality.
The second condition concerns contractual obligations. Does the foreign company have a duty—formal or informal—to pass the payment along to someone else? The Ministry’s guidance is unsparing here. An obligation to forward payment can arise not just from explicit contractual terms but from the “factual circumstances of the case.” Back-to-back loan structures, in which a company receives a loan on identical terms to one it extends, are a particular object of suspicion. So are situations in which repayment schedules are suspiciously synchronized, or in which a company demonstrably lacks the means to meet its obligations except through the payments it receives from Poland.
Consider a scenario the Ministry offers: A Polish company pays interest to a Luxembourg holding company, which in turn must pay interest on a loan from a Swiss entity that happens to be its shareholder. The terms of both loans are nearly identical—same amount, similar interest rate, correlated payment schedules. The Luxembourg company conducts no other revenue-generating activity. Even if there’s no formal agreement to “pass through” the interest to Switzerland, the Ministry argues, the factual circumstances establish such an obligation. The Luxembourg company cannot be considered the beneficial owner of the interest.
The third condition is the most expansive and, for many companies, the most devastating: the foreign entity must conduct “actual economic activity.” It must possess what the Ministry calls a “personal and material substrate” commensurate with the nature and scale of the payment it receives. This is where the new guidance becomes granular, almost forensic, in its attention to detail—and where thousands of corporate structures begin to crumble under examination.
The Mirage of Substance
What does it mean to have “substance”? The Ministry’s answer is both specific and damning. Start with personnel. Who actually manages the company? The guidance identifies several warning signs of what it terms a lack of “personal substrate.” Nominee directors are a particular concern—individuals who serve as directors purely as a service, often managing dozens of companies simultaneously. The Ministry cites the example of a Cypriot holding company managing assets worth fifty million euros, overseen by two directors provided by a local law firm. Both men serve as directors for thirty other companies. Neither has a financial background or experience in asset management. One is listed as a director of forty-five different entities.
This, the Ministry concludes, is not substance. It is theater.
The guidance introduces what might be called the “time test”: if a company’s board members live and work in Poland, with no documented evidence of time spent in Cyprus conducting company business, then the company cannot be said to possess personal substrate. The question is not whether the directors are nominally Cypriot residents but whether they are actually present in Cyprus, doing the work of management.
Then there is the question of outsourcing. The use of registered-office services, the provision of qualified resident directors by external firms, shared office space with other companies, external handling of mail and administration, consultation with advisers primarily from countries other than the company’s domicile—all of these, the Ministry suggests, are indicators of a company that exists as a legal fiction rather than an economic reality.
The material substrate—the physical and financial infrastructure of the company—is subjected to similar scrutiny. The innovation here is what might be called the adequacy test. It’s not enough for a company to have an office and employees; it must bear the costs of those assets at a level that demonstrates genuine, independent operation. The Ministry offers an example: a company with a fully equipped office in an E.U. capital receives ten million złoty in annual dividends. Decisions about how to deploy those dividends are made by “its” employees, who are technically employed under a global mobility agreement within the corporate group. But the costs of their compensation, charged internally within the group, are significantly lower than the market rate for individuals actually managing assets of comparable value—closer to an average wage in the country than to the compensation one would expect for sophisticated asset management.
The Ministry’s conclusion: regardless of the office or the formal employment arrangements, the failure to bear adequate compensation costs demonstrates that risk has been transferred to other entities. The company is not conducting actual economic activity.
The Shared Substrate Loophole
The guidance does introduce one intriguing concept: “shared substrate.” Under certain circumstances, a company can rely on personnel and infrastructure from another company in the same group without disqualifying itself as a beneficial owner. But the conditions are stringent. The entity providing the substrate must be a resident of the same area—the E.U., in the case of directives; the same country, in the case of tax treaties. The company under examination must still satisfy all other conditions for beneficial-owner status. And, crucially, it cannot be a mere intermediary; it must receive the payment for its own benefit.
