Ukraine
Technical Assistance Report-Distributed Profit Tax; Voluntary Disclosure of Assets; and BEPs Implementation

This Technical Assistance report focuses on Ukraine’s distributed profit tax, voluntary disclosure of assets, and Base Erosion and Profit Shifting Work Program implementation. The recommendations largely favor simplifying rules, improving the definition of basic concepts, eliminating potential loopholes, and adhering more closely to international standards in some cases. Thus, for the sake of simplification, the report recommends that Controlled Foreign Corporation rules should apply to the ‘first onshore’ person rather than having to trace them back to the ultimate beneficial owner in Ukraine. Also, it recommends that the proposed interest deduction limitation should eliminate the carry-forward currently permitted, limit deductions to net interest expense and exempt the financial sector from this limitation. Some key definitions can be improved too. The report suggests that if there is an urgent need to promote private investments, the accelerated depreciation tool should be applied for plant and machinery and structures housing them for say another five years.

Abstract

This Technical Assistance report focuses on Ukraine’s distributed profit tax, voluntary disclosure of assets, and Base Erosion and Profit Shifting Work Program implementation. The recommendations largely favor simplifying rules, improving the definition of basic concepts, eliminating potential loopholes, and adhering more closely to international standards in some cases. Thus, for the sake of simplification, the report recommends that Controlled Foreign Corporation rules should apply to the ‘first onshore’ person rather than having to trace them back to the ultimate beneficial owner in Ukraine. Also, it recommends that the proposed interest deduction limitation should eliminate the carry-forward currently permitted, limit deductions to net interest expense and exempt the financial sector from this limitation. Some key definitions can be improved too. The report suggests that if there is an urgent need to promote private investments, the accelerated depreciation tool should be applied for plant and machinery and structures housing them for say another five years.

I. The Distributed Profit Tax

A. Proposal to Replace the CPT with a DPT

1. It is still being debated in Ukraine whether to substitute the current CPT with a DPT, referred to in Ukraine as the Exit Capital Tax (ECT).1 The ECT applies only to legal persons and PEs established by non-residents that distribute profits or so called deemed dividends. It does not apply to legal persons and individual entrepreneurs that utilize the Simplified Tax System.2 The essential difference with the CPT is that the tax is deferred until the entity decides to distribute profits. The base of the tax is the amount of the distribution rather than a calculation of the entity’s accrued profits. This would be a fairly radical change of the corporate tax system, which has been discussed in the 2017 FAD report.3

Tax Base, Deemed Dividend Distributions, and Tax Rate

2. The tax base constitutes directly distributed dividends and capital gains to residents and non-residents in cash or in kind and who are non-ECT payers. It also includes transfers of funds by PEs to their parent companies considered remitted profits or deemed profit distributions. “Deemed” profit distributions are, for example, payments of interest and royalties to related parties or parties residing in low tax jurisdiction (LTJ), or cross-border transfer pricing adjustments to the non-payers of ECT. ECT does not apply to transactions between ECT payers who are residents of Ukraine. The ECT is charged on selected transactions rather than on financial results for the period. The accumulated deferred tax assets (including tax losses) from the previous CPT dispensation cannot serve as offsets against the ECT.4 All withholding taxes on Ukrainian-sourced income of non-residents are eliminated. ECT rates depend on the types of transactions: (1) 15 percent on dividends and capital gains payments to non-ECT payers; (2) 20 percent on “deemed profit distributions” (i.e., interest above certain threshold, royalties beyond specified limits, etc.); and (3) 5 percent on interest payments under certain conditions.

B. Alleged Benefits of the DPT

3. Intuitively, the proposed system appears to be simple as it would eliminate the need to track revenues and deductible expenses. Simplicity would be thus the main benefit from introducing the ECT. However, the ultimate argument made by proponents of the ECT is that it offers an escape route for taxpayers from the arbitrariness—or alleged corruptive practices5—of the tax authority. In all fairness, the CPT has been simplified substantially and remaining enforcement problems are rather administrative in nature. So, the main redeeming feature of the ECT is that it would solve the alleged administrative failure of the SFS by making the SFS mostly redundant. The other major benefit of the ECT is to make available more internal resources for enterprises to invest (see Box 1 for other pros and cons of the DPT).

Other Arguments Supporting the DPT

(1) The DPT does away with complex accounting-rules based measurement of taxable profit. Yet, standardized accounting rules must be followed to determine net earnings from which the dividends are declared; (2) it (partially) addresses neutrality between debt and equity; however, weak enforcement may create an incentive to distribute deemed dividends trough interest payments; (3) the DPT is levied on net cash flow; (4) there is a reduced need for debt financing; (5) tax depreciation rules become superfluous, but for accounting purposes, the corporate taxpayers would still calculate the economic depreciation according to an acceptable accounting rule, which will affect reportable net earnings; (6) because the company no longer pays tax on retained earnings, there is no longer a gross income line from which to deduct expenses; nonetheless, since losses reduce the net earnings position from which taxable distributions are made, the SFS should still track these in their effort of monitoring the correctness of the net earnings calculations; (7) since there is no tax on net revenue or earnings, losses have no significance for tax purposes either; (8) capital gains remain exempt from tax until profit distribution occurs but all the complications of determining the capital gain remain; and (9) a bit naively, since the overall tax system is simplified, the SFS staff can be materially shrunk.

C. Key Concerns About the DPT

Transitional and Permanent Decline in Revenues

4. The probable decline in CPT collections should be Ukraine’s biggest concern. Revenue of CPT from 2001 to 2018 was an average 3.9 percent of GDP. For the period 2017 – 2019, the average expected CPT revenue is 2.7 percent of GDP. This significant tax contribution is placed at risk with the tax deferral behavior commonly associated with a DPT. Estonia’s experience is indicative: Corporate Income Tax (CIT) revenue collapsed from an average of 2 percent of GDP in 1995–99 to 0.7 percent of GDP in 2001 (after the introduction of the reform). It took 14 years to reverse the initial impact. In fact, revenue never quite recovered as a percentage of GDP compared with the level it had in the second half of the nineties.6 Recent tax amendments seek to reverse this trend by applying a reduced CPT rate if companies declare more regularly dividends.7

5. By introducing the DPT, the MoF, assuming an overly optimistic 50 percent dividend payout ratio for companies, projects a UAH 65.6 billion drop in CPT collections in 2020. CPT for 2019, without the proposed ECT/DPT, is projected at 2.63 percent of GDP. It is estimated that introduction of the DPT in 2020 will translate into a revenue loss of 1.4 percent of GDP. The estimated decline in revenues for 2021 is UAH 101 billion, or 2 percent of projected GDP.8 Proponents of the ECT argue that the dynamic effects from increased employment and economic growth will at least partially compensate the direct loss in revenue. However, international experience does not support that claim.

6. Only 4 countries have adopted a DPT in substitution of a CIT. Examples are Estonia, a pioneer in 2001, North Macedonia (in 2009), Georgia (in 2017), and Latvia (in 2018).9 North Macedonia kept the regime for five years, dropping it in 2014 when it restored the CIT. Estonia is the benchmark in this policy discussion due to its long experience with this model. For the first five years after adopting DPT it averaged 0.5 percentage points below the previous five (almost 30 per cent drop), slowly recovering since then (Figures 13). Yet, revenue collections in these countries are disappointing.

Figure 1.
Figure 1.

Estonia: CPT Revenue (% of GDP)

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

Source: World Revenue Longitudinal Database (WoRLD).
Figure 2.
Figure 2.

Macedonia: CPT Revenue (% of GDP)

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

Source: World Revenue Longitudinal Database (WoRLD).
Figure 3.
Figure 3.

Georgia: CPT Revenue as % of GDP

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

Source: World Revenue Longitudinal Database (WoRLD).
Figure 4.
Figure 4.

Ukraine and Comparators: CIT Revenue, 2001 – 2018

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

7. A much more pronounced policy U-turn took place in North Macedonia, fully reintroducing the CIT beginning 2015. While in effect (2009 – 2014) the DPT revenue averaged 0.9 percent of GDP, 0.6 percentage points lower than in the prior four-year period. When the CIT was reinstated, corporate tax revenue tripled to over 3 percent of GDP (see Figure 2).

8. The cases of Georgia and Latvia are too recent to risk a conclusion. In the first year in operation, the Georgian CIT plus DPT revenue fell by a third compared to the average in the previous five years (Figure 3). A greater loss of revenue was avoided because Georgia kept the classical CIT for the financial and oil/gas sectors, extending the restriction from 2019 till 2023.10

9. The ECT initiative, for all practical purposes, amounts to a double tax cut. One for eliminating the current withholding tax on shareholders and another for permitting indefinite deferral of the corporate profit tax. There is an additional risk that wage earners, taxed at the flat 18 percent PIT rate, may incorporate themselves into independent service companies, benefiting from the new 15 percent ECT/DPT rate with a knock-on effect on PIT collections; currently, no effective anti-avoidance legislation exists to stop this tax planning avenue.

Unclear Impact on Foreign Direct Investment

10. Corporate investment in Ukraine since 1998 has averaged about 16 percent of GDP, dropping below 10 percent in 2014, the year the conflict with Russia began. It has recovered since then, but not to the levels (above 20 percent) it achieved prior to the 2008/09 financial crisis. This percentage is above the regional average and about mid-table relative to comparator countries. Also, it is clear that Estonia, despite having a DPT since 2001 cannot reverse the long-term declining trend of corporate gross fixed capital formation (CGFCF). In fact, for the period 2016–2017 Ukraine’s CGFCF exceeded that of Estonia (Table 1 and Figure 5).

Table 1.

Average Corporate Gross Fixed Capital Formation, 1998 – 2017 (in Percent of GDP)

article image
Source: WEO, OECD, Eurostat, and National Statistical Authorities.Note: Marked fields represent WEO estimates. Blank fields represent missing values.
Figure 5.
Figure 5.

Ukraine and Comparators: Corporate Gross Fixed Capital Formation, 1995–2017

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

11. Even though private (corporate) investment could be larger in Ukraine to catch up with the better performing countries in the region, the evidence from countries having introduced the DPT does not seem to indicate that this could be a significant factor. The apparent impulse from the DPT to corporate investment in Estonia, the benchmark case, seems short-lived (Figure 6, red bars represent CGFCF since introduction of the DPT). In fact, Estonia’s CGFCF is now lower than in Ukraine (Table 1). Of note is that Georgia introduced in 2017 the DPT and Georgia’s CGFCF, recording since 2013 the highest CGFCF for the comparator group; this strongly growing trend is reversed since 2017 (see Figure 5).

Figure 6.
Figure 6.

Estonia CGFCF, 1998–2017, (Percent of GDP)

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

Source: WEO, OECD and National Statistical Authorities.
Figure 7.
Figure 7.

Macedonia CGFCF, 1998–2017, (Percent of GDP)

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

Source WEO,OECD and National Statistics I Authorities.

12. North Macedonia’s experience is not encouraging either. CGFCF did not change much compared with the year prior to the adoption of the DPT regime and it slightly increased when this regime was abandoned.11

13. The IMF Staff Report (May 2019) argues that the main obstacles for greater private investment in Ukraine, especially foreign direct investment, is the lack of strong and independent institutions. Notably the judiciary (independence of courts) and weak regulations undercut the efficient operation of a modern market economy.12 There is no credible evidence that replacing the CPT for a far more lenient tax regime would contribute much to that end in Ukraine at this point. In fact, the 2017 TA Report presented empirical evidence generated by a global World Bank survey of firms investing abroad, that national taxes rank number 11 among the top 20 important factors determining their locational decisions.13

The DPT Makes the Tax System More Regressive

14. Ukraine’s tax system is relatively weak on progressivity. It increasingly favors indirect taxation (for the period 2001–2018 an average of 55 percent of total taxes) over direct income taxes (see Figure 8). The PIT by design is a flat rate regime at 18 percent, regardless of income levels. Flat PIT rates are not uncommon in the region: Estonia, Hungary, Russia and Bulgaria all adopted the flat tax for personal income. Many countries in the region, however, tax personal income at progressive rates, including Poland with a top rate of 32 percent (the country being the main destination to which currently Ukrainian labor migrates). Direct taxes contribute now only a third of tax revenues.

Figure 8.
Figure 8.

Ukraine: Growing Reliance on Indirect Taxes, 2001–2018

Citation: IMF Staff Country Reports 2019, 352; 10.5089/9781513521046.002.A001

Source: Ukrainian authorities and Fund Staff calculations

15. CIT adds an important element of progressivity to the overall tax system, since the incidence of this tax falls predominantly on owners of income from capital. Although this is a much-debated topic among economists, with differing empirical results, the emerging consensus is that an increase in CIT in the short run tends to fall almost entirely on income from capital14 and depending principally on how mobile capital moves across-borders, in the long run the incidence shifts only marginally on to labor and consumers.15 Currently, the U.S. Congressional Budget Office and the Joint Committee on Taxation assign 25 percent of the burden of the corporate tax to labor when preparing distributional results.16 A reduction in revenue from corporate taxation, as firmly predicted to happen with the adoption of the DPT, will render the Ukrainian tax system even more regressive, a move that may undermine further its legitimacy in terms of equity.

Tax Minimization Attempts to Benefit from Lower DPT Rate

16. When the CPT rate is comparatively low vis-à-vis the PIT rate, high wealth individuals or service sector employees tend to incorporate and collect dividends rather than a remuneration subject to PIT. Distributions among DPT-paying legal entities are exempt (as proposed under an DPT) and individuals find ways to have companies spend on their behalf, defying the enforcement capacity of the tax authority. At a minimum, anti-avoidance legislation should safeguard that, provided certain conditions are met, individuals rendering services through a personal service provider-company or as sole proprietor/independent contractor are deemed to be employed by the principal. Consequently, the principal is obliged to withhold wage tax and/or social security contribution from the fees paid to the personal service provider-company. Moreover, individuals or shareholders who own a substantial interest in a company are deemed to earn an arm’s length salary from which wage tax and/or social security contributions must be withheld by the company.

17. In any event, also as a protective measure for the current CPT base, it is recommended that for independent contractor companies income deeming provisions apply. This ensures that a reasonable portion of income received by such entities is treated as labor income. For example, in the Netherlands, directors’ fees in such situations are treated as ordinary employment income, and any employee who owns at least 5 percent of the shares in a company (a so-called substantial interest) is deemed to earn an annual salary of at least EUR 44,000, unless the taxpayer can prove that his actual salary is lower, or the authorities can prove that the salary is higher, on an arm’s length basis. Consequently, introduction of the DPT without accompanying anti-avoidance legislation for independent contractors or personal service companies could constitute yet another revenue leakage for the tax system.

Proposed “Compensators” Exacerbate Distributional Inequality

18. Proponents of the DPT admit that its introduction will be accompanied by significant revenue losses. In order to preserve revenue neutrality during the transition from the CPT to the DPT, they proffer a number of compensatory revenue measures that however may exacerbate the regressivity of the fiscal system: (1) e.g., increasing the VAT rate (by two percentage points); (2) cutting public expenditures (mostly adversely affecting the quality of life of low-income households); (3) letting companies pay for two years an advance payment based on their previous fiscal year’s CPT liability; (4) imposing a gross asset-based minimum tax; and (5) using privatization proceeds (this idea is unsustainable and against good practice).17

19. It is hard to imagine what the benefit of these measures would be since two proposals would cut into the free cash flow or reserves of companies that could have been used for investment. Indeed, these measures could impact adversely on start-up companies with current tax losses, thus benefiting more the established and profitable firms. The question then arises what attraction does the DPT really offer, except for only mitigating the incidence of discretionary SFS enforcement measures?

Illusory Simplified Administration and Improved Compliance

20. As dividends are distributed at the discretion of shareholders and since this event would trigger the DPT, shareholders would have a strong incentive to defer profit distribution or to find ways to disguise these through overpayments of other transactions, thereby avoiding tax altogether. To stop this practice, the new regime would also have to tax ‘deemed’ dividend distributions and at a higher rate than formal distributions—eroding the veracity of the argument that the new tax furthers greater simplicity. It will require the same transfer pricing counter measures and audits. The taxation of corporate profits is postponed until the profits are distributed as dividends or deemed to be distributed, such as in the case of transfer pricing adjustments, expenses and payments that do not have a business purpose, fringe benefits, gifts, donations, and representation expenses.

21. The DPT can only be enforced by controlling all company outflows—transaction by transaction. Highly complex administration is foreseen contrary to widespread beliefs by the proponent of the ECT/DPT. Potential deemed dividend distributions need to be checked for every transaction, requiring extensive enforcement of transfer pricing rules, which in the current system is done typically on an annual basis, not transaction by transaction. It is peculiar, that the ECT draft law disallows the tax authorities to check the financial statements of ECT payers (except for the purposes of transfer pricing adjustments). When the ECT payer transacts with a non-ECT payer and the transaction giving rise to a taxable event, the ECT payer will have to insert into the payment documents a specific code for types of transactions. The commercial banks will have to accumulate the coded payments data and share it with the tax authorities on a regular basis. This will add to the cost of banking as banks become a key player in the administration of the DPT. Only when there are discrepancies between this data and the ECT reporting system, is the SFS permitted to request explanations from ECT payers. Transactions representing taxable events, but which have been disguised as non-taxable, can be challenged by the tax authorities in court.

22. Finally, tax practitioners are of the opinion that the ECT Draft Law contains numerous easy-to-exploit tax loopholes that are difficult to police. In fact, given their complexity the SFS will be hard-pressed to enforce ECT payment and combat tax fraud, because the Draft law limits the tax authorities’ powers. Because of the loopholes as well as due to the specifics of the taxable events (ECT is a tax on specific transactions, and not on the financial result for the period), ECT could effectively become a semi-voluntary contribution.

D. Accelerated Tax Depreciation Rewarding Fixed Investments18

23. In order to attract new investment capital, countries often resort to offering tax incentives that reduce tax burdens for new investors. Incentives may be viewed as a necessary cost for offsetting other high non-tax production costs. While these arguments have merit, international experience suggests the need for careful cost-benefit assessment of tax incentives. For this purpose, the benefits should be measured as the incremental economic activity undertaken as a result of the incentive, since some investment would take place even without tax incentives. The costs of tax concessions arise through forgone revenues or the opportunity cost of public expenditures not being made due to these lost revenues. This is highly relevant when contemplating the introduction of a DPT.

24. Cost-effective tax expenditures to promote fixed investments are not profit-related incentives but are determined as a percentage of investment expenditure. They come in the form of accelerated depreciation, additional investment tax credits, and capital allowances. Where the statutory CIT rate is relatively high, decreasing the rate may be attractive – as it may be reducing tax-planning pressure on the CIT base. Ukraine’s CPT rate of 18 percent is already well below the worldwide average statutory CIT rate of 23 percent.19 Hence, reductions in the statutory CIT to attract investment should be tempered by the fact that other instruments may be more efficient in promoting investment. For example, it can be achieved by subsidizing the cost of investment, and not by cutting tax on economic rents. Generally, tax on economic profit should not influence investment decisions. Thus, surrendering tax on economic profit should be avoided on efficiency grounds, especially where profits are location-dependent.

Current Law

25. The Ukrainian CPT permits accelerated depreciation. Moreover, taxpayers may carry forward their losses indefinitely. This potentially translates for an investor into a de facto multi-year tax holiday whereby CPT tax liabilities are deferred if the deduction for depreciation generates a tax loss for more than one year. Consequently, investors in the Ukraine already benefit from generous tax depreciation rules on fixed assets: it is unclear how much additional investment will be generated by the DPT as firms may simply park funds in cash reserves.

Rewarding Fixed Investment and Not Reserve Accumulation

26. The mission suggests that if there is an urgent need to promote private investments, the accelerated depreciation tool should be applied for plant and machinery (Group 4 asset class) and structures housing them for say another five years. Since the SFS is familiar with it, there is no need to introduce alternative instruments such as an investment allowance or investment tax credits. The accelerated depreciation allows for a faster depreciation schedule than available for the rest of the economy. Tax payments in nominal terms are unaffected, but their net present value is reduced, thereby improving the liquidity of firms. Consequently, accelerating a depreciation allowance reduces the effective tax rate.

