Reading the risk in treaty access between Hong Kong and the CIS

A group with operations across the Commonwealth of Independent States and a Hong Kong holding or treasury entity faces a question that sits at the intersection of two very different tax traditions. The CIS jurisdictions – Russia, Kazakhstan, Azerbaijan, Uzbekistan, and their neighbours – have treaty networks built around residence certificates, beneficial-ownership declarations, and periodic administrative confirmation. Hong Kong operates a territorial system with no outbound withholding tax on dividends and no capital gains tax. The mismatch between those two logics is where the risk lives.

Treaty access between Hong Kong and the CIS turns on whether a Hong Kong entity can demonstrate genuine tax residence and sufficient substance to satisfy the anti-avoidance provisions built into each bilateral agreement. The governing framework is the Inland Revenue Ordinance on the Hong Kong side, read together with the relevant double-taxation agreement between Hong Kong and the CIS state in question. Where a treaty exists, access to reduced withholding rates depends on beneficial-ownership tests, substance conditions, and – increasingly – the principal-purpose test imported into CIS domestic law from the OECD's base-erosion and profit-shifting project. Where no treaty exists, the analysis shifts to unilateral relief and structural alternatives. The cases where these decisions have been made poorly are accumulating on our desk.

This analysis covers the commercial stakes, the governing instruments and how they operate across the Hong Kong–CIS interface, the comparative risk read across resident and non-resident positions, and our view on where the exposure is concentrated in the current environment.

What is commercially at stake for cross-border groups?

The commercial question is simple: how much of a dividend, interest, or royalty stream paid from a CIS operating entity reaches the group treasury or ultimate shareholder net of withholding tax? The answer determines whether a Hong Kong holding or treasury entity is commercially meaningful or merely adds a layer of cost and complexity.

Many CIS states impose withholding tax on dividends at rates between five and fifteen per cent under their domestic law. Treaty rates can reduce that materially – in some cases to five per cent on qualifying dividend payments. On a substantial dividend stream, that differential is a material recurring cash-flow item. It is not a one-time structuring benefit. It recurs every time profits are distributed, every time interest is paid on an intercompany loan, every time a royalty flows up from a CIS operating entity to a Hong Kong intellectual-property holding vehicle.

The risk, then, is not only that a structure fails at the point of implementation. It is that a structure which worked for several years is challenged retrospectively, triggering unpaid withholding tax, interest, and penalties across multiple distribution cycles. In our cross-border practice, that is precisely the pattern we see: a holding structure that was implemented without a clear substance analysis, tested years later by a CIS revenue authority with enhanced treaty-denial powers.

Kazakhstan and Russia, in particular, have both adopted domestic anti-avoidance provisions that allow their tax authorities to deny treaty benefits where the entity claiming residence lacks genuine economic content. Uzbekistan is following the same trajectory as its domestic rules modernise. The question a group GC needs to answer before the next dividend cycle is not "do we have a treaty?" but "can we demonstrate we qualify for it?"

What are the governing instruments and how does the cross-border interface operate?

Hong Kong has concluded double-taxation agreements – each formally titled as an Agreement for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion – with a number of CIS jurisdictions, including Russia, Kazakhstan, and others. Each agreement allocates taxing rights between Hong Kong and the counterpart state, sets reduced withholding rates on passive income, and includes a beneficial-ownership condition on those reduced rates. The Hong Kong side is administered by the Inland Revenue Department under the Inland Revenue Ordinance.

On the CIS side, the administering authorities interpret the treaty through their own domestic lens. That domestic lens has changed significantly over the past decade. Russia introduced a controlled-foreign-corporation regime and a de-offshorisation policy that altered the economic logic of many holding structures using Hong Kong and offshore centres. Kazakhstan updated its tax code to incorporate a principal-purpose test – a provision that allows treaty benefits to be denied where one of the principal purposes of an arrangement was to obtain those benefits, without adequate commercial justification. Uzbekistan's treaty practice is less settled, but the direction of reform is consistent with its neighbours.

