Republic of Armenia
Technical Assistance Report-Growth-Friendly Rebalancing of Taxes

With the appointment of a new government, a lively debate has ensued about redirecting fiscal policies in support of a balanced revenue-raising strategy that is conducive to investment and growth. Currently, Armenia needs to raise more revenues in support of fiscal consolidation and to generate additional funding for developing and maintaining the physical infrastructure with special reference to the need of improving the urban built-up environment. Since the Authorities requested the mission to consider tax measures that are supportive of growth and/or tradeable sector, the proposed restructuring of taxes recognizes that real estate taxes, resource rent taxes, and broad-based consumption taxes (VAT and excises) are least distortive for growth. The 2016 Technical Assistance Mission in its report reviewed already unutilized tax bases as far as excises are concerned (taxation of gambling, mobile air time, waste packaging taxes, alcohol and tobacco taxation). As requested by the authorities, this mission focused on improving personal and business income taxes, presumptive taxation, and the recurrent real estate tax as base-broadening of the latter could support the fiscal program of the new government.

Abstract

With the appointment of a new government, a lively debate has ensued about redirecting fiscal policies in support of a balanced revenue-raising strategy that is conducive to investment and growth. Currently, Armenia needs to raise more revenues in support of fiscal consolidation and to generate additional funding for developing and maintaining the physical infrastructure with special reference to the need of improving the urban built-up environment. Since the Authorities requested the mission to consider tax measures that are supportive of growth and/or tradeable sector, the proposed restructuring of taxes recognizes that real estate taxes, resource rent taxes, and broad-based consumption taxes (VAT and excises) are least distortive for growth. The 2016 Technical Assistance Mission in its report reviewed already unutilized tax bases as far as excises are concerned (taxation of gambling, mobile air time, waste packaging taxes, alcohol and tobacco taxation). As requested by the authorities, this mission focused on improving personal and business income taxes, presumptive taxation, and the recurrent real estate tax as base-broadening of the latter could support the fiscal program of the new government.

Executive Summary

With the appointment of a new government, a lively debate has ensued about redirecting fiscal policies in support of a balanced revenue-raising strategy that is conducive to investment and growth. Currently, Armenia needs to raise more revenues in support of fiscal consolidation and to generate additional funding for developing and maintaining the physical infrastructure with special reference to the need of improving the urban built-up environment. Since the Authorities requested the mission to consider tax measures that are supportive of growth and/or tradeable sector, the proposed restructuring of taxes recognizes that real estate taxes, resource rent taxes, and broad-based consumption taxes (VAT and excises) are least distortive for growth. The 2016 Technical Assistance Mission in its report reviewed already unutilized tax bases as far as excises are concerned (taxation of gambling, mobile air time, waste packaging taxes, alcohol and tobacco taxation). As requested by the authorities, this mission focused on improving personal and business income taxes, presumptive taxation, and the recurrent real estate tax as base-broadening of the latter could support the fiscal program of the new government.

The mission’s proposals are cognizant of the growth and equity trade-offs. Also, the government in its tax review is advised to internalize the following key institutional lessons from successful comprehensive tax reforms across the world: the tax reform needs a positive rallying cry such as “raising revenues for service delivery” that should continue for the reform period of say five years. It would be helpful to implement gradual, incremental changes that are aligned to the overall reform theme and to get reform-backing from a high-profile Champion in the Executive. The Minister of Finance should prominently lead the tax reforms to stabilize expectations and provide for certainty amongst investors. It would be prudent to coordinate complex tax policy changes with capacity readiness of the State Revenue Committee. Finally, consider synchronizing tax increases with the roll-out of high-visibility public service delivery to gain buy-in from the public.