Recommendations

  • Given the transitional and permanent projected revenue losses as a direct result from introducing a DPT, Ukraine should not attempt this radical corporate tax change.

  • Correct discretionary and non-proportional enforcement actions by the SFS with the view to improving the administration of the CPT—it does not justify the introduction of a new tax.

  • In lieu of the proposed ECT, consider adopting more attractive accelerated depreciation schedules that directly reward investment into plant and machinery.

II. Voluntary Disclosure Program

A. Introduction

27. Zucman20 estimates that approximately 8 percent of the global financial wealth of households is invested in tax havens, of which ¾ is unrecorded. These offshore assets are mainly owned by European residents. A large percentage of these assets are held and managed in European tax haven jurisdictions with strict bank secrecy and data privacy laws. The existence of secretive tax havens erodes public finances by abetting the avoidance or evasion of taxes, undermining tax authorities’ compliance actions.21 Exchange of information between countries became the first line of defense against international tax evasion. Yet, its effectiveness is in doubt as treaties only provide for the exchange of information about a specifically named person; a so-called “fishing expedition” for taxpayer data is not possible.

28. Some countries have moved aggressively to obtain confidential bank data from tax havens offered to them for sale by whistleblowers. This is a second line of defense but the use of such unauthorized disclosure of information by tax authorities is exceptional. Although it has been accepted by courts, it is probably not sustainable over the long run.22

29. A third counter measure, voluntary disclosure programs (VD) for foreign asset holdings with supporting tax amnesties, have gained popularity with tax authorities. Several countries have launched VD, tax amnesty, and asset repatriation as a package, utilizing in part the pressure on investors to report in terms of financial intelligence legislation the origin of their funds as they cross jurisdictional boundaries. These asset VDs can broaden the income tax base, with increasing taxes on reported earnings from offshore portfolio and fixed investments.23

30. Tax evasion is a pervasive problem to which policymakers generally respond with enhanced auditing and legal sanctions against tax evaders. Another quite common response is to offer some kind of tax amnesty. These can take many different forms, but the essence is to give taxpayers a chance to come clean in respect of past offenses and to escape or reduce taxation and/or penalties. In the short term, governments retrieve part of tax revenues previously evaded; in the longer run, they hope to bring back a number of tax evaders into the tax net, effectively expanding the tax base. Amnesties have also been used to reverse capital flight, i.e., the VD programs.

31. Yet, tax amnesties, especially frequent ones, are to be avoided since they trigger long-term declines in revenue collections due to an erosion of tax morale of compliant taxpayers. If tax amnesties are repeated with regularity taxpayers reasonably conclude that it pays to transgress tax laws as revenue authorities show leniency to lull back tax evaders who openly flout the law by forgiving penalties and interest. In contrast, tax amnesties combined with asset repatriation programs are more successful when they happen in the new governance framework of bilateral tax treaties with automatic exchange of information among tax authorities (AEOI Standard) and anti-money laundering (AML) cooperation with jurisdictions where undeclared assets are located. In tax amnesties, taxpayers are obligated to pay outstanding taxes with relief for penalties and interest due to late payments. In VD programs, penalties and interest are often waived, or immunity from imprisonment is granted, and confidentiality to taxpayers is granted only for those assets voluntarily declared and adjudicated by the authorities. This protects them against repeated audits and harassment by tax administrations.

32. VDs are a recent and new tax compliance trend that has been introduced by a number of countries. The use of relatively narrow-window VDs as an integral part of tax agencies compliance strategies is a development that will likely grow in use. Their success in encouraging targeted taxpayers to voluntarily disclose their hitherto untaxed assets depends on an increased detection capability by the authorities (central bank, financial intelligence authority, exchange of information and revenue administration) and a firm commitment to take follow on action against taxpayers who continue to hide their assets. This requires ramping up the risk of increased penalties, reputational risks, and potential prosecution. Unlike many tax amnesties, VDs are not focused on short-term capital repatriation of assets—rather the long-term benefit from a broadening tax base (i.e., foreign assets) which can be taxed prospectively. Rather VDs aim to increase voluntary compliance by targeting a high-risk area of non-compliance.

33. While VDs apply only for several months, the over-arching compliance strategy with tougher audits and enforcement steps for non-disclosed assets (supported by AEOI standards) should become a permanent feature—and this is a goal the SFS should aspire to. This more nuanced approach distinguishes VDs from many amnesties that seek to broaden the tax base in advance of promised reforms by ensuring that the promised changes become reality. Importantly, successful VDs tightly manage the benefits offered to disclosing taxpayers to ensure that these taxpayers pay a price for their non-compliance (so-called fees on gross assets) and to maintain and ideally increase the tax morale of complying taxpayers.

Key Elements of Tax Amnesties

34. Countries typically introduce tax amnesty programs mainly for the purpose for raising revenues quickly, and/or encouraging the repatriation of illegally held foreign assets. General tax amnesties not explicitly linked to a VD program, vary in terms of duration, qualifying taxes, and types of absolved penalties and interest. Irrespective of the controversial nature of amnesties, they remain popular. Evidence is mounting that tax amnesties (without a VD) generate much less incremental revenue than expected and there are high opportunity costs associated with tax amnesties for the simple reason that tax administrators are diverted from normal collection procedures of taxes, and without additional resources, enforcement and collection backlogs may develop elsewhere in the system. In any case, amnesty programs should be one-time events to clean the slate before the introduction of a new tax instrument or a new efficiency-enhancing organizational model for the revenue administration.

35. Typically, VDs are for a limited period in connection with a specific opportunity, such as increased cooperation with other tax administrations through AEOI provisions (OECD, 2010). These programs allow for voluntary taxpayer disclosure and regularization of taxes on non-disclosed foreign assets. They are often associated with the option of tax relief to repatriate these within the set timeframe of the VD program. Obligatory repatriation is less frequent but usually the program prescribes or regulates how repatriated assets can be invested domestically. Fixed assets like real estate have limited applicability with repatriation programs but their asset disclosure for income tax purposes (rental and capital gain income) is very important.

Cross-Country Experiences

36. A 2015 OECD survey found that 47 countries implemented tax amnesties, of which 13 jurisdictions administered disclosure programs with incentivized asset repatriation.24 Sizable asset amounts were regularized, and tax bases were increased: Argentina broadened the tax base by USD 100 billion; Brazil by about USD 16 billion; Indonesia by USD 321 billion; and South Africa by €7.8 billion (see Table 2 for a more detailed discussion of some VD programs).

Table 2.

Cross-Country Comparison of VD and Asset Repatriation Programs

article image
article image
Source: OECD 2010 and Transparency International, 2017.

B. Ukrainian Voluntary Disclosure (VD) Proposal

Objectives and Key Design Features

37. The planned VD program aims to promote: (1) declaration of natural persons’ foreign asset ownership; (2) payment of outstanding taxes related to these assets; (3) broadening of the tax base; and (4) future implementation of the “indirect control methods” over personal income. Taxpayers will need to disclose foreign-based assets, property and income for reporting periods also preceding the Ukrainian tax statute of limitations of three years. The VD program is scheduled from 1 July 2019 to 31 December 2019.

38. The VD program requires the payment of a one-off voluntary disclosure fee (the “Fee”) at different rates for assets declared by the resident natural person. By voluntarily disclosing the locality and existence of foreign assets, and by paying the Fee in lieu of all standard taxes, legal certainty and immunity is bought against further investigations and prosecution for prior related tax misdemeanors with respect to those assets and property disclosed as part of the VD procedure. Any parties or resources that assisted with the foreign assets’ acquisition do not need to be revealed.

Persons Qualifying for VD and Eligible Assets

39. Any Ukrainian individuals (including entrepreneurs) who own (co-own) and/or are beneficiaries/beneficial owners (controllers) of assets eligible for VD, may participate in the program. Eligible assets are portfolio investments (funds), fixed property, and other assets acquired before 1 January 2019. It includes rights or beneficial ownership of securities, rights to shares in property of legal persons or rights to funds, including a controlling interest in controlled foreign companies (CFCs).

Guarantees for Natural Persons

40. The VD immunity against prosecution for tax offenses covers the following cases:

  • If the individual pays the Fee for disclosed assets and income, which was the source of the acquisition of offshore assets or other disclosed income, the taxpayer is relieved from any other tax obligation or compulsory payments under the Tax Code (including a military levy)— but only with regard to disclosed assets/income, with the exception of the property tax.

  • If the SFS audits an individual for periods before 1 January 2019 after the VD ends, the audited tax base and tax liabilities are reduced by the amounts of assets voluntarily declared.

  • During and until the end of the VD period, natural persons may not be audited for periods before 1 January 2019. This restrict the SFS unduly and legislation should stipulate that individuals who are already under investigation by the SFS cannot apply for the VD.

  • The VD shields natural persons from criminal prosecution for crimes set forth in the draft law (i.e., sham business activity, negligence, document forgery, tax default)25 and penalties for tax transgressions with respect to the disclosed assets and sources of their acquisition. In assessing the tax penalty, chargeable income shall be decreased by the disclosed VD base.

  • Individuals, who apply for the VD, receive immunity against administrative penalties for tax and exchange control infringements, but only if they voluntarily disclose their assets.

  • Information provided by individuals as part of the VD program shall remain confidential—to protect them against repeated audits for these adjudicated and hence, regularized assets. This immunity does not apply in the case of criminal investigations—for which a VD amnesty should not be granted.

  • Any information revealed during a VD may be audited in the case of “public” persons.

  • Being the beneficial owner of assets held by a trustee/nominee owner, the individual may disclose such assets, determine the VD Fee base and pay the Fee.

Limitations on Immunity

41. Co-owners of assets must severally apply to individually secure immunity from prosecution. “Public persons” (i.e., politicians and officials) qualify for said immunity but must meet additional requirements such as audits and a disclosure about the sources of funds used for acquiring offshore assets. Immunity does not apply to assets which were acquired from proceeds of committed corruption and crimes regardless of their timing (except for Articles 205, 212, 212–1, 366, and 367 of the Criminal Code of Ukraine).26

Procedure of Disclosure, Preparation and Submission of VD Report

42. Individuals, qualifying for the VD program, may (1) include into their report any assets for which they seek immunity. Furthermore, (2) they may—but are not obligated—value the disclosed assets; (3) VD applicants self-assess the base for the Fee calculation of the disclosed asset(s); and (4) VD applicants should however seek to accurately value their assets and pay the Fee to secure immunity against further taxes and penalties. The VD report is submitted to the SFS in electronic form, which is authenticated by appending a digital signature. A natural person may only submit one VD report and if he/she wishes to amend the previous report a new VD report must be submitted before the VD deadline. In terms of the Ukrainian draft VD provisions, the exchange control amnesty only applies to disclosed foreign assets. Non-disclosed foreign assets remain fully subject to potential civil and criminal prosecution. “Public persons” disclose assets under the general VD procedure but immunity will not be granted for assets acquired out of proceeds from corruption and related crimes. Thus, public persons’ assets are considered disclosed only after the VD report has been audited by the assigned authority as per required procedures of the anti-bribery law.

Proposed VD Fee and Completion Period

43. The Fee is levied by applying the relevant rate (varied by an election of either keeping assets abroad or repatriating them) to the asset base as delineated by the individual. The following rates apply: (1) 9 percent for assets voluntarily disclosed by an individual who determines the VD Fee base and includes cash not held in bank accounts or held on bank accounts abroad; (2) a 5 percent rate for disclosed cash (in both UAH and foreign currency) held in accounts of qualifying Ukrainian domestic banks;27 and (3) a 0 percent on disclosed cash amounts if invested in state banks/state securities to incentivize such portfolio holdings.28 The Fee is payable within 30 days upon submission of the VD report. One should note that the 9 and 5 percent fees are imposed on gross assets, which in some cases may be high effectively, compared to the 18 percent PIT withholding on interest income and capital gains.

Audit of Data Contained in the Voluntary Disclosure Report

44. Information provided in the VD report may not be audited or used against a taxpayer as part of a tax enforcement action by the SFS. The assets only contained in the VD report receive “ring-fenced” immunity against such SFS enforcement action, including an in-office audit of computational errors in calculating the Fee base. If technical errors or additional asset information need to be corrected and reflected in the VD report, the individual can amend the respective data but must submit it in a new VD report within the prescribed VD deadline.

Estimated Amount of Hidden Portfolio Assets Hidden Abroad

45. With the Ukrainian exchange control liberalization efforts, accounts that have thus far been hidden from the Ukrainian authorities, may resurface. BIS data on foreign bank positions, known as Locational Banking Statistics (LBS), revealed the existence of large accounts of Ukrainian clients held with reporting banks to the amount of USD 3.5 billion by the end of 2017.29 After the introduction of the liberalization, holders of these accounts should regularly report on balance changes in them.

C. Costs and Benefits of Voluntary Disclosure Programs

Drawbacks

46. Proponents of general amnesties argue that amnesty should be offered to taxpayers in transitional periods from levels of poor enforcement to enhanced enforcement. The same argument holds for VD programs with associated tax amnesties as new currency laws with stricter financial intelligence legislation force holders of foreign assets or portfolios to come forward, disclose their existence with the view to accessing these funds in future. But the fact remains that amnesties are always discriminatory as compliant taxpayers absorb the full statutory tax burden for no reward whilst non-compliant taxpayers get a discount (waiver of interest, penalties, and even tax)—perhaps not only once, but repeatedly in case of frequent amnesties. International experience suggest that long-run revenue effects of amnesties are discouraging; frequent amnesties have a deleterious effect on long-term revenue collections (see Box 2). The mission, therefore, strongly advises the authorities, if they chose to announce a VD, to commit to a one-off opportunity which will not be repeated.

Pros and Cons of Tax Amnesties

Amnesties are advocated as a quick fix to close the tax gap, which is defined as the difference between taxes voluntarily paid and taxes that would be collected if taxpayers accurately report their tax liability.

Pros:

  • Allow taxpayers to comply in respect of past tax law infringements.

  • Permit immediate collection of outstanding taxes.

  • Allow ‘marginal’ tax evaders who have dropped out of system to come back and turn a new leaf (they may have missed one year of filing and the cost of coming clean may be too high) or those who omitted to include an item once, are forced annually to repeat the same omission.

  • May increase future voluntary compliance.

  • May aid in improving tax authorities’ records and subsequent growth of the tax base.

  • Are appropriate if they prepare for a transition period prior to a new and stricter enforcement regime or the introduction of new tax instruments.

  • Asset repatriation amnesties with supporting tax measures are marginally more effective if accompanied by anti-money laundering legislation together with a liberalization of exchange control regulations.

Cons:

  • Reduce expected cost of non-compliance and reward delinquent taxpayers.

  • Create a negative signaling effect on future compliance culture.

  • Could impact adversely on future revenue collections.

  • Reduce fear of strict enforcement and may encourage ongoing evasion because taxpayers have wiped their slate clean in relation to previous tax infringements.

  • Post-amnesty revenue effects are discouraging unless frequent audits of high-income taxpayers are not statutorily prohibited, and a tax court exists that reduces court overload and prosecution cost.

  • Compromise fairness as honest taxpayers attract a higher average effective tax rate than tax fraudsters—a VD fee applied to gross assets however mitigates this unfairness.

  • Generally, ex post amnesty compliance levels are lower than before the amnesty was rolled out.

  • Undermine effective progressivity of tax system, since tax evaders are typically wealthier than the median taxpayer and a low flat rate amnesty charge as compensation for previous tax infringements eliminates the progressive rate structure of an income tax regime.

Benefits of VDs

47. The key benefit from a VD-tax amnesty is the repatriation of flight capital and increased tax revenues from an expanding tax base. Overall, the VD program can be positive for the balance of payments as foreign assets are repatriated, together with the strengthening of the financial system. Over the short term, the VD-amnesty program will yield additional revenues cost-effectively as fewer regular audits may be required. Costly litigation can be minimized as the tax base is being broadened by taxing elements of previously undisclosed income transferred abroad and being forced into the open by tougher financial intelligence reporting and AEOI. Next to an increase in current and future tax revenues, VD programs could improve medium-term tax compliance if coupled with a more capacitated revenue administration. An amnesty eases the transition of non-compliant taxpayers towards a new enforcement environment since the reduction in penalties, interest, and tax relief, replaced by single levies, assist taxpayers in becoming compliant at a reduced cost.30

D. Strengthening the Draft Ukrainian VD Program

48. Despite the existence of exchange controls, Ukrainian individuals and entities have shifted assets offshore illegally. The foreign income earned from these assets goes unreported and untaxed even though the Ukraine taxes all passive income on a worldwide basis. The government has rightly taken the position that contraventions of currency control regulations and the Tax Code (TC) should not be condoned, should be disclosed henceforth, and income from these assets should be taxed. Given the changed environment of greater international mutual assistance in tax administration amongst nations (i.e., automatic tax information exchange) and closer AML cooperation, individuals and entities may wish to disclose and repatriate their foreign held assets voluntarily and regularize their affairs. Hence, the closer cooperation in taxation and integrated financial intelligence tracking has increased the risk of holding illegal foreign assets. Also, the world community is increasingly intolerant of tax haven countries and has reinforced measures to combat illegal money laundering. This section discusses areas where the VD concept could be sharpened together with the adoption of more pragmatic administrative measures.

49. A VD program should contain provisions that allows Ukrainian residents to disclose their foreign assets held in contravention of currency laws and tax provisions. Some of these principles are already contained in the draft VD concept note. Disclosure is needed as it will allow residents to exonerate and regularize their exchange control and income tax affairs at minimal cost. Key VD procedures are as follows: (1) Any Ukrainian resident natural person, or trust may apply for tax amnesty relief. The legislation should however explicitly provide for individuals, deceased estates, trusts and close corporations to apply for amnesty if they were facilitators in taking capital offshore. It is good practice to specifically exclude public and private companies from applying as applicants;31 (2) The application period should be a year or less;32 (3) The VD and tax amnesty should only apply to individuals and entities that come voluntarily forward – those who are already under investigation by the authorities involving their foreign assets are precluded from the amnesty process, following standard international practice; (4) The application or VD must be filed in the form of an affidavit and most importantly, the applicant must also affirm that the foreign assets disclosed do not stem from any unlawful activity, except exchange control and tax law violations or any associated misrepresentation or nondisclosure in respect of those violations. Thus, more refinement as to facilitating structures and the adoption of a reasonable VD application period is advisable.

50. The proposed VD program with associated tax amnesty intends to have foreign income-earnings assets disclosed and preferably repatriated, that may have been structured and controlled through Controlled Foreign Corporation (CFC) that are owned by Ukrainian residents. The imminent introduction of BEPS legislation, with the introduction of related CFC legislation, constitutes therefore an additional trigger for this initiative. In this context, it is a reasonable objective to encourage taxpayers to regularize their hitherto undisclosed offshore assets and to begin paying taxes on the recurrent income (rents, interest, dividend and royalties). Inherent risks such as taxpayers abusing the well-intended VD for money laundering and continued tax avoidance can be mitigated through proper planning and organizing its administration in a specially established amnesty unit. Extending the allowed period for the VD program would be helpful as well. Given the VD proposed relatively short window of opportunity to declare assets, large volumes of transactions may be declared that may overwhelm the capacity of financial institutions and the NBU. Both of these institutions need to apply a proper verification process in terms of AML and counter-terrorist financing (CTF) measures. Indeed, it may be especially burdensome for financial institutions to distinguish ordinary transactions from those related to the VD program.33

Adhering to Core Principles for a VD program

51. The VD program should be guided by key principles and the effective application of AML and CTF preventative measures. Other important do’s and don’ts are:

The Do’s and Dont’s of VD Programs1

The Do’s—The OECD refers to features that make a VD with related tax measures efficacious:

  • The VD program must be clear about its goals and scope.

  • The program must evidence a cost-effective increase in short-term revenues.

  • The adopted VD program must be aligned with the generally applicable compliance and enforcement culture of the tax administration, and which in the case of the Ukraine, needs further enhancement.

  • The VD program needs to specifically improve compliance levels of the target group for whom the VD program was designed and maintain a credible deterrent against future compliance slippage.

  • The VD program with its short-term boost to revenues must unfold in a generally improving compliance environment across all taxpayer segments with measures that improve tax compliance over the medium term.