The cross-border interface bites at two points. First, at the time of the withholding: the CIS payer is required to verify that the recipient qualifies as a beneficial owner resident in Hong Kong before applying the treaty rate. That verification requires a valid Certificate of Resident Status issued by the Inland Revenue Department, together with a beneficial-ownership declaration. Second, and increasingly, at the level of a post-payment audit: the CIS authority can look through the payment history and challenge withholding relief already granted if it concludes that the substance conditions were not met at the time of payment.

The beneficial-ownership condition is where the analysis is most frequently contested. A Hong Kong entity that functions solely as a conduit – receiving dividends and paying them up to a parent in a third jurisdiction with no independent decision-making – is exposed to beneficial-ownership denial even if it holds a valid residence certificate. The certificate confirms tax residence. It does not address beneficial ownership. Those are distinct tests, and CIS authorities are well aware of the distinction.

The sequence above describes the standard position. Your matter turns on the specific treaty in play, the structure of the entity holding the CIS interest, and the documentary record of economic activity in Hong Kong – which is where the position is won or lost.

To discuss how the treaty and substance analysis applies to your cross-border position, contact info@lockhartyip.com.

How does Hong Kong's territorial system interact with CIS treaty requirements?

Hong Kong's territorial basis of taxation is a structural advantage for many groups – but it creates a specific complication in CIS treaty access that is frequently misunderstood. Under the Inland Revenue Ordinance, profits tax applies to profits arising in or derived from Hong Kong. A Hong Kong entity can therefore receive foreign-sourced income that is not subject to Hong Kong profits tax. That is the system working as intended, and the foreign-sourced income exemption regime – in force since 1 January 2023 – conditions that treatment on economic-substance requirements for certain passive income categories.

The complication arises because CIS treaty authorities sometimes read the territorial system as evidence that the Hong Kong entity is not "subject to tax" in Hong Kong in the manner contemplated by the residency article of the treaty. If income passing through the Hong Kong entity is not taxed in Hong Kong – because it is foreign-sourced – then, the argument runs, the entity is not a resident "liable to tax" in Hong Kong for treaty purposes. This argument has been advanced in a number of CIS treaty challenges.

The Hong Kong Inland Revenue Department's position is that tax residency is determined by incorporation and management and control in Hong Kong, not by whether any particular stream of income is subject to profits tax. A Hong Kong-incorporated company managed and controlled in Hong Kong is resident in Hong Kong for treaty purposes, even if its income is not subject to Hong Kong profits tax on a territorial basis. That position is defensible and has been upheld in comparable contexts. However, a group relying on it must be prepared to document management and control in Hong Kong with genuine evidence: board meetings held in Hong Kong, decisions made by Hong Kong-based directors, banking facilities operated from Hong Kong.

This is the point at which substance and structure converge. A Hong Kong entity that is incorporated locally but directed from elsewhere – whether Moscow, Almaty, or a European headquarters – cannot credibly assert Hong Kong management and control. The residence certificate will be issued by the Inland Revenue Department on the basis of the facts presented. But if those facts are successfully challenged by the CIS authority, the treaty position collapses.

The interaction with the foreign-sourced income exemption regime adds a further layer. For a Hong Kong entity receiving dividends from a CIS subsidiary, the FSIE regime requires economic substance in Hong Kong to treat the income as exempt from profits tax. The substance required under FSIE and the substance required to defend treaty residence and beneficial ownership are not identical, but they overlap substantially. A group that has done the FSIE substance analysis properly will have addressed most of the treaty-access substance points. A group that has ignored FSIE on the assumption that dividends are simply not taxed in Hong Kong is exposed on both fronts.

Where does the principal-purpose test create the greatest current exposure?