Personal income taxation (PIT) can and should perform its main function of generating revenues in line with the distributional fairness maxim. Given that about 30 percent of the population is poor with a gradually widening inequality, the authorities should ensure that the skewed income distribution is not worsened through the joint impact of the tax system. In adjusting the PIT system, it is therefore important to maintain the progressive rate structure to ensure adherence to equity and preserving taxation according to an individual’s ability to pay. With progressivity in mind, PIT reforms should lighten the tax burden of the low-income cohorts, while minimizing the revenue impact of the reform. Hence, a personal allowance or tax-free threshold equal to the global poverty line of AMD 30,000 is recommended. This is costly to the budget but if taxes were to be adjusted within a holistic framework, other indirect tax instruments such as excises (discussed in the 2016 TA Report) could be utilized to compensate adequately for suffered revenue declines that arise from taking a large cohort of low-income taxpayers out of the PIT net. Importantly, the flat tax (PIT) should be avoided as it exacerbates income inequality by unduly benefiting high-income taxpayers at the expense of low- and middle-income households.

The Flat Tax’ negative distributional consequences are compounded by the sizable erosive impact on tax revenues if top PIT tax rates are lowered to a lower proportional flat rate. Less tax revenues would be available, possibly resulting in cutbacks on pro-poor expenditure programs. Revenue administration is important too. If the revenue administration continues successfully to close loopholes and enhance compliance (e.g., by requiring that all wage earners and sole proprietors file simplified and pre-filled tax returns that can still be audited), these additional revenues should be allocated to advance tax burden relief of low-/middle-income households.

The basic choice facing Armenian’s policy makers is whether to continue with a corporate tax (CIT) system that imposes, compared to the region, a relatively high statutory rate with several differentiated tax rates of the incentive regimes. Alternatively, one could revamp the system through base broadening measures that would afford a lower and more uniform rate structure. To make the CIT system simpler, fairer and less distorting in its impact on investment choices, the mission recommends a reform path that rationalizes CIT preferences, repeals tax allowances, and removes tax rate differentials within and across sectors of the economy. Adopting a uniform rate would be much more neutral in its impact on investment decisions, which would largely restore a level playing field for investors. Further, a tax system with a uniformly applied tax rate is less susceptible to aggressive tax-planning opportunities such as domestic transfer pricing.

With regard to the special or presumptive tax system—encompassing the revised patent fee, the turnover tax, and the family company regimes—the pertinent design weakness is that this special tax system can stunt the growth potential of small businesses. It imposes attractive relief through a turnover tax or flat charge which results in a bunching of firms under the turnover threshold. But for start-ups with accounting losses, it still taxes the first dram of turnover in that it disallows loss carry-forwards and tax deferrals. Income splitting is widespread in order not to migrate into the general tax system with a higher tax and compliance burden. A growth-oriented presumptive tax system should, however, reduce the compliance burden for start-up firms and should impose an equal, if not higher, tax burden as encountered in the general regime. This would encourage firms’ growth beyond the turnover threshold as the regime should not be a refuge for high-income taxpayers, splitting their income to benefit from a presumptive regime’s low tax burden.

Unfortunately, the recently revised presumptive regime does not address this fundamental shortcoming. Furthermore, the mission advises that VAT payers, legal persons, and certified professional service providers should never qualify for being taxed under the special tax system. Also, taxpayers should no longer be able to annually elect whether they want to operate in the presumptive or standard regime as this invites aggressive tax planning.

The taxation of real estate in Armenia has the potential to generate more revenue. Communities require additional revenue to provide services, maintain infrastructure, and upgrade and maintain poorly-maintained multi-apartment buildings. Currently, collections from the land tax and property tax on buildings generate inadequate revenue because the tax is assessed on cadastral values that are well below market values, a high country-wide value threshold, and a too low, centrally determined tax rate structure. Attempts to move cadastral values closer to market values as a step towards the implementation of a market value-based system are supported. In the interim, agricultural land values must be reviewed and commercial farms as well as vacant urban land should be more heavily taxed.