  • Clear legislation and regulations for an asset repatriation program need to create taxpayer certainty and guarantees against subsequent recurring tax audits of previously disclosed and hence regularized foreign assets.

  • The short-term gross revenue gains from the VD program should be balanced against (1) a reduction in taxpayer compliance due to continuing weaknesses in the tax administration and declining morale of compliant taxpayers; (2) the administrative cost of the amnesty; and (3) the cost in forgone revenues from waived interest, penalties and tax.2

  • When implementing a tax amnesty-asset repatriation program (i.e., a VD program), the tax authorities should be adequately resourced to deal expeditiously with applications that are commonly complex transactions. There should also be an advanced data management system for coping with the influx of taxpayers and for analyzing amnesty transactions and information, which can be used to uncover legal persons that remain outside the tax net.

  • Successful VDs provide a level of protection to facilitators of avoidance schemes so that the mechanics could be disclosed and from which the revenue administration can draw lessons on how to tackle tax avoidance in the future.

  • Tax amnesties need to be an integral part of wider voluntary compliance and enforcement strategies, as an amnesty can never on its own be effective in inducing taxpayers to honestly declare unreported income. Taxpayers only come forward, if the probability of detection is significantly increased through new enforcement strategies. This approach would also reassure compliant taxpayers if governments ramp up efforts to prosecute tax offenders.

  • Tax amnesty programs with asset repatriation involve multiple countries with dispersed information where asset and portfolio investments are held. This complicates the verification task for financial institutions and the authorities. Recent progress with AEOI strengthens enforcement and taxpayers seeking to conceal their wealth, can no longer hide from tax authorities whilst simultaneously benefiting from a secure investment environment.

  • To protect against abuse of the offered voluntary tax disclosure program, the authorities should put in place all-embracing arrangements for mutual legal assistance and exchange of information, and prosecutions This should also include asset recovery investigations and proceedings, where applications were unsuccessful (FATF, 2012). This issue refers to the OECD Standard for Automatic Exchange of Financial Account Information in Tax Matters, which invites governments to obtain detailed account information from their financial institutions and exchange this automatically with other jurisdictions on an annual basis. The Standard, implemented by several jurisdictions, reduces the scope of tax evasion, enables the discovery of undetected tax evasion, and makes the VD well-timed.

The Don’ts—

  • Do not repeat the VD but begin to enforce the tax and currency legislation to the full letter of the law.

  • Individuals who are already under investigation by the SFS cannot apply for the VD program.

1 OECD, 2010. Offshore Voluntary Disclosure—Comparative Analysis, Guidance and Policy Advice and OECD. 2007. Improving Access to Bank Information for Tax Purposeshttps://www.oecd.org/ctp/exchange-of-tax-information/39327984.pdf2 Explanatory Memorandum on the South African Exchange Control Amnesty and Amendment of Taxation Laws Bill, 2003. The Bill was promulgated as Act No 12 of 2003.

Aligning the VD and Asset Repatriation with Anti-Money Laundering Provisions

52. It is reassuring that the NBU seeks to align the Ukrainian VD and asset repatriation program with FATF principles that disallow full or partial exemption from AML requirements.34 The four principles are: (1) repatriated assets are deposited with a financial institution that is subject to AML and CTF measures; (2) assets coming from countries that do not adequately apply the FATF recommendations are given particular attention; (3) the authorities raise awareness among financial institutions on the potential for abuse and the money laundering risks in the asset repatriation program; and (4) that any documents or statements issued regarding the assets repatriated endorse the legitimacy of their origin (FATF, 2012). Thus, financial institutions must conduct a Customer Due Diligence for clients who repatriate assets and identify the beneficial owner of the account into which the assets are repatriated.

53. For purposes of the tax amnesty and asset repatriation, domestic and international coordination between the Ukrainian bank and tax authorities with oversees counterparts is crucial. First, the Ukrainian tax authorities should receive sufficient investigative powers to trace the origin of assets or refer such investigation to competent authorities in the jurisdiction where said assets are located. Second, mechanisms should exist that would allow for effective information sharing on taxpayers and their repatriated assets. Finally, countries from which assets are being repatriated should provide full cooperation to the Ukrainian authorities.

Consider Amnesty for Domestic Tax Infringements

54. The current concept note on Ukrainian VD does not provide for a penalty payment for non-disclosed and non-taxed domestic income that gave rise to the foreign asset acquisition. VD programs must therefore strike a fine balance between compliant taxpayers who declared all their assets and duly paid the taxes thereon and the non-compliant taxpayers who, due to non-disclosure, will benefit from a lower average tax burden. For that reason, the mission proposes the introduction of an additional charge on undisclosed domestic income that may have given rise to the creation of foreign assets. The underlying concern is that amnesties always benefit the tax offender vs. the compliant taxpayer. For that reason, the South African foreign asset amnesty imposed an additional 2 percent charge on all domestic tax infringements (i.e., Income Tax, Donations Tax, Withholding Tax on Dividends, Royalties, etc. and Estate Duty), but only to the extent these domestic transgressions related to foreign assets. In this case, applicants must disclose the amount and date of foreign assets shifted offshore. The price of the amnesty for a domestic tax violation comes at the price of a domestic amnesty levy of 2 per cent on previously undisclosed amounts. Also, the domestic tax amnesty will not cover other tax violations in respect of VAT, wage withholding tax, and social security contributions. These taxes were excluded from the amnesty because violations of this kind typically involve serious fiduciary violations such as the wrongful use of withheld tax from salaries.

Procedural and Administrative Matters

Institution of an Independent “VD and Amnesty Unit”

55. The proposed VD-tax amnesty program could benefit from the establishment of a centralized amnesty unit for processing the applications. An independent chairperson (e.g., a retired judge) should be appointed and the unit should contain personnel from both the currency control division in the NBU and the SFS. The unit will terminate after processing all successful VD applications and after all unsuccessful applicants have exhausted their appeals.

56. Adhering to elevated confidentiality standards is a requirement of tax amnesties with asset repatriation programs as taxpayers need the guarantee that their disclosed offshore financial affairs will not trigger repeated audit harassment by tax administrations. Information dissemination about amnesty applicants should be carefully managed as it may secure future compliance and hence, officials dealing with the VD program should operate in line with special tax secrecy provisions for officials assigned to the proposed amnesty unit. The suggested amnesty unit, centralized without any regional branches, could be a guarantor against repeated tax audits by only making successful applications available to the SFS and the NBU so that applicants receive their desired amnesty protection. With this in mind, all amnesty applications (files) should be stored physically and/or electronically in the amnesty Unit’s central filing facility and stored under access control of the Unit. Furthermore, the “amnesty unit” should provide information about the amnesty’s progress to the Minister of Finance and the Rada. This information will ensure that the amnesty is conducted in a transparent and accountable way.

57. The Amnesty Unit legislation should provide for quorum arrangements of decision-making officials from the two institutions. In processing VD applications, one approach could be that all applications must be submitted for consideration by all Unit members at a full Unit meeting (effectively guarding against corruption through single discretionary decision-making). Since this approach slows down the adjudication process it may be more efficacious to provide for quorum arrangements, whereby applications are considered by teams of two adjudicators, comprising one person from the NBU and one person from the SFS. Yet, all declines and appeals of applications will still be considered at full Unit meetings despite the quorum arrangements.

Valuation of Assets

58. In the proposed Ukrainian VD program, applicants voluntarily self-assess the base for the Fee calculation of the disclosed asset(s). Applicants should however seek to accurately value their assets to secure immunity against future taxes and penalties. Not requiring an expert valuation may hurt both parties to the VD program. For that reason, other amnesty programs do require a statement of foreign assets’ market value at a set amnesty date with a full description of assets’ identifying characteristics. Ideally, the asset values need to be confirmed by supporting documents, the particulars of which appear in Table 3:

Table 3.

Particulars of Asset Types and Required Documents of Proof

article image
Levy Collection and Repatriation of Funds

59. The VD and amnesty levy (5 and 9 percent) should be paid from foreign-held assets before taking into account any fees or commissions charged by financial institutions on the conversion of the funds. The principle is that the already discounted levies in lieu of outstanding taxes should not be further eroded by charges in the banking sector. If the amnesty levy for domestic tax violations, that enabled the placing of funds offshore, is found acceptable it can be paid within 30 days of the approval of the application from local funds. Amnesty applicants and facilitators should pay the levies through the banking system to an NBU controlled account. The proposed amnesty unit would then only report monthly on reconciliated levy payments without physically handling the levy funds.

Problems with Levy Collections

60. In managing the levy payments, multiple practical problems can arise which can be managed by the proposed Amnesty Unit. Practical issues arise in VD programs that require discretionary powers by the Unit to grant extensions for levy payments of which the following are common: (1) applicants passing away between the time of the submission of their applications and the adjudication of such applications, resulting in the blocking of such assets for probate purposes; (2) fixed investments that could not be withdrawn within the period permitted for the payment of the levy; (3) executors of deceased estates that could not access foreign assets owing to insufficient provisions in domestic wills to deal with foreign assets; (4) assets frozen abroad owing to infrequent previous use by account holders; (5) disputes may arise about ownership of foreign assets in divorce proceedings instituted after amnesty applications had been submitted, but before applications were adjudicated; (6) illiquid assets (e.g., real estate) held abroad; (7) incompetence of representatives of amnesty applicants, certain foreign banks in transferring funds to say Ukrainian banks, or domestic authorized banks, or exchange dealers; and (8) withdrawal conditions imposed by foreign banks, inter alia, in terms of disclosure legislation in foreign jurisdictions. Also, rules should exist for refunding levy overpayments.

Public Awareness and Reporting Obligations

61. A successful VD program is built on public awareness that needs to be raised continuously. This can be done by means of media coverage of the Unit’s activities and the opportunity it provides to explain amnesty procedures during the period that people could submit applications (e.g., frequently asked questions in a weekend newspaper and participation in radio and TV programs). The Unit should also prepare monthly reports on updated statistics on progress with adjudication of applications; unauthorized assets disclosure; levy imposition; and the amount of receipts and accruals disclosed from foreign sources. Of similar importance is the Unit’s responsibility to manage the reporting requirements in terms of financial intelligence legislation, in so far as assistance had been rendered with the clarification, completion or submission of amnesty applications. Exemptions granted to amnesty applicants may need to be explained to foreign jurisdictions as developed economies have increased compliance requirements for financial institutions.

The Issue of Facilitators

62. A matter not yet addressed in the Ukrainian VD program, but potentially quite important for securing a successful VD and tax amnesty, is the role of tax practitioners and facilitators. The Ukrainian concept amnesty does not generally apply to advisors and facilitators that merely assisted applicants in violating exchange control regulations and related tax acts. However, experience in other international amnesty programs suggests that advisors and facilitators need some level of protection. It is likely that advisors and facilitators may attempt to dissuade applicants from coming forward out of fear that an applicant’s request for amnesty will lead to the prosecution of these advisors and facilitators.

63. The authorities reviewing the VD applications should not force an amnesty applicant to disclose the identity of any party that assisted in a violation. The amnesty unit should erase all names of parties inadvertently revealed by the applicant on an application form. With this level of protection, no reason exists to generally extend the amnesty for advisors and facilitators as no names will be requested and no names should be revealed. However, the only extension of the amnesty could involve a limited class of facilitators, which are parties who may have physically assisted in the violation. These facilitators include individuals (such as trustees and employees), wholly owned companies and trusts that illegally held or accumulated foreign assets on the applicant’s behalf. These related parties need an extra level of protection because the SFS and NBU will automatically have an investigation trail to these related parties once the applicant discloses the need for amnesty. One option is that such limited group of facilitators may apply for relief by adding their names to an applicant’s application for amnesty. In other words, a facilitator may only come forward if the related applicant files an amnesty application.

Recommendations

  • Impose only the 9 percent VD Fee for disclosing foreign-held assets but keeping them offshore and incentivize the foreign asset disclosure and repatriation of assets with a 5 percent VD Fee.

  • Do not grant a zero percent Fee for repatriated assets and reinvested in government securities.

  • Fully align the VD and asset repatriation program with AML and CTF rules.

  • Ex ante make sure that there is no need for a follow-up VD, tax amnesty and asset repatriation program as repeated amnesties have a deleterious effect on long-term tax compliance behavior.

  • Cover facilitators (in the guise of legal persons) under the same amnesty so that tax avoidance structures and mechanisms can be revealed.

  • Establish a special central amnesty administrative unit comprised of SFS and NBU staff who can adjudicate applications and protect taxpayers by erasing their application record if not successful in getting immunity for exchange control and tax violations.

  • Extend the amnesty period to 10 or 12 months as due diligence and verification of asset registers is time consuming, especially since it involves data sharing and coordination with competent authorities in other tax jurisdictions.

  • Extend the VD, tax amnesty, or asset repatriation to trust beneficiaries and heirs of an estate.

  • The proposed amnesty unit must prepare monthly statistical reports to the Minister of Finance and Rada on the uptake of the VD program with associated tax amnesty, and asset repatriation.

III. BEPS Implementation Package

A. Introduction

64. Since the BEPS project released its final reports in 2015, Ukraine has been interested in updating its international taxation system according to the new global standards. The FAD (2017) report recommended to adopt the minimum standards (Combating Harmful Tax Practices – Action 5; Countering Treaty Shopping – Action 6; Country-by-Country Reporting (CbC) – Action 13; and Mutual Agreement Procedures – Action 14), acknowledging that other actions also could potentially benefit Ukraine in protecting its tax base.

65. The planned liberalization of exchange controls has triggered an interest in adopting BEPS actions beyond the minimum standards. Residents of Ukraine will be able to transfer increasing amounts of capital abroad, and the MoF and the NBU believe this should be accompanied with CFC rules (Action 3) to tax passive income earned abroad by residents of Ukraine. Capital outflows could also encourage increased indebtedness by resident entities, so strengthening restrictions to the interest deductions (Action 4) may be justified at this time.

66. The MoF and the NBU developed draft laws for adopting seven BEPS actions and have shared the texts with FAD for their analysis. These actions are: 3: CFCs; 4: Interest deductions; 6: countering treaty abuse; 7: defining permanent establishment; 8–10: transfer pricing; 13: CbC reporting; 14: mutual agreement procedures. The respective draft laws are discussed below in that order.35

B. BEPS Action 3 – Controlled Foreign Company (CFC) Rules

67. The Final Report of BEPS Action 3 recommended that countries introduce CFC rules to counter deferral of tax through taxpayers shifting passive income to non-resident companies that they control. In the present context, deferral means the postponement of current taxation in Ukraine of foreign income that has economically accrued to a Ukraine resident taxpayer through their controlling interest in a foreign company. In the absence of CFC rules, foreign income derived by a foreign company is not subject to Ukraine tax until the foreign company distributes the income as a dividend to a Ukraine resident shareholder or the shareholder sells their interest in the foreign company. The longer the period of deferral, the greater the tax benefit. The benefit of deferral creates a bias in favor of Ukraine residents shifting passive investments to CFCs resident in low tax jurisdictions.

68. Ukraine proposes the introduction of CFC rules to counter deferral. In broad terms, CFC rules treat the passive income of a CFC as the income of the Ukraine resident who controls the CFC. Consequently, CFC rules advance the taxing point from the time of dividend to the time when the CFC derives the passive income thereby eliminating deferral. The draft CFC rules are in proposed Article 39 of the TC. Article 39 is comprehensive in coverage and provides for all the key design features of the CFC rules as recommended in the BEPS Action 3 Report. The discussion below identifies some technical areas where the proposed rules could be improved.

Relevant Taxpayer

69. The relevant taxpayer under the CFC legislation is a resident individual who is the ultimate beneficial owner and controller of a foreign company (Article 39.1.2). The ultimate beneficial owner is determined according to the money laundering legislation. The mission was advised that the reason for applying the CFC rules only to resident individuals is that, in approximately 95 percent of cases, the person holding the shares in a foreign company is a resident individual.

70. While, currently, interests in foreign companies may be held directly by resident individuals, this is likely to change with the introduction of the CFC rules. It can be expected that resident individuals will put into place more complex domestic holding structures in an attempt to disguise their ultimate beneficial ownership of a foreign company. Consequently, tracing back to the ultimate beneficial owner of a CFC may not only add a potential layer complexity, but also give rise to an avoidance risk.

71. Further, applying the CFC rules to the ultimate beneficial owner of a foreign company may pose a technical problem in relation to dividends paid out of CFC-taxed income. If the shares in the foreign company are held by a resident entity, then such a dividend is paid to that entity. This gives rise to a potential mismatch between the resident entity receiving the dividend and the resident individual previously taxed on the CFC income out of which the dividend has been paid.

72. It is preferable that the CFC rules apply at the level of the first “onshore” resident person (resident individual or resident entity) in the chain of ownership of a foreign company. This will better protect the integrity of the rules and avoid possible mismatches in relation to dividends paid out CFC-taxed income. It also better aligns with international norms.

Foreign Entities Subject to CFC Rules

73. The primary target of CFC rules are foreign companies as they are legal persons separate from their shareholders and, therefore, can be used as a deferral vehicle. However, CFC rules should apply to any foreign entity that is treated as a separate taxpayer under the TC, i.e., any foreign entity that can be used to defer Ukrainian tax. In this respect, Article 39.1.1 applies the CFC rules to any “foreign arrangement without legal personality”, which is intended as a reference to partnerships, trusts, foundations, and the like. This is good policy and will prevent the CFC rules being avoided through the use of non-corporate entities.

74. Importantly, though, it is only necessary to apply the CFC rules to foreign entities that are taxed as separate entities under the TC. It is not necessary to apply the CFC rules to a foreign entity that is taxed as a “look through” under the TC because such an entity cannot be used as a deferral vehicle. In this regard, it is understood that partnerships are taxed as separate entities under the TC so it is appropriate to apply the CFC rules to foreign partnerships.

Control and Substantial Shareholder Rules

75. The CFC rules apply only to foreign companies that are controlled by resident individuals. The control rule is the most important aspect of CFC rules because, if there is no Ukrainian control of a foreign company, the CFC rules do not apply. It can be expected, therefore, that the control rule will be the subject of tax planning by Ukrainian residents to avoid the operation of the CFC rules. Consequently, the integrity of the CFC rules depends very much on the definition of “control”. Consistent with CFC rules in other countries, the control rule is comprehensively defined in Article 39.1. The control rule is stated separately for foreign companies and foreign arrangements without legal personality.

Foreign Companies

76. There are three alternative tests of control for foreign companies. A foreign company is a CFC if one of the following applies (Article 39.1.2):

  • (1) A single resident individual has a 50 percent or greater ownership interest in a foreign company.

  • (2) Resident individuals have, in aggregate, 50 percent or greater ownership interest in a foreign company. In this case, a substantial shareholder rule applies so that only those resident individuals that have a 25 percent or greater ownership interest are subject to tax under the CFC rules.

  • (3) A resident individual individually, or with related resident persons, exercises de facto control over the foreign company.

77. The calculation of the ownership interest of a resident individual in a foreign company is based on both voting and value rights (i.e., rights to dividends and return of capital). If there is a difference in those rights, then the higher percentage interest applies (Article 39.1.3). While control of a company is normally determined by reference to voting rights, a resident individual could arrange for the voting rights to be held by a “friendly” offshore person (such as a bank) and retain the value rights in the company. Such arrangements are avoided by determining the level of ownership interest of a person by reference to both voting and value rights.

78. Further, the ownership interest of a resident individual in a foreign company includes the ownership interest in the company of related persons (Article 39.1.4). The related person may be a resident or non-resident. The inclusion of the interests of related persons prevents the fragmentation of ownership of a foreign company among related persons. For example, a resident individual could arrange for the ownership interest in a foreign company to be held equally by the individual and four relatives. While there is 100 percent control of the foreign company by resident persons, there is no substantial shareholder as each person has only a 20 percent ownership interest. In this example, the effect of the anti-fragmentation rule is that there is a single ownership interest of a 100 percent.