There’s a catch, though. The Ministry makes clear that reliance on shared substrate is an option for the payer, not an obligation for the tax authority. In other words, a company might invoke shared substrate when making a payment, but during an audit, tax inspectors are under no obligation to accept that reasoning.
The New Due Diligence
Perhaps the most revolutionary aspect of the guidance is the introduction of differentiated verification standards based on whether the payer and recipient are related parties. For payments to related entities, certificates and declarations are insufficient. The payer must verify contracts governing substrate—office leases, employment or service agreements for key personnel. It must examine financial statements to confirm that costs are actually being borne and assets actually held. It must understand the ownership structure—the chain of relationships that might reveal whether payments are simply being passed through. It must review corporate documents: registry extracts, board structure, the scope of authority. And it must examine payment history to determine whether, in previous years, the company actually exercised economic dominion over the payments it received.
The guidance’s language here is unsparing: “When publicly available documents make it possible to establish that in the years preceding a payment, a company did not exercise economic dominion over dividends it received—for example, by paying out dividends in practically the same amount as those it received—it is necessary to examine the circumstances demonstrating that, with respect to the dividend being paid, the company will exercise economic dominion.”
For payments to unrelated parties, the standard is more forgiving: a tax residency certificate and a declaration of beneficial-owner status generally suffice. But even these must be verified for completeness and authenticity. Does the certificate cover the relevant period? Was it issued by the proper authority? Does it contain all required information? Is the declaration complete and signed by an authorized person?
For so-called technical payers—banks and brokerage houses paying out returns on securities, for instance—the standard is simpler still. They verify only the tax residency certificate, the beneficial-owner declaration, a commercial registry extract, and the most recent audited financial statement, if one exists. The key: the technical payer checks only whether these documents reveal anything contradicting the declarations. It does not actively investigate substance.
Looking Through the Veil
The guidance devotes considerable attention to the concept of “look-through approach”—the possibility of applying tax preferences based on the characteristics of an entity other than the direct recipient of a payment. For dividends covered by the E.U.’s parent-subsidiary directive, the conditions are exacting. Every company in the chain must meet the directive’s requirements: legal form consistent with the statute, E.U. or European Economic Area residency, liability for tax on worldwide income, no exemption from corporate tax in its home country. The capital requirement must be satisfied at each link: the Polish company must hold at least ten per cent of the intermediary’s shares for at least two years; the intermediary must hold at least ten per cent of the ultimate recipient’s shares for at least two years. The payments must be of the same type throughout the chain: dividend to dividend. A change in character—dividend converted to interest, for instance—disqualifies the approach. And there must be no conflict with the directive’s purpose: either the beneficial owner is in the E.U. or E.E.A., or the payment is taxed at least once within the E.U. or E.E.A.
The guidance offers a striking example: a Polish company pays a dividend to a German intermediary, which in turn pays it to a British company (majority shareholder) and a French company (minority shareholder). The Polish company can apply the exemption to the portion of the dividend corresponding to the French company’s stake in the German intermediary. This is partial application of the look-through approach—a rare bit of flexibility in an otherwise restrictive framework.
For interest and licensing fees, the requirements are harsher. The capital threshold rises to twenty-five per cent. And there’s an additional requirement: each country in the chain must effectively tax the payment. If any country exempts the payment from tax—whether under its own implementation of the directive or under a tax treaty—the look-through approach cannot be used.
For payments covered by double-taxation treaties, the standard varies depending on the treaty’s language. Some treaties require direct share ownership; in those cases, the look-through approach is possible, but only at the higher withholding rate. Other treaties permit direct or indirect ownership; for those, the look-through approach is fully available, including reduced rates. Where a treaty is silent on the nature of share ownership, the guidance adopts a taxpayer-friendly interpretation: the look-through approach is permitted.
The Expanded Scope Gambit
The most audacious concept in the guidance is what it calls “expanded subjective scope”—an attempt to solve the problem of special-purpose vehicles created within the E.U. for legitimate business reasons. The idea: even if the direct recipient of a payment (call it Entity B) is not the beneficial owner, tax preferences might still apply based on the status of an entity further up the chain (Entity C)—provided that certain conditions are met.