This Technical Assistance Report (TAR) comprises an executive summary and four chapters. Chapter 1 presents background to the current technical assistance input and an introductory overview of the authorities’ tax rebalancing strategy. Chapter 2 discusses government’s proposed adjustments to personal income taxation. Chapter 3 reviews the corporate income tax and the role of presumptive taxes. Chapter 4 presents options for increasing revenues from the taxation of real estate. The Report’s numerous appendices inform about the justifications for the previous TA Reports’ key recommendations, such as: the future of a market-value based property taxes; tax expenditures and cost-benefit analysis models; and a high-level scoping of Armenia’s involvement in the Inclusive Framework to begin the implementation of the four BEPS minimum standards and Armenia’s commitments in terms of the Automatic Exchange of Information framework.

The Report’s recommendations are recorded in Table 1.

Table 1.

Armenia: Summary of Recommendations

article image
article image
article image

I. Comparative Tax Structure and Introducing Pro-Growth Tax Reforms

A. Background to the Technical Assistance Request and Advice

1. The authorities wished to be advised on the promotion of growth-friendly tax policy reforms and a review of the immovable property tax. In particular, this entailed the following analyses: (1) the design of the profit tax (rates, tax base, tax expenditures, and development of standardized cost-benefit analyses for tax incentives); (2) personal income tax (PIT) with a review of the tax rate structure and tax-free thresholds; (3) an assessment of presumptive tax regimes (patents and turnover taxes) and reform attempts since the 2015 tax policy (TP) Technical Assistance (TA) mission suggested major structural reforms in this regard; and (4) fuller utilization of the recurrent property and land taxes. Finally, the authorities expanded the TA Mission’s terms of reference to include TA advice on BEPS-related implementation of minimum standards and required capacity in international standards of exchange of information.

2. Armenia is at a historic crossroad. The peaceful change of government three months ago created opportunities for much-needed structural and governance reforms. The new government aims at dismantling oligarchic structures, removing barriers to domestic competition, restoring a level playing field in the economy, improving the business and investment environment, ultimately targeting a market-led economic growth strategy.

3. Currently, Armenia needs to raise more revenues in support of fiscal consolidation and to provide more funding for capital (physical infrastructure) and social spending. Armenia’s General Government Revenue—especially its tax revenue—records significant growth since 2004 (Table 2). This was predominantly driven by persistent annual increases of collections from taxes on income, profits and capital gains. Yet, the level of government revenue collection is relatively low compared to Armenia's immediate peers as well as other comparator groups1 (Figure 1). Efforts are needed to bring down public debt that exceeded 55 percent of GDP in 2017 and the budget deficit, ambitiously targeting 2.7 percent of GDP in the 2018 budget (down from 4.8 percent in 2017). Poverty and inequality are socially pressing issues that require concerted policy efforts on all fronts. As such, the tax system needs to effectively balance the country’s broad range of social and development objectives against the need to create an investment-friendly business climate. Tax policies should, therefore, be sufficiently nuanced to promote economic growth, pursue distributional fairness, but also remain regionally competitive—all being difficult tradeoffs. Table 3 compares Armenia’s tax structure against her immediate peers, being mostly the member states of the former Soviet Union.

Table 2.

Armenia: General Government Revenue (percent of GDP), 2004 -2016

article image
Source: GFS General Government and/or Fiscal File.Note: “…” fields represent missing values.
Figure 1.
Figure 1.

Total Government Revenue as percent of GDP

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF, World Economic Outlook
Table 3.

Armenia: Regional General Government Revenue by Source (percent of GDP) 2016

article image
Revenue Source: GFS. GDP Source: WEO.Note: Tax Revenue does not include Social Contributions. “…” fields represent missing values.

4. This TA mission’s review and outputs are preceded by the 2015 FAD Technical Mission that analyzed the Authorities’ tax policy proposals underlying the 2015 Draft Tax Code with its subsequent 2016-promulgation. That TA mission’s recommendations covered approaches to tax reform, taxation of employment and capital income, taxation of business income and protecting its base against tax avoidance schemes, the VAT and the turnover-based presumptive business tax regimes, the recurrent property tax, excise taxation and VAT base broadening. Finally, the 2016 TA Report made suggestions on how to embrace the revenue potential of environmental taxes. Consequently, the current mission’s advice and proposals should be read together with the analysis and proposals contained in the 2016 TA Report.2 For ease of reference, Appendix 1 provides a summary of the 2016 TA advice but key policy discussions of the 2016 Report will not be repeated.