79. Tracing rules apply where the first foreign company in the corporate chain is simply a holding company with a foreign company lower down the corporate chain holding the passive assets that derive CFC income (Article 39.1.4). However, under the rules, it is provided that tracing is based on a simple multiplication of interests. This will limit the effectiveness of the tracing rules. For example, a resident individual may have a 50 percent ownership interest in Foreign Company 1 that in turn has a 50 percent ownership interest in Foreign Company 2. The other 50 percent ownership interest in both Foreign Company 1 and Foreign company 2 is with unrelated non-residents. In this example, there is control (50 percent) at each tier, but the multiplication of interests means that the resident individual has only a 25 percent indirect interest in Foreign Company 2 and, therefore, Foreign Company 2 is not a CFC.

80. Tracing should be based on the assumption that, if there is control at a particular tier in the chain of companies, the tracing percentage is 100 percent. This is the position with de facto control (Article 39.1.4), but it should apply generally in calculating indirect ownership interests. In the example above, this would mean that Foreign Company 2 is correctly treated as a CFC.

81. Importantly, with indirect interests, there are two measurement concepts. First, there is the measurement of control interests. As stated above, if there is control at a particular tier, then the control interest should be deemed to be 100 percent. Second, once control is established, there is the measurement of the resident individual’s share of CFC income. This should be based on a straight multiplication of interests to reflect the other ownership interests in the foreign company. Consequently, in the example above, Foreign Company 2 is a CFC based on tracing control interests, but the resident individual is taxed only on 25 percent of the CFC income of Foreign Company 2.

82. The substantial shareholder threshold under control test (2) above is 25 percent. It is understood that this is based on an alignment of domestic thresholds (including for related persons) with the participation dividend threshold under Article 10(2) of tax treaties. While there is a degree of simplicity in using the same threshold for different purposes under the TC, the threshold should also be consistent with policy considerations. The substantial shareholder rule is intended to limit the application of CFC rules to those resident individuals who have access to sufficient financial information concerning the CFC so as to be able to calculate CFC income. The 25 percent threshold could be reduced to 10 percent to align with the usual commercial threshold between direct and portfolio investment. Given that CFC income is calculated by reference to the financial accounts of the CFC, a resident individual with a 10 percent or greater ownership interest should have access to sufficient information to do the calculation.

83. The de facto control test is based on economic control exercised individually by a resident individual or collectively with resident related persons. The de facto control test should take account of all related persons, resident or non-resident. This is the case with the calculation of ownership interests and the same should apply for de facto control.

84. The basic principle of de facto control is that a resident individual can exercise significant influence over the decision making by a foreign company regardless of the legal rights (if any) that the individual has in the company (Article 30.1.6). There are five circumstances specified where a person is treated as having de facto control over a company. While the listed circumstances capture the main examples de facto control, it should be clearly stated that the list is not exhaustive of all situations that may involve de facto control.

85. It is provided that an individual can declare to be a person exercising de facto control over a company (Article 39.1.6). The reference to “individual” is not limited to a resident individual. Hence, non-resident individuals could declare themselves to have de facto control with the result that the foreign company is not a CFC. Even if the making of the declaration is limited to resident individuals, it is possible that an individual may declare themselves to be the de facto controller but not have the financial resources to pay the tax on CFC income. It is better to retain SFS flexibility to determine who is the de facto controller of a CFC.

Foreign Arrangements Without Legal Personality

86. Foreign trusts can be structured so that income accumulates in the trust without anyone being immediately entitled to the income and, therefore, a foreign trust can be used as a deferral vehicle. Because it is unclear who will ultimately benefit from the income accumulated in the trust, it is usual for CFC rules to tax the settlor of the trust. Such trusts are usually set up as a “blind” trust under which the trustee has complete control over the assets of the trust. The settlor usually has no control over the trust and may not even know what the assets of the trust are. Despite the lack of control over the trust, it is usual to apply the CFC rules to the settlor of the trust on the assumption that the settlor is unlikely to have transferred substantial funds to the trustee in settlement of the foreign trust without the expectation that either the settlor or persons close to the settlor will ultimately benefit from the trust.

87. A settlor (i.e., creator) of a foreign trust is effectively treated as the controller of the trust (Article 39.1.5). However, there are a number of exceptions to this (Article 39.1.5.1), including where: (1) the settlor is not able to direct the distribution of the profit of the trust; (2) the settlor does not have a reversionary interest in the assets of the trust; or (3) the settlor does not exercise de facto control over the trust. The exceptions will, in effect, mean that the settlor of a blind trust is unlikely to be treated as the controller of the trust and, therefore, CFC rules will not apply. It is best practice to treat a resident individual who is the settlor of the trust as the controller of the trust in all cases.

Calculation of CFC Income

88. The adjusted taxable profit of a CFC is included in the personal income of a resident individual controller of the CFC based on their percentage interest in the CFC. The percentage interest of a de facto controller is deemed to be 100 percent and the same rule should apply to a resident individual who is the settlor of a foreign trust. CFC income is taxed at the standard PIT rate of 18 percent.

89. The adjusted taxable profit of a CFC is the pre-tax financial accounting profit of the CFC based on audited accounts, but subject to certain modifications. This essentially aligns with the taxable profit calculation under the TC which is based largely on financial accounting profit. There are some adjustments in calculating CFC income, including the add-back of impairment write downs and limiting the deduction of interest expense to 30 percent of EBITDA.

90. The resident controller of the CFC is entitled to a credit for any foreign tax paid by the CFC in relation to CFC income (Article 170.13.2.4). This is consistent with CFC income being calculated on a pre-tax basis. The amount of the tax credit is proportionate to the controller’s share of CFC income. There are a number of exemptions applicable to the calculation of adjusted taxable profit. The main exemptions are discussed below.

Comparable Tax Exemption

91. Adjusted taxable profit does not include comparably taxed income (Article 39.4.1). If the income of a CFC has been comparably taxed, then there is no deferral of residence tax.36 There are two bases for determining whether CFC income has been comparably taxed:

  • (1) There is a tax treaty or an information exchange agreement between Ukraine and the country of residence of the CFC. This exception should apply only when the country of residence does effectively exchange information with Ukraine (i.e., the existence of an information exchange arrangement should not, of itself, support the exemption).

  • (2) The country of residence of the CFC is not a listed jurisdiction (black list) and the CFC has paid an effective tax rate on its income that is not lower than five percentage points below the nominal rate of corporation tax in Ukraine. Based on the current rate of corporation tax, the effective rate of tax paid by the CFC must not be lower than 13 percent. The effective rate of tax is calculated as the tax paid to the CFC’s country of residence as a percentage of pre-tax financial accounting profit.

92. The use of a jurisdictions list for the purposes of the comparable tax exemption is a common simplification measure. However, the risk in using a “black” list is that every country not in the list is treated as a comparable tax country, which may not necessarily be the case. It is positive, therefore, that the black list is supplemented by a differential effective tax rate approach, as described above (par 92(2)).

93. Territorial tax countries can be used to “disguise” tax haven activity. A CFC can be established in a territorial tax country with PE (passive) operations in a tax haven. As the PE’s income is foreign income, it is not subject to tax in the CFC’s country of residence based on territoriality. It would be prudent, therefore, to include territorial taxation on the black list even though, in a particular case, this may be captured under the effective tax rate approach.

Active Income Exemption

94. There is an exemption for active income (Article 39.4.1). This is consistent with the international norm that CFC rules target passive income in low tax jurisdictions. The active income exemption applies on an entity basis. If the share of the passive income of a CFC as a percentage of its total income for a year is less than 50 percent, the CFC passes the active income test and there is no CFC taxation of any of its income (passive or active) for the year. Similarly, if the share of the passive income of a CFC as percentage of its total income for a year is 50 percent or more, the CFC fails the active income test and there is CFC taxation of all its income (passive or active) for the year.

95. The important issue with the active income exemption is the definition of “passive income”. The definition should include the standard items of passive income, such as dividends, interest, royalties, rents, annuities, and capital gains on the sale of passive assets. In particular, the definition of “interest” needs to be broadly stated to include any form of return under a financial instrument and not be limited to interest payable on a debt instrument.

96. It is usual to have an exception for “active” passive income. For example, if a CFC has genuinely developed the industrial or intellectual property right in respect of which a royalty is paid to the CFC, then the royalty should be treated as active income. The same applies also to interest derived by a bank.

De Minimis Exemption

97. A de minimis exception applies where the total revenue of a CFC for a year does not exceed EUR 1 million (Article 39.4.2.1) Where a resident individual is the controller of more than one CFC, then the EUR 1 million threshold is tested by reference to the total revenue of all CFCs. It is common for there to be a de minimis exception under CFC rules as a simplification measure and it is good design to apply it to all CFCs to prevent the splitting of passive assets among several CFCs.

Public Company Exemption

98. There is an exception for a CFC that is a public company listed on a recognized stock exchange (Article 39.4.2.2). Again, this is a standard exemption under CFC rules and is based on the assumption that a publicly listed company is unlikely to be a deferral vehicle because it would be expected to regularly pay dividends to meet the demands of shareholders. Given the 25 percent substantial shareholder threshold, it is likely that this exemption would rarely apply. However, if the substantial shareholder threshold is reduced to 10 percent as suggested above, this exemption may have a greater role to play.

Dividends Paid by CFCs

99. The treatment of dividends needs to be considered in two cases: (1) a dividend paid by a CFC to another CFC; and (2) a dividend paid by a CFC to the resident individual controller. A dividend paid by a CFC to another CFC of the same controller is, in effect, treated as exempt income of the recipient CFC (Article 39.3.2.5). This avoids double CFC taxation on profits as they pass through a chain of CFCs of the same controller.

100. The taxation of a dividend paid by a CFC to the resident individual controller of the CFC depends on the timing of the dividend payment. For an interim dividend received by the controller in a Ukraine bank account, the CFC income supporting the dividend is taxed at 9 percent (rather than 18 percent). Similarly, a dividend that is paid by a CFC to a Ukraine bank account of a resident controller by the end of the calendar year following the reporting period is also taxed at 9 percent. In this case, the resident controller can file an adjusted tax return to give effect to the reduction in the tax rate on CFC income from 18 percent to 9 percent. A dividend paid out of CFC income outside of this time period is not subject to further tax (i.e., the 18 percent tax on CFC income applies).

101. The policy underlying the 9 percent rate is to encourage repatriation of CFC income. The 9 percent rate applies only where the dividend is paid into a Ukrainian bank account. It is noted, though, that the 9 percent rate may have the opposite effect, namely to encourage passive investments offshore. For example, a resident individual who invests in debt instruments issued in Ukraine is subject to tax at 18 percent on the interest paid on those instruments. If, instead, the individual uses the funds to establish a CFC, which invests the funds in debt instruments in a country that does not tax interest (such as Hong Kong), the CFC rules apply and, initially, the interest income of the CFC is taxed to the resident individual at the 18 percent rate. However, if the CFC repatriates the interest income as a dividend within two years, the tax rate is reduced from 18 percent to 9 percent (assuming no dividend withholding tax applies to the CFC). It is possible, therefore, that the repatriation incentive may simply encourage resident individuals to make passive investments offshore through a CFC rather than locally.

102. The way that the exemptions are applied means that all the income of a CFC for a tax year will either be fully taxed CFC income or non-taxed CFC income. For example, the adjusted taxable profit of a CFC that fails the active income test for a tax year includes all the income of the CFC (passive and active) for that year. Similarly, a CFC that passes the active income test for a tax year has no CFC income for that year. However, a CFC may fail the active income test in one year but pass the test in another year. This means that a CFC may have both CFC taxed-income and non-taxed income. If a dividend is paid outside the two-year period for the 9 percent rate to apply, it will be necessary to know which profits support the dividend. If it is the CFC-taxed profits, then the dividend is exempt from tax. If it is the non-taxed profits, then the dividend is taxable at the 18 percent rate.

103. There does not appear to be any dividend ordering rules to apply where a CFC has both CFC taxed and non-taxed profits, although the mission was advised that the CFC could nominate the profits that support the dividend. It is accepted that this is one methodology that could be used. However, there are other possible methodologies, such as: (1) first out of CFC taxed income; (2) first out of non-taxed income; or (3) pro-rata out of CFC taxed and non-taxed income. It is important that there is a clear rule applied for the ordering of profits supporting a dividend. Similar issues arise in relation to the taxation of gains on disposal of shares in a CFC.

Recommendations

  • Apply CFC rules at the level of the first “onshore” person in Ukraine.

  • Only treat a foreign entity as a CFC if it is not taxed as a look through entity in Ukraine.

  • Tracing should be based on the assumption that, if there is control at a particular tier in the chain of companies, the tracing percentage is 100 percent.

  • Consider reducing the substantial shareholder threshold to 10 percent.

  • Apply the de facto control test by reference to all related persons and not just resident persons.

  • Limit or delete the rule that permits an individual to declare themselves to be a person exercising de facto control over a company.

  • Treat a settlor of a foreign trust as the controller of the trust in all cases.

  • Out of abundant caution, include territorial taxation in the list of situations that would have the effect of a LTJ (as in Australia).

  • Ensure that there is a comprehensive definition of “passive income”, including a definition of “interest”. Exclude obvious examples of “active” passive income.

  • Delete the 9 percent rate for repatriated CFC income.

  • Include dividend ordering rules to apply where a CFC has both CFC-taxed and non-taxed income.

C. Limitation on Interest Deductions (Action 4)

The Earnings Before Interest, Tax, Depreciation and Amortization (EBITDA) Rule

104. Ukraine has had in the books for some time a relatively weak thin cap rule, which has been discussed in previous FAD reports.37 The current initiative to strengthen the rule following BEPS Action 4 report38 is a positive measure. The proposal would replace the existing rule39 by limiting interest deductions to 30 percent of EBITDA (the ‘EBITDA rule’).40 The excess interest payments up to the limit set by the Arm’s Length Principle (ALP) could be carried forward indefinitely (explained below), but with a yearly reduction of that excess of five percent.

105. However, there are different approaches to the design of an EBITDA rule, with varying degrees of complexity. The approach proposed by the MoF is rather complex. The mission encourages the MoF to consider a simpler version as discussed in this section. Nonetheless, the rule proposed by the MoF would correct several deficiencies of the current regime. Namely, the proposal would:

  • Apply the limitation to all interest payments, to non-resident and resident lenders, or to independent or related parties. Currently, the thin cap rule applies only to non-resident related parties, which can be avoided, for example, through back-to-back loans;41

  • Eliminate the debt to equity ratio of 3.5 to 1 as a precondition to limiting the deduction of interest payments, which was an unnecessary qualifier and the debt ratio was too high in any case to be significant;

  • Reduce the limit to qualifying interest deductions from 50 to 30 percent of EBITDA; and

  • Eliminate the ad hoc definition of related parties for the sole purpose of applying the thin cap rule (50 percent participation in equity – TC Article 141.2, instead of 20 percent in the general rule – TC, Article 14.1.159).

Interaction with the Arm’s Length Principle (ALP)

106. According to the MoF, the proposal would also improve the current provision by clarifying how the interest deduction limitation would interact with the ALP. The draft proposal states that, if the ALP allows for an interest payment (to a related party) higher than 30 percent of EBITDA, the interest limitation would apply anyway. The proposal is not explicit however about how the interaction with the ALP should be understood in the opposite situation, when the ALP sets an interest payment lower than the EBITDA rule, but it can be inferred that the intent is that the ALP would apply in this case.

107. There is more than one way to understand how the ALP can interact with a limitation to interest deductions. On the one hand, it could be argued that the MoF’s proposed asymmetric approach is fair because the EBITDA rule is designed to stop abusive deductions only, without qualifying whether payments below it are necessarily correct. Thus, in such circumstances the ALP would be binding for taxpayers. On the other hand, the EBITDA rule controls interest deductions resulting both from the amount of debt and the applicable rate, limiting simultaneously thin capitalization and transfer pricing issues. Accordingly, once an EBITDA rule is in place, it could be argued, arm’s length (AL) considerations are redundant. This approach is evidently much simpler (although it may sacrifice some tax base).

108. The proposed EBITDA rule also interacts with the ALP on the carry-forward provision, which further complicates the regime. According to Article 14.1.49 of the proposed reform, all non-AL payments to non-residents should be recharacterized as dividends, meaning that they are not deductible and subject to the corresponding withholding. This implies that excess interest payments (above the EBITDA threshold) can be carried-forward only up to the AL amount; payments above this level would have been recharacterized as dividend and thus not carried-forward. Alternatively, whenever the AL interest is equal or below the EBITDA threshold, none of the interest paid above this threshold can be carried-forward. Thus, the possibility of carrying-forward (some, if any) excess interest payments comes at the cost of calculating the AL interest level every year.

Net vs Gross Interest and the Financial Sector

109. The proposed rule applies to (gross) interest expenditure. Such deduction limitation is considerably stricter and generally more protective against abuse than one on net interest expenditure (i.e., interest expenditure reduced by interest income). But it should not apply to the financial sector, where holding financial liabilities is an essential element of the business model and EBITDA is not a suitable measure of economic activity in that sector.42 If the MoF is committed to the stricter version of the rule, it should exempt the financial sector.43 Applying the deduction limit to net interest expense, generally more lenient as an anti-abuse measure, would be mostly irrelevant for the financial sector given that banking and insurance institutions normally have a net interest income.44 Adopting this version of the rule may not require an exemption for the financial sector.

110. There is no internationally agreed standard for limiting interest deductions in the financial sector. But admittedly there are risks of aggressive tax planning through interest deductions in the sector. A cautious approach would be to exempt the financial sector from the EBITDA rule and assess specifically the base erosion problems faced by this sector in Ukraine and develop tailored made measures to address them.45

Policy Options

111. Simplicity is generally preferable. The MoF proposal is complex, for it requires to assess the AL level of interest payments that are deductible, regardless of the interest deduction limitation rule. The AL level of interest is also necessary to determine how much interest can be carried forward. One option is to apply only the EBITDA rule based on net interest expense, eliminate the carry-forward, and possibly lowering the threshold for the EBITDA rule (proposed at 30 percent); the BEPS Action 4 report suggestion ranges from 10 to 30 percent. However, given the MoF’s reluctance to adopt a lower threshold and its concerns that the AL interest payment could in some cases be below it, another option is to disallow the deduction of all net interest payments above the AL level when this is below the threshold and eliminate the carry-forward for interests that are recharacterized as dividends.

Unsolved Problem with Interest Deductions

112. Interest is deductible when accrued but withholding applies only when it is paid. This asymmetric treatment of interest paid to foreign residents allows for tax planning and it is not solved by the EBITDA rule. Income sourced in Ukraine is deducted (by a resident entity) but remains untaxed. The party abroad could even be exempt if its residence country has a territorial regime and benefit from double non-taxation (and possibly not affected by the Ukrainian regulation on LTJ). The EBITDA rule could be supplemented by another measure requiring that that no deduction be allowed for an expenditure subject to withholding until withholding tax is paid (as in Australia).

Recommendations

  • Replace current thin cap rule for the EBITDA rule proposed by the MoF, considering the following adjustments:

    • o Eliminate carry-forward of non-deducted interests.

    • o Apply EBITDA rule to net interest expense.

    • o Exempt the financial sector from the rule.

  • Defer deduction of accrued interest subject to withholding until withholding tax is paid.

D. Anti-Treaty Abuse Measures (Action 6)

113. BEPS Action 6 deals with abuse of tax treaties. The Final Report on BEPS Action 6 identified treaty abuse as a major base erosion issue. While the main example of treaty abuse is treaty shopping, the BEPS Action 6 Final Report identified a number of arrangements that can give rise to treaty abuse. For existing treaties, the Multilateral Instrument (MLI) is the instrument developed under BEPS Action 15 for the implementation of anti-treaty abuse measures in existing treaties.

114. It is proposed to amend the TC to include an anti-treaty shopping rule and to provide for the taxation of indirect transfers of Ukrainian immovable property as provided for in Article 13(4) of the OECD and UN Models. There is an important difference in the nature of these two amendments. The anti-treaty shopping rule is, in effect, a domestic law provision that overrides tax treaties. Whereas the taxation of indirect transfers under Article 13(4) is a jurisdictional issue that needs to be replicated in the TC to be effective, as discussed below.

Treaty Shopping

115. The inclusion of an anti-treaty shopping rule in tax treaties is a minimum standard under the MLI. Based on BEPS Action 6, the MLI specifies two methodologies for dealing with treaty shopping: (1) the principal purpose test (PPT); and (2) a limitation of benefits (LOB) rule. A signatory to the MLI can use either the PPT or LOB approaches, or both.