The principal-purpose test – a provision now embedded in the domestic tax legislation of several CIS states following OECD recommendations – is the most significant current source of treaty-denial risk for Hong Kong holding structures. The test allows a tax authority to deny a treaty benefit where it is reasonable to conclude, having regard to all relevant facts and circumstances, that obtaining the benefit was one of the principal purposes of an arrangement.

The critical word is "one". The test does not require that obtaining the treaty benefit was the sole or even the dominant purpose. It requires only that it was among the principal purposes. That is a low threshold, and CIS tax authorities have shown a willingness to apply it broadly.

For a Hong Kong holding entity that was placed between a Mainland China or European parent and a CIS operating subsidiary, the structuring history is often directly relevant. If contemporaneous documents – board resolutions, investment committee papers, structuring memoranda – show that the reduced withholding rate was a primary driver of the choice of Hong Kong, that documentation becomes evidence for the authority applying the principal-purpose test. Groups that retained those documents without considering how they read in a treaty-denial context have found them unhelpful in audit.

The counter-argument is that the treaty benefit was not the only purpose of the structure. Hong Kong's common-law system, the efficiency of its corporate registry, access to capital markets, and proximity to Mainland China operations are all genuine commercial reasons for using a Hong Kong holding entity. Where those reasons are substantiated by contemporaneous evidence, and where the entity has genuine substance, the principal-purpose test can be rebutted. The question is whether the group built that evidence base at the time the structure was implemented, or is trying to reconstruct it years later under audit pressure.

In our cross-border practice, the latter is far more common than it should be. The cost of building a proper substance and documentation record at implementation is modest. The cost of trying to rebuild it during a CIS tax audit – with deadlines, information requests, and the possibility of treaty-benefit clawback across multiple years – is substantially higher.

If an earlier filing, structure, or treaty-access position has produced an adverse or stalled result with a CIS authority, a second read of the position can identify the strategic error and the routes still open. Write to info@lockhartyip.com.

How does the analysis differ across CIS jurisdictions?

The CIS is not a single tax environment. The bilateral treaties between Hong Kong and individual CIS states vary in their terms, and the administrative capacity and enforcement posture of each state's revenue authority differs materially. A group with exposure in multiple CIS states cannot apply a single treaty-access analysis uniformly across its structure.

Russia represents the most developed treaty-enforcement environment in the CIS. The Russian Federal Tax Service has invested significantly in transfer-pricing audit capacity, beneficial-ownership analysis, and treaty-denial procedures. The de-offshorisation legislation introduced over the past decade created a regime of controlled-foreign-corporation rules that altered the economics of many holding structures. For groups with Russian operating entities, the treaty-access analysis must account for CFC disclosures, the potential application of the anti-avoidance provisions, and the question of whether the group's ultimate beneficial owners have made the disclosures required under Russian domestic law. The Hong Kong–Russia double-taxation agreement remains in force, but its practical application has been tested by both the de-offshorisation regime and the broader geopolitical environment since 2022.

Kazakhstan presents a different set of issues. The Kazakhstani tax code has been modernised in stages, and the principal-purpose test is now a feature of domestic law. Kazakhstan has also introduced transfer-pricing rules and a register of beneficial owners of legal entities. For a Hong Kong entity holding a Kazakhstani operating company or receiving interest on a loan to a Kazakhstani entity, the substance and beneficial-ownership analysis runs in parallel with the transfer-pricing analysis on the intercompany terms.

Azerbaijan, Uzbekistan, and the smaller CIS states occupy different points on the same trajectory. Their treaty networks are less extensive, their administrative capacity varies, and the practical risk of a sophisticated beneficial-ownership challenge is lower in some jurisdictions than in Russia or Kazakhstan. However, the direction of travel is consistent: each jurisdiction is building the legal tools to challenge treaty access where substance is absent, and the enforcement of those tools is increasing over time.