B. Comparative Tax Structure, Investments and Growth Behavior

5. Pursuing greater simplicity of the tax system is advised (especially with regard to the presumptive regimes). The Armenian system distorts the level-playing field for investors by imposing non-uniform effective tax rates on different businesses, depending on their size, residency, business activity, or location of their investment. Preferential tax treatment on the level of individual companies and products is widespread, while compliant businesses within the tax net face significant compliance burdens. Informality thrives for a number of reasons, with complexity in the tax provisions being leading causes. Large variances in effective tax rates and the remaining loopholes in the tax system aid opportunities for tax arbitrage. Overall, the fundamental requirement of a business- and growth-friendly tax system—i.e., neutrality with respect to investment decision-making—is compromised.

6. Armenia's tax system may have a significant impact on capital investment decisions. While non-tax factors, including geopolitical ones, had a role in constraining private sector development, the complexity of some tax instruments, the lack of predictability, and discretionary tax-related decision-making, may have contributed to discouraging investments given these investor uncertainties. Indeed, after the steep plunge following the world financial crises, gross capital formation as a percent of GDP has stagnated over the last several years (Figure 2). Equally, the inbound foreign direct investment (FDI) stock declined significantly in the aftermath of the Russia crisis (post 2014) and has been slow in recovery thereafter (Figure 3). This trend is far more pronounced relative to the performance of the FDI stock in the comparator countries.

Figure 2.
Figure 2.

Investment, as percent of GDP

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: Source: IMF, World Economic Outlook
Figure 3.
Figure 3.

Inbound FDI, as percent of GDP

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: UNCTAD

7. Given these pressures of attracting FDI, support for economic growth without ignoring the distributional fairness imperative, requires a comprehensive view on the tax system. Since Armenia’s tax structure (2016) reveals some unique features compared to the peer group (see Table 3 and Figure 4).; most of the tax instruments should be included in the review to afford a revenue-neutral rebalancing.

Figure 4.
Figure 4.

Armenia: Regional General Government Revenue by Source (percent of GDP)

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF, World Economic Outlook, and OECD

8. A OECD study3 reviewing the impact of different tax instruments on per capita growth found a relative strong relationship between taxes, investment and economic growth. The study assessed the design of a tax structure and whether it would be conducive to economic growth. It provides a “tax and growth” ranking of taxes, confirming results from earlier literature but providing a more detailed disaggregation of taxes. Corporate taxes (CIT) are found to be most harmful for growth, followed by personal income taxes (PIT), and then consumption taxes. Yet, no country can do without a CIT given the revenue needs and for the fact that generated economic rents by companies can be taxed heavily without affecting behavior. There are other good reasons for imposing CIT and PIT. Recurrent immovable property taxes appear to have the least negative impact on growth—but are costly to administer given the difficulties with registration of properties, their required periodic revaluation, and their general unpopularity with taxpayers. In addition, central government cannot simply substitute its income taxes with recurrent taxes on immovable property since in most countries property taxes are the exclusive tax domain and revenue source of sub-national governments. Also, only a small number of countries raise significant revenues from property taxes (Tables 3).

9. Against this background, some pertinent features of Armenia’s tax structure, possibly informing future tax base and rate adjustments, need to be highlighted:

  • A heavy reliance on indirect taxation (i.e., broad-based consumption taxes) which is conducive for economic growth: in Armenia, 42 percent of tax revenues are generated from indirect taxes vis-à-vis the comparators’ 38 percent.

  • PIT in Armenia raises 6.5 percent of GDP—significantly higher than the regional average of 3.5 percent. This is explained by Armenia’s discontinuation of the social security contribution system in 2013 and assuming these requited payments into the PIT regime (see Chapter 2).

  • Armenia’s CIT revenues of 2.5 percent vs. the comparators’ average of 2.6 percent of GDP.

  • Property tax in Armenia raises revenues of 0.5 percent of GDP (mainly on vehicles and to a much lesser extent on immovable property), against the regional average of 0.6 percent.