116. Under the PPT, a benefit under a tax treaty is denied if it is reasonable to conclude that obtaining the benefit was one of the principal purposes of an arrangement that resulted directly or indirectly in that benefit. The reference to “reasonable to conclude” implies that the purpose of a person in entering into an arrangement is determined objectively having regard to all the facts and circumstances. The onus is on the SFS to establish that a principal purpose of a person was to obtain treaty benefits.

117. Under the LOB rule, treaty benefits are limited to qualified persons. In broad terms, a qualified person is a genuine resident of a Contracting State (such as a resident individual). Generally, an entity is a qualified person of a Contracting State only if resident individuals hold, directly or indirectly, 50 percent or more of the voting or value rights in the entity. There is an exception where an entity carries on an active business in its country of residence and the payments for which treaty benefits are claimed are sufficiently connected to that business. The business of making or managing investments (i.e., holding passive investments) is not an active business. Consequently, a company that is established in a Contracting State that is beneficially owned by non-resident individuals will not be a qualified person unless it carries on an active business in that Contracting State.

118. The mission was advised that the MoF is proposing to implement the PPT approach because of concerns about the complexity of the LOB approach. While a LOB Article is more detailed than an Article based on the PPT, an LOB Article has greater certainty and, therefore, is considered simpler to administer. In particular, the SFS should be able to readily access the information necessary to determine whether a resident of the other Contracting State is a qualified person entitled to treaty benefits. While the PPT has the advantage of flexibility, it is less certain than the LOB rule as it requires the determination of the purpose of a person in relation to a transaction or arrangement. Further, it is possible that the two Contracting States may form different views as to the relevant purpose. On balance, it is considered that the LOB approach is the better approach for Ukraine in countering treaty shopping. However, both approaches are acceptable within the MLI framework.

119. While Ukraine has signed the MLI and proposes to include the PPT approach in its existing treaties covered by the MLI, this process may take some time. As an interim measure, it is proposed that the PPT is incorporated into Article 103.2 of the TC as a treaty override rule. A treaty override is effective only if, constitutionally, treaties and the TC are of equal status. If there is a constitutional priority in favor of treaties, then a proposed treaty override rule will have no effect. The assumption underlying Article 3 of the TC is that treaties and the TC are of equal status (otherwise Article 3 would not be necessary). However, Article 3 of the TC provides for a legislative rule under which a treaty has priority over the TC in the case of conflict. Article 3 needs to be amended to provide that the PPT in Article 103.2 is an exception to the legislative priority in favor of tax treaties.

120. It should be made clear that the PPT rule to be included in Article 103.2 of the TC is an interim measure only. It should cease to apply to an existing treaty once it is agreed with the other Contracting State that the MLI applies to the treaty. Further, the PPT rule in Article 103.2 of the TC should not apply to future treaties as the entitlement to benefits Article will be part of future treaty negotiations.

Indirect Transfers of Immovable Property46

121. Currently, the TC provides for taxation of the income derived by a non-resident from the sale of immovable property located in Ukraine (Article 141.1.1(f)). This taxing right can be avoided through an indirect transfer of the immovable property. This involves a nonresident, directly or indirectly, selling the shares or other interest in the entity holding the immovable property rather than the entity selling the property. Given that the value of an interest in an entity equates to the value of the assets of the entity, it would be relatively easy to use indirect transfers of immovable property to avoid Ukrainian tax on the disposal of immovable property. Consequently, the taxation of indirect transfers is a measure intended to protect the taxing right applicable to gains arising from the disposal of Ukrainian immovable property.

122. It is proposed to treat a gain made by a non-resident on an indirect transfer of Ukrainian immovable property as Ukraine-source income (Article 141.1.1(g)). The rule applies only to the sale of an interest in a foreign legal person. This is presumably because a gain on the sale of an interest in a resident legal person (regardless of its assets) is already treated as Ukraine source income (Article 141.1.1(f)).

123. There are some technical issues with the design of the indirect transfer rule. First, there should be a broad definition of “immovable property” for the purposes of the rule. In particular, the rule should apply to mining and petroleum rights relating to land in Ukraine. This may be covered by the general law (or the tax law) defining the meaning of immovable property, It is noted that some countries have experienced planning under which the holder of a mining or petroleum right allocates substantial value to mining and petroleum information (rather than to the mining or petroleum right) and then argues that the 50 percent threshold is not satisfied (local asset value of immovable property as a proportion of total value of all assets indirectly transferred offshore). This has been countered by including mining and petroleum information in the definition of immovable property. It is important that the definition of immovable property is clarified in the TC because it is also the treaty definition of immovable property (Article 6(2) of tax treaties).

124. Second, the rule applies only where the Ukraine immovable property is held by a resident legal entity. It is important that the rule applies in relation to both resident and nonresident legal entities that hold Ukraine immovable property otherwise the rule will be easily avoided through a non-resident holding the immovable property. Even if there are limitations on non-residents owning land in Ukraine, the definition of immovable property will include other interests (such as mining or petroleum rights) that may be owned by non-residents.

125. Third, in broad terms, the indirect transfer rule applies where a foreign legal entity sells an interest in another foreign legal entity that, directly or indirectly, holds shares in a Ukraine legal entity that owns Ukraine immovable property. There are two thresholds that must be satisfied before the rule applies. First, 50 percent or more of the value of the interest in the foreign entity that is sold is derived, directly or indirectly, from an interest in a Ukraine legal entity. Second, 50 percent or more of the value of the interest in the Ukraine legal entity is derived from Ukrainian immovable property. The inclusion of the first threshold is a departure from Article 13(4) and it may unnecessarily narrow the scope of the rule. It is recommended that the drafting of Article 14.1.54(j) is closely aligned to the drafting of Article 13(4) of the OECD and UN Models.

126. Finally, the rule may be avoided by, immediately before the indirect transfer, shifting other assets into the entity that owns the Ukraine immovable property, so as to fall below the 50 percent immovable property threshold. The revised Article 13(4) of the OECD and UN Models provides that the rule applies where the 50 percent threshold is established at any time in the period of 365 days prior to the indirect transfer. This should also be provided for in Article 14.1.54(j) of the TC.

127. An issue arises as to how to collect Ukraine tax on an indirect transfer as it is likely to be a transaction between two non-residents involving the sale of an interest in a nonresident entity. It is proposed to collect the tax on indirect transfers by withholding with the non-resident purchaser required to withhold tax at 15 percent from the profit on disposal. There are two difficulties with this. First, the purchaser will know the sales price but not the seller’s profit as that would require knowledge of the seller’s cost. Secondly, and more importantly, given that the transaction is between two non-residents, it is not clear how the withholding tax liability can be enforced.

128. There are two options for the collection of tax on indirect transfers. First, the entity that owns the Ukrainian immovable property could be treated as a payment agent of the nonresident who has made the gain (as in the MoF’s proposal). Alternatively, the entity that owns the Ukrainian immovable property could be deemed to have made a pro-rata disposal and re-acquisition of the property if there is a 10 percent or greater change in the underlying ownership of the entity.

129. The outcome under the two approaches is essentially the same (namely, the entity that owns the property pays the tax owing by the non-resident). However, there is a difference under the two approaches in relation to the legal liability for the tax. Under the agency approach, the primary liability remains with the non-resident, whereas, under the deemed disposal approach, the primary liability is shifted to the entity that owns the property. Under the agency approach, the non-resident can claim a foreign tax credit (FTC) in their home country for the Ukraine tax paid on the gain from the indirect transfer. Instead, there is likely to be a loss of FTCs under the deemed disposal approach as the liability is shifted to the owner of the immovable property. The agency approach is often preferred as it operates in a way that is consistent with international norms.

Recommendations

  • Use the LOB method to counter treaty shopping abuses.

  • Make clear that the domestic anti-treaty shopping rule is an interim measure applicable only until the MLI is implemented.

  • Align the drafting of the rule for taxation of indirect transfers of Ukrainian immovable property with terms of Article 13(4) of the OECD and UN Models.

  • Include a definition of immovable property that covers mining and petroleum rights, and mining and petroleum information.

  • Collect the tax on a gain on an indirect transfer by treating the owner of the Ukrainian immovable property as the payment agent of the non-resident who has made the gain.

E. Definition of Permanent Establishment (Action 7)

130. The draft BEPS law revises the TC definition of permanent establishment (PE) to align with the definition of PE in the OECD and UN Model Treaties as amended based on the BEPS Action 7 Report (Article 14.1.193). The amended definition of PE in the OECD and UN Models closes some longstanding loopholes in the former definition with the effect of strengthening source taxation of business profits.

Relevance of the PE Concept Under the Tax Code

131. There is an incomplete use of the concept of PE under the TC. While a non resident carrying on business in Ukraine through a PE is treated as a taxpayer (Article 133.2.2), the jurisdictional rules in the TC relating to source of income (Article 14.1.54) do not rely on the concept of PE. Despite this, it is provided that the allocation of business profits to PEs is based on the functionally separate entity approach (Article 141.4.7) (see below). Thus, a Ukraine PE of a non-resident is a taxpayer and the functionally separate entity approach applies in calculating the taxable profit of the PE, but the standard sourcing rules for PEs do not apply under the TC in determining the income taken into account in calculating the taxable profit of a non-resident.

132. It is clear, though, that the concept of PE has an important administrative role in regulating the taxation of non-residents. A non-resident must apply for registration before commencing to carry on business in Ukraine through a PE (Article 64). It is provided that the PE registration requirement applies regardless of whether or not the non-resident is subject to tax in Ukraine. This highlights the disconnect between the concept of PE and the sourcing rules for business profits. The SFS is empowered to undertake an audit of a non-resident that is conducting a business in Ukraine through an unregistered PE (Article 78.1.14).

PE as a Jurisdictional Concept

133. In many countries, the concept of PE is used in tax law as a jurisdictional rule for: (1) the taxation of business profits; and (2) the taxation of interest, royalties, and management and technical fees paid to non-residents. The jurisdictional rule for business profits relates to both the taxation of domestic income of non-residents and the taxation of foreign income of residents. The standard jurisdictional rules relating to PEs are briefly outlined below.

134. There are three contexts in which the concept of PE is normally used in the taxation of non-residents. First, it is usually provided in the tax law that the business profits attributable to a PE of a non-resident are domestic source income. This aligns the sourcing rule under the tax law with Article 7(1) of tax treaties.47 The definition of PE is fundamental to this sourcing rule. In particular, the broader the definition of PE, the broader are the source country taxing rights over business profits.

135. Second, the withholding tax rules for non-residents apply to income that originates from within the jurisdiction. The PE concept is relevant for determining the geographic origin of withholding income. This is usually stated by reference to two sourcing rules:

  • (1) Income paid by a resident of a jurisdiction is sourced in that jurisdiction unless the payment of the income is an expenditure of a foreign PE of the resident.

  • (2) Income paid by a non-resident of a jurisdiction is sourced in that jurisdiction if it is an expenditure of a PE of the non-resident in the jurisdiction.

136. Third, a branch profits tax commonly applies to branches of non-residents so as to equalize the tax treatment of branches and subsidiaries. A “branch” is the commercial term for the tax concept of PE. A branch profits tax is relevant where withholding tax applies to dividends paid to non-residents. In the absence of a branch profits tax, a branch (i.e., PE) is more favorably taxed than a subsidiary.

137. The PE concept is also relevant for the taxation of residents. The jurisdictional issue for residents relates to double taxation relief. Residents are liable for tax on worldwide income but are usually entitled to relief from double taxation in relation to foreign income. The business profit sourcing rule for residents normally provides that all business income is treated as domestic source income except income attributable to a foreign PE of the resident. In other words, business income attributable to a foreign PE of a resident is foreign income and, therefore, qualifies for double taxation relief.

138. These basic jurisdictional rules do not apply in Ukraine. First, Article 14.1.54 states the general principle that Ukraine-source income is any income derived by residents or nonresidents from any activities in Ukraine.48 In the case of non-residents, there is no qualification that Ukraine-source income is limited to income from activities that amount to a PE. Second, interest and royalties paid by a resident person is Ukraine source income under Article 14.1.54(a). This is both over-inclusive as there is no exception for payments that are expenditures of a foreign PE of a resident and under-inclusive as it does not include payments made by a Ukraine PE of a non-resident. Third, there is no branch profits tax in the TC. Fourth, there is no unilateral relief from double taxation under the TC so there does not appear to be any necessity to distinguish between domestic and foreign income of residents.

139. The relevance of PE as a jurisdictional concept in Ukraine appears to be limited largely to the application of tax treaties. Under Article 7 of a tax treaty, Ukraine has jurisdiction to tax the business profits of a resident of the other contracting state that are attributable to a PE in Ukraine. Under Article 23, Ukraine is required to give a credit for tax paid by a resident on the business profits of the resident attributable to a PE in the other Contracting State. Where there is a conflict between a tax treaty and the TC, the tax treaty prevails (Article 3 of the TC). Consequently, if there is a conflict between the broad statement of jurisdiction to tax business income in Article 14.1.54 of the TC with the narrower taxing rights permitted under a tax treaty through the PE concept, the narrower treaty jurisdiction applies.

140. The updating of the definition of PE in the TC should be accompanied by a clearer statement of jurisdictional rules in Article 14.1.54 for the taxation of business income, interest, and royalties based on the concept of PE. This would achieve greater internal consistency in the use of the concept of PE and better align the taxation of non-residents under the TC with taxation under tax treaties.

Treaty and Tax Code Definitions of PE

141. The relationship between the TC and treaty definitions of PE is particularly important where the concept of PE is used as the jurisdictional rule for taxing the business profits of non-residents. It is common for the tax law definition of PE to be broader than the treaty definition. The scope of the definition of PE is usually an important part of treaty negotiation with the negotiated treaty definition usually narrower than the TC definition.

142. A broader tax law definition of PE will be particularly relevant where no treaty applies. While Ukraine has 73 tax treaties, there are still many countries with which Ukraine does not have a treaty. Even where there is a treaty, the broader tax law definition can apply in situations not covered by the treaty. For example, the registration requirement applicable to PEs in Article 64.5 of the TC can be based on the TC definition of PE, although a treaty may exclude taxing rights. The key point is that there is no obligation to exactly align the tax law definition of PE with the treaty definition.

143. Some examples of a broader tax law definition of PE are:

  • (1) The tax law definition may not have a preparatory and auxiliary activities exception or, the exception may be limited to a representative office (i.e., an office that has representation as its sole activity).

  • (2) The tax law definition may include shorter temporal limits for the construction and services PE rules.

  • (3) The tax law definition may not include an independent agent exception to the agency PE rule.

144. While less common, it is also possible that, under a particular tax treaty, the definition of PE may be broader than the definition in the tax law. The impact of this will depend on the legal relationship between tax treaties and the tax law. The general position is that a tax treaty does not impose tax, rather the tax is imposed under the tax law based on the residence of the taxpayer and the geographic source of the income. Instead, a tax treaty sets out the permitted scope of source country taxation of income derived by a resident of the other Contracting State. As a tax treaty is permissive only, taxation permitted under a tax treaty must be provided for in the tax law to be effective.

145. It is important, therefore, that the jurisdiction to tax asserted under the TC is at least aligned with the taxing rights permitted under Ukraine’s tax treaties (although, as noted above, it may be broader than the treaty definition). For PE, for example, the definition in Article 14.1.193 should be based on the broadest definition in Ukraine’s tax treaties to ensure that Ukraine can fully tax business profits as permitted under each of its tax treaties. This issue is not limited to the definition of PE and can arise in other contexts under tax treaties49 and, therefore, a simple way to align the taxing rights asserted under the TC with the taxing rights permitted under a tax treaty is to provide in Article 14.1.54 that any income that Ukraine has the right to tax under a tax treaty is treated as Ukraine-source income.

Technical Issues with the Revised PE Definition

146. The revised definition of PE in Article 14.1.193 largely follows the amended definition in the OECD and UN Models. A few technical points are raised with the definition:

  • (1) The new definition of PE operates one way in that it only defines a PE of a non-resident in Ukraine. Hence, it has no application in determining whether a resident has a foreign PE. This likely reflects the current role of the PE concept as largely an administrative concept used for registration of non-residents. If, as suggested above, the concept of PE has a greater jurisdictional role to play, then the definition should be drafted in generic terms without reference to the residence of the person conducting the relevant activity. This is the way that PE is defined under Article 5 of the OECD and UN Models.

  • (2) The reference to “server” at the end of the first paragraph of the definition should be clarified. A server (being a physical asset) located in Ukraine that is owned or leased by a non-resident can constitute a PE provided the location of the server is fixed. However, a non-resident that has a website “hosted” on a server owned by a third party Internet Service Provider (ISP) is not usually regarded as having a PE as the non-resident does not have any rights over the server.50 The ISP is not regarded as a dependent agent as it does not have any authority to contract on behalf of the non-resident nor deliver goods. Any attempt to treat the ISP as an agent for the non-resident is problematic as the ISP does not have any knowledge of the transactions conducted through the non-resident’s website on its server nor does it handle any payments on behalf of the non-resident.

  • (3) It is suggested that the references to “six months” in paragraphs (a) and (b) of the definition are changed to “183 days” to avoid the need for the complicated rules dealing with part months.

  • (4) It appears that paragraph (c) of the definition does not fully capture the amendments to Article 5(5) of the OECD Model and it does not appear to capture commissionaire arrangements (although this may be a translation issue).

  • (5) As indicated above, the preparatory and auxiliary activities exception could be deleted, or the exception could, at least, be limited to a representative office. In this regard, it was mentioned several times in the discussions that a representative office of a non-resident is registered as a PE. It is noted, though, that a genuine representative office usually only conducts market research, advertising, and other non-transactional operations and, therefore, the activities of such a representative office usually come within the preparatory and auxiliary activities exception and, therefore, do not constitute a PE.

  • (6) If the preparatory and auxiliary activities exception is retained, then, as the PE definition has the delivery agency inclusion, the reference to “delivery” in paragraph (a) needs to be deleted. Further, it needs to be made clear that paragraph (e) applies only where the activity is of a preparatory or auxiliary nature.

Attribution of Profits to a PE

147. Article 141.4.7 provides that the taxable profit of a non-resident conducting business in Ukraine through a PE is calculated on the basis that the PE is a separate entity with internal dealings valued by reference to the arm’s length principle. This implies that the “functionally separate entity” approach applies to the attribution of profits to a Ukraine PE of a non-resident and this was confirmed in the discussions. Under the functionally separate entity approach, the “remuneration” charged for internal dealings can include a mark-up when supported by functions, assets, and risk as determined by analogy to Article 9 of the OECD Model and the OECD Transfer Pricing Guidelines.

148. This can be illustrated by the example of an internal division within the head office of a company that is set up to perform a treasury function for the rest of the company. The Treasury division borrows externally and then on-lends to other parts of the company, including a PE in Ukraine, at a notional interest rate that includes a mark-up on the external interest cost Under the functionally separate entity approach, the Ukraine PE can deduct the notional interest paid to the head office. The mark-up has the effect of allocating part of the company’s profit to the treasury function performed by the head office. To fully apply the functionally separate entity approach, withholding tax should be charged on the deemed interest paid to the head office. The deemed interest is also subject to the EBITDA limitation on interest deducibility. If the internal interest charge is not supported by functions, assets and risks, then the Ukraine PE can deduct only the portion of the actual external interest cost properly allocated to the PE.

149. While the functionally separate entity approach is the authorized OECD approach as articulated now in Article 7(2) of the OECD Model, not all countries use this approach. Some countries attribute profits to a PE by reference to the “relevant business activity” approach. Under this approach, internal transfers are not recognized as separate transactions based on the legal reality that a PE is not a separate legal person. Instead, the overall profit of the relevant business activity conducted by the non-resident is determined and a part of the profit is then allocated to the PE based on sourcing rules.

150. If Ukraine applies the functionally separate entity approach, but the other country applies the relevant business activity approach, there is risk of either double taxation or double non-taxation. While the functionally separate entity approach is the OECD standard, it is noted that Article 7(2) and (3) of the UN Model continues to support both methods. It would be prudent, therefore, to retain some flexibility in the method used so as to ensure consistency in profit allocation between a PE and head office.