The practical implication for a group with CIS exposure across multiple jurisdictions is that the substance and treaty-access analysis must be done jurisdiction by jurisdiction. A structure that works for a Kazakhstani subsidiary may not work for a Russian one. The common thread is that substance in Hong Kong – real decision-making, real directors, real banking – is the foundation for any treaty-access position across the CIS.

What does the decision matrix look like for a group reviewing its position?

A group reviewing its Hong Kong–CIS treaty-access position is likely to fall into one of several situations, and the appropriate response differs by situation.

Where the Hong Kong entity has genuine substance – directors based in Hong Kong, board meetings held in Hong Kong, banking and treasury operated from Hong Kong, and a clear investment rationale for the structure that is documented in contemporaneous records – the position is defensible. The priority is to ensure that the Certificate of Resident Status is current, that the beneficial-ownership declarations made to CIS payors are accurate, and that the FSIE substance conditions are being monitored and maintained. The risk is low but not absent, because even a well-substantiated structure can be challenged; the value of proper documentation is that a challenge can be met efficiently.

Where the Hong Kong entity has nominal substance – incorporated locally but directed from elsewhere, with directors who are not genuinely engaged in management – the position is exposed. The treaty-access risk is real, and a CIS audit could result in denial of treaty benefits retrospectively across multiple years of distributions. In that situation, the options are to build genuine substance in Hong Kong prospectively, to restructure the holding arrangement so that the beneficial-ownership point is addressed at a different level of the group, or to model the net cost of operating without treaty access and determine whether the structure remains commercial. Each option has different time and cost implications, and the right answer depends on the specific treaty, the CIS jurisdiction, and the group's commercial plans.

Where no Hong Kong–CIS treaty exists for the relevant jurisdiction, the analysis shifts. Unilateral relief under the CIS state's domestic law may apply in some cases. An alternative holding jurisdiction with a more favourable treaty position may be considered, subject to the same substance and beneficial-ownership analysis being applied to that jurisdiction. A group moving from one holding centre to another without addressing the underlying substance question simply transfers the risk.

A micro-scenario illustrates the pattern. An Asian group with a manufacturing subsidiary in Kazakhstan used a Hong Kong treasury entity to make intercompany loans to the subsidiary and receive interest payments at treaty rates under the Hong Kong–Kazakhstan double-taxation agreement. The treasury entity had two nominally local directors but was managed day-to-day from the group's Singapore regional office. When a Kazakhstani transfer-pricing audit expanded into a treaty-access review in late 2026, the authority questioned both the beneficial-ownership status of the Hong Kong entity and the arm's-length nature of the loan terms. The group came to us after the audit had produced an initial denial. We reviewed the structure and identified that a substantial portion of the decisions attributed to the Hong Kong entity had been made in Singapore. The beneficial-ownership position was reconstructed prospectively, but the retrospective exposure for prior interest payments required a negotiated resolution with the Kazakhstani authority. The outcome was qualitatively better than the initial denial, but the process took two audit cycles and consumed significant management time.

A second scenario: a European family group with a historical holding structure through Cyprus and a secondary Hong Kong entity sought to consolidate its CIS dividend flow through the Hong Kong entity following changes to the Cyprus–Russia treaty position. The Hong Kong entity had genuine substance – an experienced director based in Hong Kong with independent decision-making authority, documented investment-committee records, and a clear rationale for the choice of Hong Kong tied to the group's expansion into Mainland China and Southeast Asia. We reviewed the treaty-access position, confirmed the substance analysis against the FSIE conditions, and prepared the documentation file for use in any future Kazakhstani or Russian beneficial-ownership enquiry. The structure was sustainable and was implemented with a clear evidence base.

What do foreign advisers most commonly get wrong?

The most common error we see when reviewing structures built by advisers unfamiliar with the Hong Kong–CIS interface is conflating residence with substance. A Hong Kong Certificate of Resident Status is a useful document. It is not a guarantee of treaty access. It confirms that the Inland Revenue Department is satisfied that the entity is resident in Hong Kong for tax purposes on the basis of the information provided. It says nothing about beneficial ownership, nothing about the principal-purpose test, and nothing about whether the entity has the economic content to withstand a CIS audit.