  • The revenue productivity of VAT is close to the world’s average (Figure 6) but Armenia’s VAT C-efficiency of 0.47 vs. the average 0.59 for comparators (Table 4), suggests that Armenia could still do further work on closing the VAT policy and compliance gaps.4

Figure 5.
Figure 5.

VAT Rates

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF, VAT rates database
Figure 6.
Figure 6.

VAT Productivity

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF, World Economic Outlook and IMF Staff
Table 4.

Comparative CIT Productivity and VAT C-Efficiency, Latest Available Year

article image
Sources:CIT Productivity: CIT Revenue from WEO/GFS (OECD and Eurostat, where data available), GDP from WEO; CIT Rates from FADTP Rates Database (Internal)VAT C-Efficiency: VAT Revenue data and C-Efficiency ratios from FADTP (Internal C-Efficiency Database) and GFS (OECD and Eurostat, where data available), Consumption Data from WEO, VAT Rates from FADTP Rates Database (Internal)VAT Productivity: VAT Revenue data from GFS (OECD and Eurostat, where data available), GDP Data from WEO, VAT Rates from FADTP Rates Database (Internal)Note: “…” fields represent missing values. Formulas:CIT Productivity = (CIT Revenue as % of GDP) / (CIT Rate)VAT C-Efficiency = VAT Revenue/(Total Final Consumption net of VAT Revenue * VAT Rate)VAT Productivity = VAT Revenue % GDP/VAT Rate (%)

10. With growing recognition of the tax system bottlenecks in Armenia, there is a need to thoroughly analyze the impact of tax policy options. To be effective in evaluating the various tax policy proposals, it is essential that tax reform options are carefully assessed, quantitatively analyzed, and openly debated. This requires that decision-makers and all stakeholders in the debate have access to the best available data and independent, evidence-based analysis, including information about the impact of tax reforms on revenue, the income distribution, and the economy. There is a critical need for the MoF to play a leading role in explaining the economic rationale and intent behind changes in tax policy and tax legislation. Global experience suggests that insufficient debate about tax reform options can lead to a tax system that fails the test of legitimacy in the eyes of the public.

C. Growth-Friendly Fiscal Measures

11. Governments often regard taxes as a policy instrument for stimulating economic growth to the extent its distortionary effects are minimized on factors that generate growth such as utilization of capital and labor. Before entertaining the roll-out of tax incentives, the mission wishes to caution that the relationship between taxes and economic growth needs more careful consideration. The above-mentioned tax structure analysis describes Armenia’s relatively greater reliance on indirect taxes, but a higher reliance on trade taxes (1.1 percent of GDP vs. the comparator group’s proportionally lower 0.8 percent of GDP, Tables 3)). Correcting the tax structure before introducing new tax incentives is, therefore, advice underpinned by the OECD’s empirical evidence provided above. 5

12. A revenue neutral, growth-oriented tax reform could include a partial shift of the future tax burden from income taxes to less distortive taxes such as consumption or recurrent taxes on immovable property. For this purpose, the Armenian tax policy design should guard the integrity of broad-based consumption taxes (VAT and excises) by cutting back on exemptions and seeking to raise more from these instruments. This could pay for a lowering of the basic CIT rate (which at 20 percent may be seen by investors as relatively high given the regions average statutory rate of 15.4 percent in 2017, see Figure 7).

Figure 7.
Figure 7.

Regional Comparators: Average Statutory CIT Rate, 1992 – 2017

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF, Fiscal Affairs Department, Tax Rates Database (Internal)

13. Since 2006, Armenia’s CIT rate became less attractive than the regional average rate. Thus, broadening the CIT base—which could afford a gradually lower CIT rate in a revenue neutral fashion—could signal a more inviting investment environment. Also, it would provide less incentives to embark on tax avoidance as lower rates make this endeavor much less rewarding.