Step one: verify the structure. Entity B is not the beneficial owner. Entity C is domiciled in the same area as Entity B—the E.U. for directives, the same country for tax treaties. If there were a payment from B to C, it would be of the same type as the payment from Poland to B. And Entity C satisfies the formal requirements for preferential treatment: legal form, capital threshold, holding period.
Step two: examine substance. Entity B is an intermediary because of Entity C’s authority—C can demand that the payment be forwarded. Entity C actually conducts real economic activity. Entity C independently decides how to use payments “at the level of” Entity B. And Entity C bears the costs and risks of managing the payment.
There’s a crucial caveat: this is an option exclusively for the payer or taxpayer. Tax authorities are not obligated to apply this concept. The taxpayer must explicitly indicate in its filing that it is using the expanded subjective scope.
The guidance offers a practical example. A French company (C) wants to finance its Polish subsidiary (A). The bank extending the loan requires security in the form of share collateral. But the Polish company’s bylaws prohibit encumbering its shares. The solution: create a French special-purpose vehicle (B), contribute the Polish company’s shares to it, and use the shares in B as collateral. Funds flow from bank to C to B to A. When the Polish company pays a dividend, it goes to B, which lacks substance and is merely an intermediary. But C has full substance and actually makes decisions about how to use the dividend. C bears risk costs, including the costs of providing security to the bank.
According to the guidance, it may be possible to apply the exemption based on C’s status, even though the payment goes to B. This is elegant in theory. In practice, it’s a minefield—an invitation to disputes over what constitutes genuine business necessity and what constitutes artifice.
The Presumption Trap
The most controversial section of the guidance introduces a presumption that the beneficial-owner condition is satisfied for dividends covered by the parent-subsidiary directive—but only when the dividend is subject to at least one round of taxation within the E.U. If a Polish company pays a dividend to a German company, which in turn pays it to individual shareholders (who will pay withholding tax in Germany), the presumption applies. No need to examine whether the German company is the beneficial owner. Similarly, if a Polish company pays a dividend to a German company, which pays it to another company in a country with which Germany has a tax treaty providing for withholding tax on dividends, the presumption applies.
But the presumption does not apply when the beneficial owner is outside the E.U. or E.E.A., or when the payment is not taxed at all within the E.U.—for instance, because of exemptions in successive countries. Example: a Polish company pays a dividend to a Greek company, which pays it to a Swiss company. Greece has a tax treaty with Switzerland that exempts dividends from withholding tax. The dividend is not “taxed at least once” in the E.U. The presumption cannot be used.
The guidance acknowledges the tension: “The possibility of a payer applying this presumption cannot modify the formal statutory obligations incumbent on the taxpayer.” In other words, even if the payer relies on the presumption, the taxpayer filing for a refund of withholding tax must still submit a declaration of beneficial-owner status. And the Ministry reserves the right to challenge a structure as artificial under anti-avoidance provisions, even when the presumption has been applied.
The End of One World, the Beginning of Another
What the Ministry of Finance has done, with this guidance, is to collapse the distinction between legal form and economic substance—to insist that a company cannot claim the benefits of E.U. law or treaty provisions unless it exists not merely on paper but in fact. For the thousands of Polish businesses that have relied on holding companies in Cyprus, Luxembourg, or Malta, the implications are stark. A structure that was designed to be efficient may now be vulnerable. A strategy that was defensible may now invite challenge.
The guidance is not law. It does not create new obligations; it interprets existing ones. But as any lawyer will tell you, interpretation is everything. And the Ministry has made clear that this is how it will interpret the rules—how its inspectors will conduct audits, how its officials will evaluate claims for preferential treatment. To disagree is to invite a prolonged legal battle, the outcome of which no one can predict.
The Cyprus shell game, it seems, is over. What comes next is anyone’s guess.