The second common error is treating the CIS as a single jurisdiction. Advisers who have built a treaty-access position for a Kazakhstani entity and then applied the same analysis to a Russian entity, without accounting for the differences in the treaties and the very different enforcement posture of the Russian Federal Tax Service, have exposed their clients to risks they were not warned about.

The third error – and the one with the most lasting consequences – is failing to maintain the substance record after implementation. A group that builds genuine substance in Hong Kong at the outset but then allows the director appointments to lapse, moves the banking relationships to a more convenient jurisdiction, or stops holding formal board meetings in Hong Kong loses its treaty-access position gradually and invisibly. By the time a CIS audit arrives, years of distributions have been received at treaty rates on a position that has been quietly hollowed out.

The Inland Revenue Ordinance and the bilateral treaties do not require extraordinary levels of substance. They require genuine, documented, maintained substance. That is not a high bar for a group with real commercial reasons for using Hong Kong. But it does require deliberate attention at the implementation stage and ongoing monitoring thereafter.

Where is the risk concentrated in the current environment?

Our read of the current environment is that treaty-access risk for Hong Kong–CIS structures is elevated on two fronts simultaneously.

The first front is the continued development of CIS domestic anti-avoidance law. The principal-purpose test is now embedded in the treaty practice of the major CIS jurisdictions. Beneficial-ownership analysis has become a routine feature of withholding-tax audits in Russia and Kazakhstan. The administrative capacity to conduct those audits has improved. Groups that have not reviewed their treaty-access positions since their structures were implemented – in many cases five or more years ago – are operating on an analysis that predates these developments.

The second front is the interaction between the FSIE regime in Hong Kong and the substance conditions required for treaty access. The FSIE regime, in force since 1 January 2023, imposes economic-substance conditions on Hong Kong entities receiving certain categories of foreign-sourced income. For a Hong Kong entity receiving dividends or interest from a CIS subsidiary, the FSIE conditions and the treaty-access substance conditions run in parallel. A group that has not addressed the FSIE position is potentially exposed on both the Hong Kong tax side and the CIS treaty side simultaneously.

The window for addressing these positions without the pressure of an active audit is narrowing. CIS authorities that are building audit capacity are selecting cases from their registers of treaty-relief applications. A group that has claimed treaty rates on significant CIS income flows over multiple years is a visible target. The cost of a proactive review and remediation – substance analysis, documentation file, FSIE compliance check – is considerably lower than the cost of managing the same issues under audit conditions.

Groups using a Hong Kong holding or treasury entity as part of a CIS structure should, at minimum, verify the current Certificate of Resident Status, review the substance record for the past three years, confirm that the FSIE conditions are being met, and document the commercial rationale for the Hong Kong entity in a form that can be presented to a CIS authority in response to an information request. That is not a complex or time-consuming exercise for a well-run entity. It is, however, one that should not be deferred until a formal challenge arrives.

Our practice covers the full tax-positions and treaty-access analysis across the Hong Kong–CIS interface, including the FSIE substance review, beneficial-ownership documentation, and coordination with locally licensed advisers in the relevant CIS jurisdictions. For a preliminary read on your treaty-access position and the specific risk points for your structure, email info@lockhartyip.com.

Related practices

• Holding Structures – cross-border holding design, offshore centres, and substance requirements for Greater China groups
• Private Wealth – succession, asset protection, and family-office structuring across CIS and Asian jurisdictions

Speak with Lockhart & Yip

For a scoped view of your matter, contact info@lockhartyip.com. Discuss your matter →

Related

• Tax Positions
• Hong Kong Source Territorial Position Foreign Group Guide
• Profits Tax Position Hong Kong Trading Entity Analysis

This publication is general information and does not constitute legal advice. For advice on your situation, contact info@lockhartyip.com.