14. The OECD paper suggests that a restructuring of taxes within each of the broad tax categories could secure significant efficiency gains. Realistically, a greater revenue shift could probably be achieved into consumption taxes—but it has distributional consequences. Since consumption taxes are less progressive than PIT, or even regressive, transferring the tax burden from PIT to consumption taxes would reduce progressivity. Similarly, shifting revenue reliance from CIT to consumption taxation would increase share prices and wealth inequality as well as increase income inequality by lowering capital income taxation—and make income and wealth distributions more skew. These are not insignificant trade-offs in tax policy design but should be tackled soon as it is less costly than applying a plethora of blunt tax incentives. In the aftermath of the global financial crisis (2008 – 09), the IMF released key proposals on growth-friendly fiscal policies—see Appendix 2 for a detailed discussion.

D. Designing a Comprehensive Tax Reform Strategy

15. Armenia announced an overhaul of the tax system—yet, caution should be exercised with respect to its timing and the process to ensure sustainability of collections. The entire reform plan needs to be presented as a package of "give and take”. The amendments should work towards a tax policy consolidation period whereby amendments are kept to a minimum with the view to stabilizing the tax system and maximizing certainty for taxpayers. All significant tax policy proposals should be conditioned on the readiness of the State Revenue Committee (SRC) and Customs to implement such changes. Rushed tax policy changes could backfire if taxpayers misunderstand these or they cannot be effectively administered. In this regard, it would be useful to pay attention to the lessons and experiences with successful tax reform approaches (Bird, 2004) in developed and developing countries (see Box 1). Success is defined as achieving the intended result of the tax restructuring: i.e., attaining the revenue target, attracting investments (often the reason for introducing tax expenditures), stimulating growth, or enhancing fairness and tax morale:

Lessons from Successful Tax Reform Programs

  • The essential requirement for success is strong leadership and political backing for the Ministry of Finance in reforming the tax system. Strong political leadership is exemplified by appointing a Champion who is prepared to put his/her reputation on the line for the entire period of the tax reform.

  • The champion should be backed by the chief executive, i.e., the prime minister or the cabinet of ministers.

  • A holistic package of tax reform interventions, if implemented properly, can be more successful than piecemeal adjustments which lose sight of the bigger picture. Taking such systemic view can help short-term political pressures and counter the proliferation of vested interests.

  • A comprehensive tax reform program is to be preferred whereby “the pluses of the reform are real and significantly enough so that taxpayers would accept the minuses” that must go with it to make a reform program fiscally sustainable.

  • The more transparent tax consultations and discussions are, the more successful reforms could be but it also means that the appointed champion (the executive or ministry responsible for taxation) maintains throughout the reform process its leadership position and drives it with passion and a sustained pace.

  • The MoF should be supported in the tax reform program by a strong tax policy team (tax policy unit6 or directorate), which integrates tax policy analysis with revenue estimation, and legal drafting. This team should also draw on private tax expertise.

  • The tax reform program should be communicated through all available media and the taxpaying public should be continuously educated about the need for the reforms—and this is for the whole duration of the reform.

  • In order to create a positive message about the unfolding tax reform initiative, the authorities should select a sequence of reforms, beginning with the generation of early revenue gains or a fairness-enhancing rebalancing. It would address the short-term nature of political goodwill by the electorate and could be part of so-called comprehensive gradualism in tax reform.

  • Consolidate one reform before attempting the next phase or tax intervention—providing sufficient time for the SRC to prepare IT systems and train assessors adequately.

  • Crisis-driven or hurried reforms due to poor preparation tend to fail, even though they may contain a solid idea.

  • Importantly, tax reforms tend to be more successful when they involve simplification of tax systems and legislation with accompanying lower compliance and administrative costs—again, the readiness of the tax administration is of central importance.

  • Paying sufficient attention to the implementation, proper legal drafting of tax amendments with guideline notes (secondary legislation), IT system adjustments, the training of tax administration officials, and the printing of new tax return forms, will contribute to a successful reform.

  • Continuity among decision-makers responsible for tax policy and implementation aids successful tax reforms.