Recommendations

  • Amend Article 14.1.54 to provide that the business profits of a non-resident are Ukraine-source income if attributable to a PE in Ukraine.

  • Amend Article 14.1.54(a) to provide that interest and royalties are Ukraine-source income when paid by: (1) a resident other than as an expenditure of a foreign PE of the resident; or (2) a Ukraine PE of a non-resident.

  • Revise the definition of PE so that it applies to both residents and non-residents.

  • Clarify the scope of treatment of a server as a PE.

  • Review the agency PE rule to ensure that it properly covers commissionaire arrangements.

  • Consider deleting or narrowing the preparatory and auxiliary activities PE exception.

  • Apply the functionally separate entity approach to allocating profits to PEs but retain flexibility to modify the outcome when the other country uses a different method of profit allocation so as to limit the possibility of double taxation or double non-taxation.

  • Where the functionally separate entity approach applies, apply withholding tax to internal dealings recognized as interest or royalties.

F. Transfer Pricing (Actions 8 – 10)

151. TP regulation in Ukraine has been contentious and difficult to implement. Since 2013 Ukraine has attempted to have a functional TP regime, paying special attention to commodities, a large proportion of the country’s total exports. Authorities believe that aggressive tax planning through TP is still a large problem and wish to adopt new measures, in line with OECD/G20 BEPS recommendations, to strengthen the control of tax avoidance.

152. Previous FAD missions51 have discussed extensively Ukraine’s TP regime as well as BEPS report on Actions 8–10.52 This report therefore will omit a general background discussion and will assess directly the new proposal for amending the TP regime.53 The proposal would introduce changes in five major areas, some but not all inspired by the BEPS TP action report: (1) definition of related parties and controlled transactions, (2) anti-abuse measures, (3) low value-added services, (4) TP for commodity transactions, and (5) new guidance on risk. This section of the report discusses these issues in that order.

Definition of Related Parties and Controlled Transactions

Related Parties

153. A new definition of related parties is to be introduced (Article 14.1.39). Two entities are associated, or be part of a group, if they have consolidated accounting obligations under International Financial Reporting Standards (IFRS). The same would apply if the obligation arose because an entity in the group traded in a stock exchange. Although it was explained to the mission that the intention of the draft law is to introduce this definition exclusively for CbC reporting (BEPS Action 13 – see section G of this chapter), the text of the proposal makes it applicable for Article 39 of the TC, which encompasses all transfer pricing. The differentiation should be clarified.

154. The proposal would also change the current definition of related parties resulting from participation in equity, making the rule more lenient. Specifically, it would increase the participation threshold from 20 to 25 percent. Also, it would create a higher threshold for international financial organizations which have ‘privileges and immunity’ according to international treaties of Ukraine; associated enterprises in this case would require (directly or indirectly) at least 75 percent participation. The first measure would gain in independent parties that could be used as comparable observations and could be justified, but it is unclear the need for the second rule. It must be noted that the tax exemption which may be granted to an international financial organization does not include the right to erode the tax base of a related party in the financial private sector that is subject to the CPT in Ukraine. This special rule is highly unusual and unclear as to the problem that may be addressing. The mission did not find this last change justified and recommends keeping a uniform 25 participation threshold.

155. The definition of control determining when two parties are related would also be amended, introducing specific thresholds of commercial integration. Association among entities is generally defined based on participation in equity (as discussed above) or control. The concept of control of one entity over another is inherently difficult to define objectively and some situations of control will inevitably escape any definition. The TC describes some circumstances of control and the draft law would add others based on commercial integration. Namely, any taxpayer that obtains or spends more than 75 percent of its income or expenditures from a single legal entity is presumed to be related to that entity.

156. This rule can easily become an overreach, however. There might be legitimate or inevitable commercial reasons for concentrating purchases or sales in such percentage, and the rule would impose TP obligations to unrelated parties. Moreover, any entity that has a strong position vis-à-vis a client or supplier will have some market power that may result in a pricing advantage, but that is how the market works, it does not mean that there is a TP manipulation for tax purposes. Also, the rule is somewhat naïve, for the supplying or purchasing company can split in different legal entities to escape this threshold.54

157. Over-indebtedness is another reason proposed to assume that two entities are related. The threshold is a debt 3.5 times equity capital and does not apply to banks. This rule is unusual. Normally such threshold triggers the non-deductibility of interest payments for debt above the ratio. The rule is designed instead to control all transactions among the parties, as if they were related transactions. It would seem odd for an entity to loan three times the equity of another to attempt shifting profits through transactions other than interest payments. The rule is also naïve in that the constraint applies to indebtedness to a single entity, which can be easily avoided by originating loans in different (but related) lenders. The rule is therefore redundant, and the reform proposal would gain by eliminating it.55 Instead the debt to equity ratio should be retained as a risk criterion for audit selection.

158. Finally, partners in a joint venture are also considered related, but this might be misguided. The threshold is that participants have 25 percent or more of a common property. However, partners in a common property or joint business venture are independent shareholders, with separate interests. It does not follow from their partnership that they would be indifferent as to how the profits are divided among them. It is conceivable that the partnership could participate in other entities to which it would be related but the concept should not extend to entities owned independently by each of the partners. It should not be assumed that these are related parties.

Controlled Transactions

159. The current general definition of “controlled transaction” is too wide-ranging.

Strictly, it could include any transaction whatsoever, since it refers to all transactions that “may affect the taxpayer’s taxable income” (Article 39.2.1.1). Instead, a controlled transaction should be that which takes place among related parties only. The obligation to comply with the ALP may vary depending on certain circumstances, for example, whether the controlled transaction is carried out by a small enterprise or with another domestic entity, but the general principle should be kept very simple: controlled or related party transactions are all those between related parties, including those which are presumed to be related by law.56

160. The TC also defines the type of transactions that could be considered for TP purposes (Article 39.2.1.4). It does this by way of listing examples, such as the ‘purchase of final or intermediate goods’, with a number of etcetera in the list, making it a loose listing. This would be unnecessary if ‘controlled transaction’ was defined as ‘any transaction among related parties’.

161. The law assumes that the foreign resident is a related party in some cases, for example, when it resides in a LTJ. This approach is standard international practice, but Ukraine takes it a step further by extending the presumption (in the current law) to cases where the foreign resident is a ‘agent’ (Article 39.2.1.1 – b). This is too broad and can affect transactions which are not controlled in reality. Transactions with legitimate brokers residing in a non-listed jurisdiction should not trigger TP obligations. In fact, by defining such operations as controlled the law reduces the availability of market price observations to apply the Comparable Uncontrolled Price (CUP) method for commodities. So, the rule could be counterproductive.

Recommendations

  • Clarify that the new definition of related parties as a function of IFRS obligations only applies to CbC reporting.

  • Do not introduce the special 75 percent participation threshold defining related parties for international financial organizations.

  • Do not introduce commercial dependency ratios as a deemed control threshold for defining related parties.

  • Eliminate the definition of related parties as a function of debt to equity ratios (make it a risk analysis parameter for selecting audits).

  • Eliminate “classifying partners” in joint venture as related parties.

  • Define “controlled transactions” as all those that take place among related parties.

  • Eliminate the presumption that trading with a non-resident agent (in a non-listed jurisdiction) is a controlled transaction.

Anti-Abuse Measures

Deemed Dividends

162. The proposal would recharacterize as (deemed) dividends profits transferred abroad through TP manipulation, intending to collect withholding taxes that otherwise would go unpaid (Article. 14.1.49). This would clearly be the case when the transfer was done by mispricing traded goods or services. However, overpayment of royalties or interests may also be subject to withholding taxes, and possibly at rates higher than dividends, depending on the tax treaty. If tax treaty rates apply, the measure could yield unintended results. However, consistency in the treatment of dividends (either deemed or actual distributions) would probably imply some loss of revenue as compared to applying domestic law withholding rates when recharacterizing payments into dividends.

Business Restructuring

163. The proposal would subject intra-group business restructurings to TP rules (Article. 39.2.1.4 – e). The intention is correct and complies with OECD TP Guidelines, but the language in the Article to be amended does not provide a good legal framework for it. Rather than describing business restructuring as another business transaction ‘for the purposes of transfer pricing’, the law should clearly state the TP principle applying to business restructurings. The fundamental idea is that the ALP applies whenever something of value is transferred among related parties, whether a function, asset or business risk, even if not recorded as a transaction in the affiliate’s accounting. A second condition is that independent parties would not enter into such arrangement without a compensation.57 The compensation should make the taxpayer indifferent, in net present value terms, before and after the restructuring.

Business Rationale

164. MNEs can engage in intra-group transactions which have no parallel among independents for both commercial or tax-planning reasons. The challenge for the ALP is to design rules that will clearly distinguish one type of transactions from the other, so that only the second (tax planning-led transactions) can be disallowed. However, the ALP is weak in this regard and the attempt by the MoF to legislate on it (Articles 39.2.1.10–11) inevitably reflects this problem.

165. The draft law attempts to distinguish these types of transactions by describing the conditions in which the tax authority could disallow a transaction. However, the principle behind these conditions remains rather vague. The lack of ‘business reasons’ test is not used consistently58 and it suggests that the transaction in question must have a business profit, when in fact legitimate business dealings can result in losses. Unfortunately, the state of the art in this matter does not provide an easy or very convincing solution. The ultimate test to verify that a transaction has not been entered principally for tax purposes is that an independent person would be willing to engage in it, even if no actual comparable market transaction exists. So, the argument becomes very hypothetical.

166. The draft law proposes that the burden of the proof should be on the tax authority (Article 39.2.1.11). This will be very difficult to do without documentation provided by the taxpayer showing how the transaction was planned to benefit both entities engaged in it. So, the drafting of the law could be improved by defining ‘rational business reason’ more precisely, requiring that there be a clear profit motive for the transaction from the perspective of the individual entities.59 Also, the burden of proof should be first on the taxpayer to show that the transaction complies with the principle just described.

TP and the Base of Profit Level Indicators

167. TP is often solved by applying a market profit level indicator (PLI) whose denominator is an accounting variable of the tested party, such as costs and expenses or turnover. One precaution to be taken is that such variable is not in turn affected by TP, otherwise the calculation may not be compliant with the ALP. The draft law would require in such cases that the taxpayer notify the authority about the proportion of the variable affected by related party transactions and to justify why it is used as the denominator of the PLI (Article 39.4.6). This seems to be too lenient. Aside from requiring that information, the taxpayer should first attempt at changing the denominator of the PLI to one which does not have TP issues. If this were not possible, the taxpayer should show that the related transactions in the base of the PLI comply with the ALP. Failing this, the taxpayer should consider changing TP methods.

Recommendations

  • Clarify that recharacterized amounts are deemed dividends and are subject to the rate of withholding as per domestic law.

  • Instead of listing business restructuring as another controlled transaction, define the TP principle applicable to business restructuring.

  • Define rational business reason consistently as having a clear profit motive for the individual entity (where no party a priori would be left worse-off).

  • Place the burden of proof for a rational business reason of a transaction onto the taxpayer.

  • Priority options for a PLI containing TP should be to change either the denominator or the TP method.

Low Value-Added Services

168. The proposal adopts the new OECD (2017) guidance on low value-added services.60 While this is an attempt to simplify the application of TP on non-core, support and relatively minor (no intangible or risk related) intra-group services, the guidance is far from simple. Indeed, segregating low value-added services from all other types of services is not trivial: they must be differentiated from services that are only pass-through costs (Article 39.3.2.11), from high value-added services and—very importantly, from those which are exclusively for the benefit of the entity providing the service (Article 39.3.2.12). Not surprisingly, the provision in the draft law is insufficiently defined by way of examples and thus very hard to apply.61

169. This regulation is fundamentally to the advantage of MNEs which have heavy centralized corporate overhead expenditure which they wish to charge to the global operation. The special regime is a simplified safe harbor of sorts with a cap on the mark-up on costs (no TP method needed, nor market benchmarking) which avoids double taxation across the enterprise.62 Nonetheless, the MNE still must demonstrate that the service has been rendered and that it passes the benefit test in the interest of the recipients. Developing countries often see these charges with skepticism, given the difficulties in segregating them from those which are for the exclusive benefit of headquarters’ shareholders. Moreover, expenditure on behalf of shareholders activities is not well defined in the OECD TP Guidelines,63 which makes the low value-added safe harbor administratively burdensome for tax authorities to verify.

170. This proposal is, at best, premature, as the OECD is in the process of revising this guidance. In May 2018 the OECD requested public inputs for scoping future revision of chapter VII (intragroup services) of the TP Guidelines (2017). Responses indicate that current guidance is not satisfactory.64 Ukraine should wait until final guidance has been developed and assess then the extent to which the guidance is beneficial for service recipients. Given that these types of services may represent a significant reduction to returns in the recipient country and can also be an instrument to shift profits, countries that typically host subsidiaries, like Ukraine, should be cautious with regards to these simplifying TP regimes. An option, while new guidance is considered by the OECD, would be to eliminate all references to low value-added services and keep some general rules for related services that condition their deducibility to the test “about actual rendition of service” and to a “benefit test”. A second-best option would be to put a cap to the deduction of low value-added services (which nonetheless have passed those tests) as a (very low) percentage of EBITDA.

Recommendations

  • Eliminate the regime for low value-added services (second best: place a strict limit to total deduction for this item as a low percentage of EBITDA).

  • Keep rendition and benefits tests for allowing deduction of service payments to related parties.

Commodities

171. The proposal prioritizes the application of a CUP TP method for commodities, which is consistent with prior FAD recommendations.65 This is an appropriate restriction, limiting taxpayers’ ability to use just any TP method when it concerns commodities, for other methods can be more easily manipulated, leading to profit shifting in this key sector of Ukraine’s economy (mainly grain exports). The selection of comparable market commodity transactions is the key to safeguard the tax base and by using quoted prices (from organized commodity markets) is generally most reliable. However, the proposal is somewhat lenient as to the types of market comparable observations that may be used with CUP (Article 39.3.3.4). Aside from quoted prices, it mentions ‘internal comparables’ and ‘prices between unrelated entities’, without further considerations. While conceptually correct, these options have to be qualified.

172. MNEs can generate their own internal comparable transactions by trading occasionally small amounts at discounted prices with unrelated parties. A convenient market benchmark can be created this way that will be difficult for the tax authority to challenge. Regulations should specify that such internal uncontrolled transactions, to be acceptable as a market benchmark, would have to be contracted on a regular basis, with several clients, in volumes similar to those negotiated with related parties. Also, regulations should specify that any other market price references would have to be publicly available and issued by a reputable independent source (e.g., Reuters or Bloomberg).66

173. TP methods allow for adjustments to market observations so that they are more comparable to the controlled transaction being valued. The draft law allows such adjustments in applying CUP to commodities without restrictions (Article 39.3.3.4), leaving too much discretion to the taxpayer (or the tax authority). Given the homogeneity of certain commodities (grains) and the availability of quoted FOB prices for Ukraine exports from different Black Sea ports, the types of adjustments allowed of these prices should be limited only to a few (e.g., quality specifications and freight costs when exports are from a port different than those referenced for quoted prices), and explicitly specified in the law.67

174. In case the taxpayer is unable to apply CUP, it can choose another TP method under certain strict conditions. It must reveal the chain of participants in the transactions beginning from the producer to the last entity which sells the commodity (in whichever stage of transformation) outside the MNE group, i.e., to the first independent entity—Article 39.3.3.8). The TC does not clearly say however that if the profit margins so revealed lead to the conclusion that the margin in Ukraine is not reasonable, the SFS could adjust it, or disregard intermediates and use the final price out of group as a benchmark.

Recommendations

  • Qualify internal comparable observations used in CUP method for commodity exports so that they are regular and significant operations.

  • Qualify external unrelated comparable observations used in CUP method for commodity exports so that they are publicly issued by independent and reputable sources.

  • Limit type of TP adjustments for grain exports to, for example, quality of commodity and freight costs.

  • Clarify that complying with information release when applying a TP method other than CUP on grain exports, tax authority can nevertheless modify profit margin or revert to CUP.

New Guidance on Risk

175. Risks assumed in entering a controlled transaction are a comparability factor to find a market benchmark. This is a complex factor to identify, especially because it partly depends on how contracts are drawn up among related parties to the transaction. The OECD BEPS report on TP (Actions 8–10) attempted to put some limit to the discretion with which MNEs could allocate those risks, based on each affiliate’s capacity to assume the risks and their ability to control them. Control of risk is a vague concept and OECD TP Guidelines remit it to the party in charge of decision-making. But this standard has several weaknesses (e.g., joint decisions, relevant /important decisions, sub-contracting management, simulation) and it is often very difficult apply. The MoF draft proposal updates the legislation with the new OECD standards and inevitably reproduces these weaknesses (Article 39.2.2.5). Although it is hard to repair this problem given that the allocation of risks among affiliates makes little conceptual sense (for the group is indifferent), some safeguard could be considered in stating that the risk profile of an individual affiliate should also comply with the ‘rational business reason’ principle. Also, comparability analysis should focus on business traits and functions which are normally observable in independent firms. Functions such as “implementation of strategic management”, “price policy” or “strategy of production and sales” play into the language of tax planning (in current Article 39.2.2.5).

Recommendations

  • Condition risk profile of an affiliate to ‘rational business reason’.

  • Eliminate from current law references to “implementation of strategic management”, “price policy” or “strategy of production and sale’ as examples of functions that may indicate risk assumption.

G. Country-by-Country Reporting (Action 13)

176. BEPS Action 13 involved the development of new transfer pricing documentation requirements for MNEs. These requirements are referred to as country-by-country (CbC) reporting and are central to the BEPS measures dealing with transfer pricing. CbC reporting under BEPS Action 13 applies only to globally significant MNEs based on an annual group turnover of €750 million.

177. It is intended that the CbC Report is filed only with the jurisdiction where the global parent company is resident with the residence jurisdiction then automatically exchanging the Report with the competent authorities of other jurisdictions where the MNE operates. However, this mechanism may not be effective for Ukraine in all cases. There are several potential limitations on Ukraine’s ability to receive CbC Reports:

  • (1) There may be no obligation under the tax law of the parent company’s country of residence to file a CbC Report.

  • (2) Ukraine may not have an information exchange arrangement with the parent company’s country of residence.

  • (3) Even if there is an information exchange arrangement in place, as a practical matter, there may be delays in the CbC report being exchanged with Ukraine.

178. In the case of (1) above, BEPS Action 13 provides that a MNE may appoint a “surrogate parent entity” to prepare and submit the CbC Report. This is intended to avoid multiple obligations for group companies in different countries to file a CbC Report where the parent company has no obligation to file the Report. The draft transfer pricing legislation obliges a resident taxpayer belonging to an international group with an annual turnover of €750 million or more to file the group’s CbC report with the SFS in the following cases:68

  • (1) The resident taxpayer is the parent company of a globally significant group.

  • (2) The resident taxpayer is the surrogate parent company of a globally significant group.

  • (3) The resident taxpayer is a member of a globally significant group and there is no obligation on the parent company to file a CbC Report to its country of residence and no surrogate parent company has been appointed.

  • (4) The resident taxpayer is a member of a globally significant group and there is no information exchange between parent company’s country of residence and Ukraine or, if there is, parent company’s country of residence has repeatedly failed to comply with information exchange requirements with Ukraine.

179. The obligation to file a CbC Report to the SFS in the circumstances specified above is broadly consistent with BEPS Action 13. There is a concern, though, that (3) may be overly burdensome as it is doubtful whether a Ukraine resident subsidiary would be able to compel the global parent company to provide it with the necessary information to prepare the report. The global parent company controls the subsidiary and not the other way around. The circumstances in (4) differ from (3) in that, under (4), the report has been prepared and the issue is the mechanism by which the SFS accesses the report. Further, the circumstances specified in (4) are not likely to be common as Ukraine has entered into 73 tax treaties and is a signatory to the Convention for Mutual Administrative Assistance in Tax Matters.

180. The transfer pricing rules specify an additional CbC reporting obligation on Ukraine MNEs with an annual global turnover of USD50 million or more. As this obligation is imposed solely only on Ukraine MNEs, this is a purely domestic matter for Ukraine. It is for Ukraine to determine what information it requires from its own MNEs. Importantly, though, there must be a clear reason for requiring the relevant documentation and it should be not be an overly burdensome obligation.