  • Good reforms exclude quick fixes such as premature tax amnesties before the tax administration has built sufficient competence; tax lotteries to force honest recording of turnovers, and the issuance of VAT invoices, or presumptive taxes that turn into tax relief measures.

  • Careful synchronization of incremental reform steps is ensuring a measure of success.

  • Tax structure reform is best accompanied by modernization of the tax administration and expenditure policies.

  • Tax reform should take account of unique conditions and limitations of a country.

  • During the reform, considerable thought and planning should go into transitional provisions to ensure sustainability and credibility of the reform.

The Political Economy of Tax Reform

16. The challenge for the new government is to deliver highly visible and substantive improvements to the quality of life of the population in the aftermath of the Velvet Revolution. Modern democracies view paying taxes as a social contract between citizens and the state (i.e., state-building) by which the citizens pay their dues in return for public services. Tax reform initiatives should therefore enhance tax morale, which is the willingness to accept say a more revenue-productive property tax in exchange of improved public services under the banner of Revenue for Service.7 If taxes are well designed, effectively collected, and prudently spent on reliable public services, taxes could indeed support economic growth. Similarly, proper tax design should advance distributional fairness—granting selective tax concessions achieves the opposite.

17. Thus, the Armenian tax reform message should be linked to these social compact cornerstones. But the message is only credible if budget outlays are indeed reducing extreme inequalities and if they would fund the post-revolution expenditure priorities. It could be the rallying cry for the tax reforms to get buy-in from skeptical taxpayers. The mission, in its discussion with stakeholders, received unanimous feedback that increases in taxes—such as a more revenue-productive property tax—would be rejected as taxpayers see no commensurate benefit returned in the form of municipal service provision. This suggests the attraction of benefit taxes for any future tax reform, such as road use taxes, environmental charges, and recurrent property taxes. To increase taxes in an environment of low tax morale is very difficult even though it has become a burning issue for providing sufficient resources to local communities for the refurbishing backlog of condominiums. However, taxpayer resistance could be managed down if local communities would commit to allocate a set percentage of property taxes towards improved maintenance of more than 55 percent of the condominium housing stock of Armenia. This would, for example, constitute a tax reform initiative with high visibility.

18. The tax reform program should be cognizant of the effects of a currently corroded social contract and firm belief in the society that the state and service delivery have little credibility. However, the political goodwill period of waiting for improvements in the tax system and the livelihood of especially the poor executed by the New Government may last only a few hundred days. Hence, given the urgency in changing perceptions about the effectiveness of the new government some overarching tax reform guidelines should be considered:

  • Visible improvements in service delivery or alternatively, material changes to the tax burden of low and middle- income households through a rebalancing of taxes and tax rates.

  • Some taxes would require restructuring to enhance their acceptability in terms of fairness, efficiency, predictability, and serving the interest of the majority rather than the select few.

  • Consequently, consider introducing the concept of gradual, predictable, well-communicated decrease of the tax burden of low-income households which can be afforded by broadening of the tax base due to the rationalization of tax privileges and better tax compliance.

Roadmap for a Short- to Medium Term Tax Reform Program

19. A comprehensive review on the tax system is critical: all the tax instruments should be included in the reform to afford a revenue-neutral rebalancing. The recommendation of preparing a package of tax measures is based on IMF FAD policy advice to many countries, emphasizing that policies must be designed and backstopped with realistic revenue estimates. Formulating a grand structural tax change and implementing it incrementally over say five years will more likely succeed. Given the constraints in revenue administrations, incremental tax changes within a thoroughly researched policy framework are about the only way to success. Overly ingenious ideas and a high reliance on discretionary administration should be avoided.8

20. A medium-term tax reform strategy should stabilize tax collections and improve certainty for investors by reforming tax instruments in a synchronized manner. Elements of consultative processes in relation to the introduction of a Comprehensive Tax System Reform Plan can be stylized as per Figure 8. As for the current tax reforms, the authorities have prioritized adjustments to PIT, CIT, and a modern property tax—with VAT and excises providing compensatory revenues if afore-going structural reforms may result in initial revenue losses. The mission addresses the following elements of such tax reform strategy:

  • Extend the coverage of the PIT—increase its progressivity by taxing high incomes at progressively higher rates—as addressed in Chapter 2.