181. It has been suggested that this obligation should be separated from the CbC reporting obligation. However, the mission was advised that this may reduce the chances of it being enacted. Ultimately, this is an internal issue for Ukraine. If it is determined that this additional reporting obligation is reasonably required and is not overly burdensome, then the legislative mechanism for requiring filing of the report is a matter for Ukraine.

Recommendations

  • Reconsider the feasibility of asking a Ukraine subsidiary of a globally significant MNE to file a CbC Report when no report is required to be prepared by the parent company and no surrogate parent company is appointed.

  • Ensure that the obligation on Ukraine MNEs with turnover between USD50 million is not an overly burdensome obligation.

H. Mutual Agreement Procedures (Action 14)

182. Ukraine is a member of the Inclusive Framework (IF), sponsored by the OECD with member countries having the obligation to adopt BEPS minimum standards. BEPS Action 14 on Mutual Agreement Procedures (MAPs)69 is a minimum standard, which requires countries to include Aticle 25(1)-(3) of the OECD Model Tax Convention70 in their double taxation treaties (DTAs). This Article provides a mechanism for competent authorities to resolve differences in the application of DTAs. Additionally, domestic law must contain provisions that make these treaties operational. This minimum standard will be monitored by peer reviews in the IF.

183. The MoF proposed a draft text to be included in the TC to regulate MAP under DTAs. This is a positive initiative that will enable Ukraine to comply with BEPS Action 14 minimum standard. The motivation for the Action 14 package is to ensure that:

  • Treaty obligations with regards to MAP are fully implemented;

  • Administrative processes to do so are in place; and

  • Eligible taxpayers have access to MAP.

184. The BEPS Action 14 Report also proposes compulsory arbitration as an option to resolve MAP disputes in a timely manner. Ukraine has incorporated the arbitration clause in several of its treaties but did not opt for the arbitration clause in the MLI.71 This is probably prudent, since there is no consensus on the usage of arbitration to ensure the resolution of MAP disputes.72 While arbitration could be most beneficial in countries where MAP cases have been unresolved for a long time, this is not the case with Ukraine. Some commentators have expressed reservations about using compulsory arbitration given national sovereignty issues.73 Arbitration decisions are confidential and do not set precedents. So, they do not necessarily add to systemic certainty. As Ukraine will be building its MAP experience in the future, training its staff as they negotiate settlements with their treaty counterparts, it may be useful delaying the consideration of introducing under Action 14 compulsory arbitration in all its treaties until a later date.

185. Ukraine has had little experience with MAPs. Only one MAP has been concluded (2018) and there are no additional requests in the pipeline. As the SFS develops its international taxation auditing capacity, the number of cases will increase and having a clear set of rules on how to proceed with MAPs will also help in attracting more cases, since it will reduce the friction in dispute resolutions with foreign investors.

The Powers of the Competent Authority

186. The reform proposal would benefit from greater precision in certain aspects. Particularly important is to have a well-defined and carefully balanced relationship between the competent authority (the MoF) and those of the SFS. While facilitating taxpayers’ access to MAPs procedure is important, this should be done without undermining the enforcement capabilities of SFS. The MoF does not have the technical capabilities to assess many of the issues that may be subject to MAP cases, for example, bilateral advance pricing agreements (APAs). In such cases, the MoF will have to delegate the function, but there are no rules on how this should be done. Administrative regulations must be introduced so that this does not become a random or an arbitrary process that skips the expertise of the SFS.74

187. The main concern is that the competent authority should also be under some kind institutional scrutiny. The mandate of the competent authority is to solve MAPs, to reach an agreement with the foreign authority, which is also acceptable to the taxpayer. It is expected that negotiations by the competent authority will sometimes sacrifice national tax revenue, because it is not its role to raise revenue but to solve disputes. In this role it can overrule SFS audits. These are very substantial powers which should not operate outside institutional mechanisms of checks and balances. Although this is strictly a tax administration issue and thus beyond the scope of this mission, generally, the SFS needs to have a voice in the MAP process and the resolutions of the competent authority must clearly justify and record when they deviate from the SFS position.

Corresponding Adjustment

188. According to the draft law, the taxpayer can request the “controlling authority” (presumably the SFS) to authorize a ‘corresponding adjustment’. This refers to an adjustment to its tax return mirroring the change in TP done by a foreign related party to a controlled transaction. This is the right approach, for the adjustment should in every case be authorized by the SFS rather than being done spontaneously by the taxpayer and subject to a probability of scrutiny after the fact.

189. The language in the draft law should more consistently articulate this proposal. For example, Articles 39.5.5.1 – 3 should—

  • Do not define the corresponding adjustment as a “right”, rather refer to the taxpayer’s right to ‘request’ a corresponding adjustment.

  • Grant the right to request the corresponding adjustment to the ‘taxpayer’ and not, as the proposal reads presently, to a “party of the controlled transaction”. The TC in this regard does not apply to the other (foreign) party in the controlled transaction.

  • Define who is the ‘controlling authority’, and this should be the SFS.

  • Base the right to request a corresponding adjustment on domestic law.

  • Establish that the request for a corresponding adjustment can be made to the SFS when the foreign resident entity has made its own adjustment, 75 not when the foreign resident notifies the ‘intention’ to make the adjustment, as the proposal currently reads.

MAP and Other Remedies

190. If a corresponding adjustment is rejected by the controlling authority, the taxpayer can request a MAP to the competent authority, “irrespective of remedies provided by the Code” (Article. 108.1.1). While the general procedure described is correct, it is not clear if the taxpayer can proceed simultaneously with other remedies and MAP, including litigation. While taxpayers could be granted the right to request MAP without having exhausted other administrative dispute resolution mechanisms available to them, they should not be allowed to do it simultaneously, including litigation. Also, if a MAP concludes with no agreement, the taxpayer should not be able to resort to other administrative remedies, except for litigation. This should be made clear in the TC.

Who Can Request MAP

191. The proposal allows both resident and nonresident parties to a controlled transaction to request a MAP (Article 108.1.2). While countries can opt for this approach according to the BEPS Action 14 report, it may lead to more MAP cases than necessary. Another approach, also permissible under BEPS minimum standard, is that parties must go through their competent authority in order to request a MAP. Under this approach, a non-resident entity must present its case of alleged double taxation to its own competent authority, which, after examining the case, would contact Ukraine’s competent authority to initiate a MAP. Similarly, Ukrainian taxpayers can request a MAP to the MoF when they seek protection under a treaty for their operations abroad. This procedure has the advantage that cases which the foreign competent authority does not think have the merits or warrant the administrative costs of negotiating, will not engross the MAP inventory.76 The MIL allows countries to choose this second option, 77 which should be preferred in Ukraine given resource constraints (but the MoF has already ratified MLI having chosen the other option).

When to Request a MAP

192. The draft law allows for MAP requests to be made either “before or after a tax audit”, which would disallow them at any stage during an audit (Article 108.2.1). However, it can be inferred from the text of Article 108.2.2 that this is a possibility as well; the provision reads as follows: “… the decision of tax assessment … shall not be approved until completion of MAP ...”. This could be interpreted as saying that a MAP could interrupt an audit in progress, stopping its final tax assessment from being issued. It should be clarified if this is indeed intended. If so, it should be clear as well that MAP should not suspend an audit before the summary of findings is issued to the taxpayer, so that the competent authority has at its disposal SFS’ formal position about the case in question. MAPs should not become an instrument simply to stop audits in their tracks before the SFS can assess the case and provide its findings to the competent authority why the position proposed by the taxpayer should or should not proceed. MAP should be an instrument for resolving disputes on double taxation rather than replacing the audit process.

Statute of Limitations

193. Another cause for concern is how MAPs affect (or not) the statute of limitations. While it is established in the draft law that the tax audit would be suspended while the case is taken up for MAP (to be renewed if negotiations fail), the version reviewed by the mission says nothing explicit about suspending the statute of limitations. Without a rule that suspends all statute of limitations during MAPs, there would be no guarantee that the audit could in fact be finalized if a MAP lasts long enough and does not find a solution to the dispute. This could create an incentive for taxpayers requesting MAPs for the sole objective of playing the clock against the tax auditors.78

194. The draft law explicitly allows taxpayers to reject in the end any agreement reached by the tax authorities under MAP. Although it is highly unusual for tax authorities to reach an agreement on an adjustment which is not satisfactory to the taxpayer, such situation is conceivable. If this were the case, the agreement per se, arguably, should be left to stand and the taxpayer should be notified accordingly. The affected entity could still choose to litigate the MAP outcome but would have to do so against a position that has officially resolved double taxation. It is recommended therefore to eliminate the text in Articles 108.5.4 -5 that allow the taxpayer to terminate the MAP because it disagrees with the outcome. This would not limit the possibility of a competent authority (in agreement with its taxpayer) to terminate a MAP at any time.

Recommendations

  • Clarify that MAP cannot stop an audit before the SFS has issued summary of findings.

  • Develop administrative regulations on how the MoF will delegate the competent authority for resolving MAPs.

  • Clarify the language in the proposed text so that taxpayers have the right to request a corresponding adjustment, not the right to the adjustment.

  • Trigger corresponding adjustment requests with an actual adjustment made by a nonresident, not just with notifying the intention of doing it.

  • Allow the taxpayer to access MAP without having exhausted other means of dispute resolution but he/she should not be allowed to pursue them simultaneously.

  • Opt whenever possible for MAPs to be initiated only through bilateral competent authority negotiations.

  • MAP should not suspend an audit before the SFS has issued its summary of findings.

  • Make sure that the statute of limitations is suspended for audits while MAP is being negotiated.

  • Do not expand in further agreements the acceptance of the arbitration option.

Appendix I. Tax Treaties and the Multilateral Instrument (Action 15)

195. A tax treaty is an international agreement between two or more countries (referred to as “Contracting States”). In broad terms, the effect of a tax treaty is to allocate taxing rights (and, therefore, tax revenues) in respect of income or gains arising from economic activity occurring between the Contracting States. Essentially, this is done by a tax treaty limiting the taxing rights of a Contracting State as a source country. Ukraine is a party to 73 tax treaties currently in force.1 This is a significant number of tax treaties and means that, in many cases, the application of Ukraine’s international tax rules is impacted by tax treaties.

196. The implementation of some BEPS recommendations, particularly in relation to BEPS Action 6 (abuse of treaty), Action 7 (avoidance of PE), and Action 14 (MAP), will require amendments to existing tax treaties. The Actions 6 and 7 amendments are designed to protect against abuse of treaty and strengthen source country taxing rights. A tax treaty is normally amended by the two Contracting States negotiating a protocol to the treaty. However, this would be a resource-intensive and time-consuming process for implementing BEPS changes, particularly for a country like Ukraine that has a significant number of existing tax treaties. To avoid the need to negotiate protocols to existing tax treaties, BEPS Action 15 involved the development of the MLI as the mechanism for the timely implementation of the BEPS amendments to existing tax treaties.

197. Ukraine signed the MLI on July 23, 2018 and has recently lodged the instrument of ratification with the OECD. While signing the MLI will allow Ukraine to implement the BEPS treaty amendments in relation to existing treaties in a timely manner, the MLI is a complex document that requires careful consideration of its implications for existing tax treaties and the negotiation of future treaties. It is important, therefore, that the implementation of the MLI is done as part of a broader strategy of developing a tax treaty policy and updating the Ukraine Model Treaty. A tax treaty policy will guide decision making concerning the scope of application of the MLI to Ukraine’s existing tax treaties, and also guide decisions concerning future treaty partners and the content of new treaties. The updating of the Ukraine Model will be an important component of the tax treaty policy and will strengthen Ukraine’s position in the negotiation of future tax treaties.

198. It is desirable that there is a moratorium on the negotiation and signing of new tax treaties. This will allow the MoF to focus on developing a tax treaty policy, updating the Ukraine Model Treaty, and applying the MLI to existing tax treaties. It will also ensure that, as far as possible, there is consistency between existing and future tax treaties. The discussion below considers first the development of a treaty policy and updating the Ukraine Model, and then the application of the MLI to existing treaties.

Tax Treaty Policy

199. Because of the potential negative impact that tax treaties can have on tax revenues in the Ukraine, it is important that the MoF develops a tax treaty policy to provide a framework for MLI implementation and future treaty negotiations. The tax treaty policy should include the following components: (1) clear guidelines for choosing treaty partners; (2) a baseline for negotiations; (3) the use of the Ukraine Model Tax Treaty in negotiations; and (4) the preparation of a Tax Treaty Impact Statement once a treaty has been negotiated but before it is signed.

Guidelines for Choosing Tax Treaty Partners

200. It is essential that clear guidelines are developed for choosing treaty partners. Importantly, is not incumbent on the Ministry to negotiate a tax treaty with every country that seeks a treaty with Ukraine nor would this be in the country’s best interests.

201. The main argument for capital importing countries, like Ukraine, to enter into a tax treaty is that it will attract trade and investment into the country. Consequently, in choosing a treaty partner, the starting point should be to consider the current level of trade and investment into Ukraine from the potential treaty partner. If that level of trade and investment is significant, there is likely to be a greater justification for negotiating a tax treaty with that country. Where there is little or no current trade and investment with a potential treaty partner, then greater care must be taken in making the decision to negotiate with that country. In particular, an investigation should be undertaken to determine whether there are impediments to such trade and investment, and a realistic assessment made as to whether a tax treaty will help remove or overcome those impediments. If it is unlikely that a tax treaty will have any significant impact on the level of trade or investment coming into Ukraine, then there is likely to be little justification in negotiating a treaty with that country.

202. The guidelines should prohibit negotiations with any country whose tax rules or practices pose a revenue risk to Ukraine. Countries that either have no income tax, a preferential regime for foreign income, or that tax only on a territorial basis, pose a potential revenue risk for Ukraine. A treaty negotiated with a such a country may result in double non-taxation of any income that Ukraine is required to exempt from tax under the treaty.

203. Further, in negotiating a tax treaty, the clear intention of Ukraine must be that treaty benefits are limited to genuine residents of the other Contracting State. Consequently, in choosing a treaty partner, it should be clear that the potential trade and investment coming to Ukraine from that country is genuinely sourced in that country. Some countries negotiate a broad treaty network to facilitate treaty shopping (see Chapter III D above). In this case, the aim is to attract residents from outside the country to establish a base company in the country to take advantage of the country’s tax treaty network. The base company may be established under a preferential tax regime in the country or the country may be a territorial tax country. If Ukraine negotiates a tax treaty with such a country, the revenue loss from the treaty can be expected to be much greater than would be the case if the application of the treaty were confined to genuine residents of the other country.

204. It is acknowledged that the inclusion of an anti-treaty shopping rule (such as a “LOB” Article) in a tax treaty will limit the impact of treaty shopping. However, such a rule will not necessarily counter all possible treaty shopping abuses, particularly in relation to services. Consequently, it would not be sensible to enter into a tax treaty with a high-risk country with the expectation that the inclusion of a LOB Article will completely protect Ukraine from treaty shopping under that treaty.

Baseline for Negotiations

205. The tax treaty policy should set out the baseline for negotiation of a tax treaty. This will identify those provisions of a tax treaty that Ukraine is prepared to negotiate on and those that are non-negotiable. At a minimum a treaty should include the following:

  • A positive rate of tax on dividends, interest, royalties, and management and technical fees.2

  • The taxation of indirect transfers of immovable property in Ukraine as provided for in Article 13(4) of the Model Treaties.

  • An anti-abuse rule.

206. Where the tax treaty policy provides that there is to be no negotiation on a particular subject matter (such as zero rates), it is essential that the negotiation all tax treaties abides by this baseline. As soon as one treaty is negotiated that departs from the agreed baseline, Ukraine will be under pressure to give the same benefit in all future treaty negotiations, which may be difficult to refuse. In other words, such a treaty will, in effect, set a new baseline for future negotiations.

Model Tax Treaty

207. Ukraine has a model tax treaty that it uses in treaty negotiations. The Ukraine Model was prepared in 2005 and largely follows the UN Model. Having a Ukraine Model is consistent with best practice and allows treaty negotiations to focus on areas of disagreement as between the two countries. However, the Ukraine Model needs to be updated for BEPS changes to tax treaties. Further, the Ukraine Model should be fully integrated with the development of the tax treaty policy so that it sets out the baseline for negotiations. There is further discussion below on the Ukraine Model Treaty.

Tax Treaty Impact Statement

208. As a tax treaty will involve a loss of revenue through the reduction in Ukraine’s taxing rights as a source country under the treaty, a Tax Treaty Impact Statement should be prepared before a treaty is signed. This will facilitate the undertaking of a cost/benefit analysis in relation to potential future tax treaties. A Tax Treaty Impact Statement should set out the following:

  • (1) The reasons for choosing the other country as a treaty partner, including a statement of the current volume of trade and investment into Ukraine from the country.

  • (2) An assessment of how the tax treaty will increase the level of trade and investment into Ukraine, or other otherwise lead to closer economic relations.

  • (3) A statement of any other benefits to Ukraine that may be obtained under the treaty.

  • (4) A quantification of the potential revenue loss for Ukraine under the treaty.

Updating the Ukraine Model Treaty

209. There are two main model tax treaties – the OECD Model3 and UN Model4. While described as a separate model treaty, the UN Model is based on the OECD Model with some important modifications that provide for greater source country taxing rights over some classes of income. While tax treaties are largely based on either the OECD or UN Model (or a mix of the Models), there may be some departures from the Models in particular treaties reflecting the fact that a tax treaty is the outcome of negotiations between the two Contracting States. The latest version of each Model was published in 2017 and includes amendments to tax treaties recommended in the Reports for BEPS Actions 6, 7, and 14.

210. The UN Model was specifically developed for use in negotiations between developed and developing countries. The main differences in the UN Model are:

  • (1) A broader definition of PE in Article 5 that allows for greater source taxation of business profits under Article 7. The key differences in the UN Model are: (1) a services PE inclusion; (2) delivery agent inclusion; (3) an insurance inclusion: (4) a shorter time period (183 days) for the establishment of a construction PE; and (5) the inclusion of supervisory services as a construction PE.

  • (2) The application of a limited force of attraction principle in determining the taxable profits of a PE (Article 7(1)). In addition to the business profits attributable to the activities of a PE, there is also source taxation of income from transactions in the source country (such as sales or services) of the same or similar kind as those conducted through the PE.

  • (3) The PE profit attribution rules are based on the pre-2010 OECD Model rules and, therefore, retain flexibility for the Contracting States to use either the functionally separate entity approach or relevant business activity approach for the calculation of PE profits (Article 7(2) and (3)).

  • (4) Source taxation of royalties is permitted subject to a rate limit (Article 12).

  • (5) The definition of “royalties” includes equipment lease payments (Article 12(3)). Under the OECD Model, these payments are business profits and, therefore, the source country can tax only if the lease payments are attributable to a PE.

  • (6) Source taxation of fees for technical, management, or consultancy services is permitted subject to a rate limit (Article 12A). This Article was included in the 2017 UN Model and is an important limitation on base erosion given the high level of trade in services today.

  • (7) The UN Model has retained a separate Article for the taxation of payments for independent personal services (Article 14). The OECD Model treats such payments as business profits covered by Article 7.

  • (8) The UN Model retains the right of the source country to tax “other income” that is sourced in the country under domestic law.

211. As stated above, the Ukraine Model was prepared in 2005 and is based largely on the UN Model. It is accepted that the Ukraine Model needs to be updated, but an issue has arisen as to whether the updated Ukraine Model should continue to be based on the UN Model or be more closely aligned with the OECD Model. This is an important issue for an emerging economy like Ukraine and concerns the balance between residence and source taxation of income arising from cross-border transactions between the two Contracting States.

212. Broadly, the effect of a tax treaty for a Contracting State is to “trade off” reduced taxing rights as a source country in return for increased taxing rights as a residence country. If the trade and investment flows between the two Contracting States is roughly equal (say between two OECD countries), a tax treaty is likely to have limited impact on the overall revenues of each Contracting State. In this case, the effect of a tax treaty is to change the composition of revenues collected from cross-border transactions with reduced tax revenue collected from non-residents offset by increased tax revenue collected from residents.

213. For a capital importing country like Ukraine, however, tax treaties result in an overall reduction in tax revenue. The loss in tax revenue collected from non-residents is not offset by any increase in tax revenue collected from residents. Where outbound investment is relatively low, tax treaties can result in a significant revenue loss for the government. Thus, the choice between the OECD and UN Models essentially involves a decision as to the extent of the revenue loss from tax treaties. The revenue loss will be greater under the OECD Model and the issue then is whether that is offset by non-tax benefits, particularly an increase in trade and investment into Ukraine.

214. Given Ukraine’s status as an emerging economy, it is preferable that the Ukraine Model continues to be based largely on the UN Model. However, that is not to say that the Ukraine Model must exactly replicate the UN Model. For example, the UN Model continues to tax income from independent personal services separately under Article 14 by reference to the concept of “fixed base”. Article 14 was deleted from the OECD Model some years ago with the income from independent personal services taxed under Article 7 as business profits. The OECD approach is considered the better approach because of the clarity around the definition of PE as opposed to the undefined term “fixed base” used Article 14. Ultimately, the Ukraine Model may be a mix of the OECD and UN Models, but with greater reliance on the UN Model to provide for greater protection source taxation.

Implementing the Multilateral Instrument

215. A country that signs the MLI is referred to as a “contracting jurisdiction” and an existing tax treaty to which the MLI applies is referred to as “covered tax agreement” (CTA). A contracting jurisdiction is required to implement the BEPS minimum standards for tax treaties, namely limitation of benefits and improving dispute resolution processes. A contracting jurisdiction can opt out of the other provisions in the MLI generally or for particular CTAs. Consequently, the MLI applies to an existing treaty only if both contracting jurisdictions have signed the MLI and agree on the scope of its application to the existing treaty. The scope of application of the MLI may be limited to the minimum standards or extended to include other provisions in the MLI. The instrument of ratification lodged by a signatory country sets out the intended scope of application of the MLI to the country’s treaties.

216. The MLI does not amend a CTA in the same way as a protocol to a tax treaty amends the treaty. Instead, a CTA and the MLI “coexist” with the MLI modifying the application of a limited number of Articles in a CTA (such as the Article 5 definition of PE) and also adding some new provisions into a CTA (such as the Entitlement to Benefits Article). Both the MLI and a CTA are international treaties and, therefore, under international law, in the case of conflict, the later in time prevails. Consequently, the MLI being later in time has priority over the CTA in the case of conflict. However, the MLI does provide some compatibility clauses for greater clarity.

217. The mission was advised that Ukraine intends to implement only the minimum standards, and Article 9 of the MLI dealing with taxation of indirect transfers of immovable property. In relation to entitlement to benefits, it is intended to adopt only the principal purposes test. It is recommended above that the MoF develops a tax treaty policy and updates the Ukraine Model before agreeing on the application of MLI to existing treaties. This will ensure that the implementation of the MLI is consistent with the tax treaty policy and updated Ukraine Model, and that there is a degree of consistency as between Ukraine’s existing tax treaties and its future treaties.

218. It is recommended that Ukraine adopts two other optional provisions in the MLI that will protect against the avoidance of Ukraine tax.5 These are:

  • (1) The application of a 365-day holding period before a non-resident qualifies for the lower 5 percent tax rate on participation dividends under Article 10 of tax treaties (Article 8 of the MLI). The holding period will ensure that the lower treaty rate does not apply to corporate shareholders in non-resident companies that engage in certain short-term share acquisitions (such as under securities lending arrangements).

  • (2) Amending Article 4 of tax treaties to permit the treaty residence of a dual residence company to be determined through mutual agreement based on place of effective management, place of incorporation or formation, or any other relevant factors (Article 4 of the MLI). This is particularly relevant for Ukraine as its uses place of incorporation as the sole test of corporate residence and, therefore, the tiebreaker rule in current Article 4(2) of tax treaties will rarely apply to allocate residence to Ukraine.

219. Importantly, the MLI does not address all weaknesses in Ukraine’s current tax treaties. In particular, a number of current treaties include a zero rate on dividends, interest, or royalties. While the PPT and/or LOB will limit treaty shopping abuses to take advantage of zero rates, zero rates can only be fully addressed through either the renegotiation of the treaty or the negotiation of a protocol to the treaty. It is understood that this is being prioritized by the MoF. The development of a treaty policy and the updating of the Ukraine Model will also help identify non-MLI issues with existing treaties.

Recommendations

  • Develop a Tax Treaty Policy including guidelines on treaty partners, a baseline for negotiations, and the preparation of a Tax Treaty Impact Statement.

  • Update the Ukraine Model Treaty for BEPS changes to tax treaties and align the Ukraine Model Treaty with the UN Model 2017.

  • Reconsider the scope of application of the MLI to existing treaties once the tax treaty policy is developed and the Ukraine Model updated to ensure that, as far as possible, there is consistency between existing treaties and future treaties.

  • Put a moratorium on negotiating new tax treaties until after the tax treaty policy is developed and the Ukraine Model updated.

1

The label ECT can be misconstrued since an exit tax in other countries has a different meaning, related to a capital gains tax when the taxpayer changes residence. Also, the name would seem to imply that the tax applies to all capital withdrawn from the corporation, including the initial capital contributed by shareholders, which should not be subject to a tax when reimbursed to investors.

2

There has been some discussion to adopt the ECT for small taxpayers only. This is not analyzed here since it would involve an assessment of the simplified tax regime, which is beyond the scope of this report.

3

The IMF Capacity Development Mission on Tax Policy, Schatan, R., M. Grote, and M. Kobetsky, 2017. Rethinking Dividend Distribution Tax, Rationalizing Simplified Taxes, and Adopting BEPS Measures (Washington, DC), evaluated and commented extensively on the DPT.

4

Disallowing past losses creates a disadvantage for ‘old’ capital versus new investors, which is manifested as a loss in value of assets that have not fully amortized those losses.

5

Reanimation Package of Reforms Team, 2018. Roadmap of Reforms for 2019–2023, see Chapter on Tax and Budget Reform, and related PowerPoint Presentations, Kiev.

6

The lowest level of corporate tax revenue was in 2001, the first year the DPT was in force, registering less than 0.7 percent of GDP. However, CIT revenue had been dropping fast before 2001, so this low number cannot be entirely imputed to the tax policy change.

7

From 2019 onwards, a lower 14 percent CIT rate applies for companies making regular profit distributions.

8

Figure 4 compares Ukraine’s CPT revenue performance since 2001 – 2018. CPT collections have recovered since 2014 remarkably, being slightly above the average for EU-28, OECD member states, and Balkan states. Hence, the resilience of the Ukrainian CPT is a virtue and it is a key tax instrument needed for maintaining fiscal balances.

9

Moldova is sometimes referred as another example when it instituted a zero rate for CIT in 2008. However, the reform is not equivalent to substituting CIT for a DPT since withholding on dividends was not increased in the expectation to compensate revenue. The zero rate on CIT was ended in 2012.

10

Data provided by the German Advisory Group on all four countries group point in the same direction, a significant drop in tax revenue after adopting the DPT. See D. Saha, et al, Taxation of Distributed Profits: International Experience, Institute for Economic Research and Policy Consulting, May 2017.

11

In Georgia CGFCF dropped the year DPT was adopted, but since only one-year data is available, this is not very meaningful.

12

IMF, Ukraine—Policy Note for the 2019 Article IV consultation and first review under the SBA, May 2019.

13

Schatan et al (2017), p 14.

14

Seminal paper on this topic is A. Harberger (1962), The Incidence of the Corporation Income Tax, Journal of Political Economy, 70(3).

15

See A. Auerbach (2006), Who Bears the Corporate Tax? A Review of What We Know, Tax Policy and Economics; H.A. Vazquez-Ruiz, A New Approach to Estimate the Incidence of the Corporate Income Tax, Dissertation, Georgia State University, 2012; available at: https://scholarworks.gsu.edu/econ_diss/82

16

Jane G. Gravelle, Who pays the corporate tax? Congressional Research Service, September 29, 2017.

17

Privatization proceeds should be utilized to reduce the principal amount of public debt and is not a sustainable revenue raiser given its one-off nature.

18

“Fixed investment” is the purchasing of newly produced fixed capital or the accumulation of physical assets such as machinery, land, buildings, installations, vehicles, or technology.

19

Nov 27, 2018, Corporate Tax Rates Around the World, 2018, Tax Foundation.

20

Zucman, G., 2013. “The Missing Wealth of Nations: Are Europe and the US Net Debtors or Net Creditors?”, The Quarterly Journal of Economics, 128(3): 1321–1364.

21

Bethmann, D. and M. Kvasnicka, 2017. “International Tax Evasion, State Purchases of Confidential Bank Data and Voluntary Disclosures”, Beiträge zur Jahrestagung des Vereins für Socialpolitik 2017: Alternative Geld- und Finanzarchitekturen – Session: Taxation II, No. B19-V3 ZBW – Deutsche Zentralbibliothek für Wirtschafts-wissenschaften, Leibniz-Informationszentrum Wirtschaft, Kiel, Hamburg.

22

Bethman, et al.

23

Martin, L. and J. Oldfield, 2017. “Best Practices in Tax Amnesty and Asset Repatriation Programmes”, Transparency International and the European Commission and Le Borgne, E., 2006. “Economic and political determinants of tax amnesties in the U.S. States”, Proceedings of the Ninety-Eighth Annual Conference on Taxation, National Tax Association: Washington, DC.

24

Transparency International, 2017.

25

Clauses 205, 212, 212, 366 and 367 of the Criminal Code of Ukraine.

26

Article 205 refers to setting up a sham business; Article 212 refers to the willful evasion of taxes, duties (compulsory payments) which are part of the taxation system established by law; Article 212–1 refers to the evasion of payment of premiums on obligatory state pension insurance or social security contributions; Article 366 refers to forgery in office; and Article 367 refers to any neglect of official duty by an official causing significant damage to the legally protected rights and interest of individual citizens or state or public interest.

27

The successful South African foreign asset amnesty (2003)27 imposed a 5 percent levy on the value of disclosed foreign assets that were repatriated and a 10 percent levy on the market value on non-repatriated foreign assets minus any permissible allowed threshold of foreign investments held by a South African resident.

28

This is an exceedingly generous tax incentive that can hardly be justified as it rewards handsomely all previous exchange control and accompanying tax violations. Indeed, it provides an additional benefit to non-compliant taxpayers vis-à-vis compliant taxpayers, compromising severely the horizontal equity principles. The granting of this tax expenditure should be resisted as the overriding priority is to grow the tax base and regularize foreign asset and portfolio holdings.

29

Lub, H., 2019. Ukraine—Report on Balance of Payments Statistics Mission (November 12 – 23, 2018), Statistics Department of the International Monetary Fund (Washington, DC).

30

In terms of VD and asset repatriation amnesties some country experiences suggest a significant growth in tax base: South Africa grew it by EUR 7.8 billion; Argentina by USD 100 billion; Brazil by USD 16 billion; Indonesia by USD 321 billion (see Martin, L. and J. Oldfield, 2017. “Best Practices in Tax Amnesty and Asset Repatriation Programmes”, Transparency International and the European Commission).

31

The South African VD program with related tax amnesty provided for a joint facilitator application with an amnesty applicant. The objective of this provision was to extend amnesty to parties who rendered assistance to applicants in transferring funds abroad from South Africa without the necessary authorization.

32

Some amnesty experiences indicate that initially only a small number of applications is lodged but as prospective applicants became aware that the adjudication unit dealing with application grants amnesty without any negative consequences for applicants who applied shortly after the submission period had opened, interest in applying for amnesty will increase. Experience also suggest that the majority of applicants tend to delay the lodgement of their applications until the very last moment, inter alia, because they expected that the deadline for submission would be postponed.

33

Financial Action Task Force (FATF), 2012. “Best Practices Paper. Managing the anti-money laundering and counter-terrorist financing policy implications of voluntary tax compliance programmes”. http://www.fatf-gafi.org/publications/fatfrecommendations/documents/bestpracticesmanagingtheanti-moneylaunderingandcounter-terrorirstfinancingpolicyimplicationsofvoluntarytaxcomplianceprogrammes.html

34

Not all tax amnesties do. See Said, E.W., 2017. “Tax Policy in Action: 2016 Tax Amnesty Experience of the Republic of Indonesia”, MDPI Laws, School of Law, Politics and Sociology, University of Sussex, Brighton. The mission received assurances from the NBU that if funds are repatriated to Ukraine, they would be transferred from foreign bank accounts. Thus, it is assumed that relevant source jurisdictions would have carried out AML and CFT compliance checks before fund transferals to Ukraine. When no income/assets are repatriated into Ukraine (disclosure of assets not held in bank accounts) separate AML and CFT procedures appear irrelevant.

35

BEPS measures are discussed here as a way to strengthen the CPT; they are not discussed in the context of a DPT.

36

This is a general rule under CFC regimes, although, in the case of Ukraine, there is no unilateral double tax relief.

37

See Ruud de Mooij, et al, Tax policy to strengthen the revenue base, IMF/FAD, April 2013; R. Schatan, et al, Rethinking dividend distribution tax, rationalizing simplified taxes, and adopting BEPS measures, FAD/IMF, May 2017.

38

OECD (2017), Limiting base erosion involving interest deductions and other financial payments, Action 4 – 2016 Update, OECD/G20 BEPS Project, OECD Publishing, Paris.

39

TC, arts 140.1 – 140.3.

40

MoF, Restrictions on interest deductions (internal document, undated).

41

An additional advantage of the EBITDA rule is that it does not breach Article 24 of tax treaties as it applies to both residents and non-residents. Traditional thin cap rules require an arm’s length debt exception to avoid breaching Article 24.

42

The OECD BEPS Action 4 Final Report (2015) advised that the rule based on gross interest expenditure should not apply to the financial sector (par 188). The updated version of this report (OECD (2017)) did not change this conclusion.

43

OECD (2017), par 507. Exempting the financial sector would be reasonable anyway if the financial regulatory framework in Ukraine is sufficiently robust to avoid base erosion through excessive indebtedness. This assessment was not done by the mission.

44

OECD (2017), par 513.

45

OECD (2017), par 506.

46

SeeThe Taxation of Offshore Indirect Transfers – A Toolkit (2017) by the Platform for Collaboration on Tax accessed at http://documents.worldbank.org/curated/en/322921531421551268/a-toolkit-draft-version-2.

47

It is noted that Article 7(1) of the UN Model extends source taxing rights to include also income from transactions (sales or services) of the same or similar nature as the transactions undertaken through a PE. Consequently, if Ukraine has tax treaties that follow Article 7(1) of the UN Model, then this will need to be reflected in the sourcing rule in the TC.

48

Article 14.1.54(i) expressly refers to business activities and independent personal services.

49

Similar issues commonly arise with “royalties” as the treaty definition may be broader than the tax law definition. For example, the treaty definition may follow the definition in Article 12 of UN Model and include equipment lease rentals, but this may not be provided for in the tax law definition.

50

Paragraph 124 of the Commentary to Article 5 of the OECD Model.

51

See Ruud de Mooij, et al, Tax policy to strengthen the revenue base, IMF/FAD, April 2013; Roberto Schatan, et al, Restoring a strategic approach to tax reform, FAD/IMF, March 2015; R. Schatan, et al, Rethinking dividend distribution tax, rationalizing simplified taxes, and adopting BEPS measures, FAD/IMF, May 2017.

52

OECD (2105), Aligning Transfer Pricing Outcomes with Value Creation, Action 8–10 – 2015 Final Report; OECD/G20 BEPS Project, OECD Publishing, Paris.

53

“Summary of proposed amendments to the Tax Code of Ukraine in terms of implementation of BEPS Actions 8–10 and 13”, document provided to the mission by Ukraine’s MoF.

54

The threshold should be used as a risk criterion in risk analysis for selecting tax audits.

55

It was explained to the mission that in one case an entity had not declared to have transactions with related parties but had a debt to equity ratio above the cited threshold and when audited it was found that it had controlled transactions under the general definition of the concept. Thus, the debt to equity criteria served really as an element to select taxpayers for audit.

56

For example, when the foreign counterpart resides in a low tax jurisdiction.

57

OECD TP Guidelines.

58

Article 39.2.1.10 refers to “business reasons” and Article 39.2.1.11 refers to “rational business reason” and the conditions describing each are different.

59

This would exclude any arrangement that a priori would make any of the parties worse-off.

60

OECD (2017), Transfer pricing guidelines for multinational enterprises and tax administrations, OECD Publishing, Paris, chapter VII.

61

Provisions include eight paragraphs listing examples of services that could be low value-added (and not an exhaustive list) and describes ten types of services that may not be recognized as such (Article 14.1.183), plus another 5 paragraphs describing costs associated with the maintenance of the parent company (Article 39.3.2.12).

62

Low-value added services would be considered complaint with the ALP when charged with a mark-up of 5 percent over costs (Article 39.3.2.10).

63

OECD (2017), par 7.9 (2010).

64

OECD, Comments received on the request for input, Scoping of the future revisions of Chapter VII (intra-group services) of the Transfer Pricing Guidelines, 28 June 2018.

65

TP for commodities was amply discussed in the 2017 FAD technical report.

66

While such terms may not be sufficiently defined to settle the issue entirely, the regulation could include specific examples to clarify. R. Schatan et al (2017) contains a more detailed discussion on TP methods and their application to Ukraine’s commodity export.

67

Costs of port logistics should be part of the selected grain export price for TP purposes (i.e., as if paid by the Ukrainian exporter), so the benchmark should be FOB Black Sea port. Each commodity may require its own type of adjustments.

68

Article 39.4.10 of the Tax Code

69

OECD (2015), Making Dispute Resolution Mechanism More Effective, Action 14 – 2015 Final Report, OECD/G20 BEPS Project, OECD Publishing, Paris.

70

OECD (2017), Convention on Income and Capital, OECD Publishing, Paris.

71

See OECD (2015), Developing a Multilateral Instrument to Modify Bilateral Tax Treaties, Action 15 – Final Report, OECD/G20 BEPS Project, OECD Publishing, Paris. This is further discussed the appendix of this chapter.

72

Tax Certainty, IMF/OECD Report for the G20 Finance Ministers, March 2017, p. 57.

73

BEPS Monitoring Group, ‘Explanation and Analysis of the Multilateral Convention to Implement Ta Treaty Related Measures to Prevent Base Erosion and Profit Shifting,’ reproduced by Tax Analysts (2017), p. 5.

74

For example, it would be undesirable to have the SFS ruling one way on unilateral APAs and the same issue solved differently under a Bilateral APA through the competent authority. One way to solve this would be to have the SFS as the competent authority, at least for Bilateral APAs.

75

The adjustment can be documented with a filed modified tax return by the foreign resident.

76

Experience in other countries indicate that MAP inventory cases often reach problematic levels, which is one of the main reasons motivating BEPS Action 14. It is important therefore to cut down on unsubstantiated MAP applications.

77

MIL, Article 16.

78

The statute of limitations for TP audits in Ukraine is seven years. It could be expected that MAP cases would be resolved before that long (targeted average period to solve a MAP by Action 14 is two years). However, not all cases taken up by MAP are about TP and all other cases have a statute of limitations of three years.

1

70 tax treaties were negotiated by Ukraine as a sovereign nation and 3 treaties of the former USSR continue to apply to Ukraine as the successor to the USSR.

2

While there is currently only limited taxation under the Tax Code of fees paid to non-residents for services, this may change in the future as many countries now impose withholding tax on management and technical fees paid to non-resident. This taxing right is now preserved under Article 12A of the UN Model. Consequently, Ukraine’s treaties should preserve some level of source country taxing rights over management and technical fees.

3

OECD Model Tax Convention on Income and on Capital. The first OECD Model was published in 1963 and the Model has been revised regularly since that time. The latest version of the OECD Model was published in 2017.

4

UN Model Double Tax Convention between Developed and Developing Countries. The first UN Model was published in 1980 and the Model has been amended several times since that time. The latest version of the UN Model was published in 2017.

5

This may require a modification to the instrument of ratification.

Ukraine: Technical Assistance Report-Distributed Profit Tax; Voluntary Disclosure of Assets; and BEPs Implementation
Author: International Monetary Fund. Fiscal Affairs Dept.