  • Design a CIT that is broad-based to allow for a uniform, statutory rate at regionally competitive levels. Having in the CIT regime 0 percent and positive tax rates invites domestic transfer pricing practices with elevated revenue leakage risks as the anti-transfer pricing capacity in the SRC is rudimentary but developing. This is exacerbated by the absence of binding regulations on how to prepare transfer pricing documentation9 (see Chapter 3).

  • Avoid tax incentives that jeopardize revenue and good governance, are hard to reverse, or generate no offsetting social benefits (see previous 2016 FAD TP advice). Instead, the mission wishes to re-emphasize the FAD TP advice in 2011 that attractive accelerated depreciation allowances encourages investments more cost-effectively.10 Similarly, the full taxation of agriculture remains a high priority. Convincing taxpayers about the need for this intervention would, however, need synchronization with ramped-up support to farmers in the form of improved infrastructure as the political resistance against such adjustment is palpable. The cost of tax expenditures and pragmatic cost-benefit analyses of tax incentives is discussed in Chapter 3 and Appendix 4.

  • As previously advised (2016 TA Report: 37–47), the Special Tax Regimes for taxing business income—the Turnover Tax System, the Family Entrepreneurship System, and the Patent Tax System—require simplification, rationalization and reduction of tax benefits. Globally, the purpose of such regimes is to reduce the regressivity of tax compliance burdens for small and medium-sized businesses.11 The intention is not to provide tax relief to these classes of taxpayers. The Special Tax Regimes should encourage businesses to grow and should not serve as a refuge for high income taxpayers to avoid tax—through say income splitting. Unfortunately, the current system falls short in all these areas. Legal entities and all registered professional service providers should not benefit from the Special Tax Regimes’ privileges, nor should VAT payers—see Chapter 3.

  • The authorities’ indicated tax reform plan for the PIT system is a revenue loser which may require a VAT rate increase. Given the already high VAT rate, reforms should focus on how to improve Armenia’s VAT C-efficiency by addressing the policy gap (exemption creep) and compliance gaps (it may require as a first step a VAT gap analysis by FAD).

  • Given these same revenue risks, enhance collections from imposing excises on untaxed activities such as gambling, betting, and mobile phone use—see 2016 TA Report: 55–64.

  • Using a property assessment value that is closer aligned to the market value, consider various options for introducing an adjusted value-based or unit area-based immovable property tax to generate additional resources for the financing of local government—Chapter 4.

  • For the outlined tax reform strategy, the overarching imperative should be the pursuit of simplification in tax legislation—i.e., avoiding excessive rate differentiation in the patent regime or immovable property tax.

  • The mission, on request by the authorities, suggests ways to strengthen the capacity to deal with profit-shifting by developing guideline notes on transfer pricing documentation—and accelerate compliance with the four minimum standards under BEPS—see Appendix 9.

Figure 8.
Figure 8.

Circular Processes followed in a Comprehensive Tax Reform

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Key Tax Reform Steps and Advice

  • A comprehensive package of tax reform measures should be announced upfront, where after incremental implementation of individual tax instrument adjustments can ensue.

  • For the sequencing of reforms, develop a communication theme that includes “revenues for service” and target fairness-enhancing rebalancing of taxes.

  • Seek Cabinet support and approval for the rationalization of tax incentives.

  • For purposes of the PIT reform, and with revenue neutrality in mind, design compensatory tax measures under the excise and VAT regimes.

  • Consolidate one incremental tax reform, before embarking on the next adjustment.

  • Synchronize carefully tax increases with the roll-out of visible public service programs.

  • Pay sufficient attention to proper planning for implementation, timely legal drafting, release of guideline notes, preparation of new IT systems, and training of administrative staff.

II. Taxation of Employment Income

A. Overview

21. There are commendable elements in Armenia’s PIT system: it is generally simple, transparent, and relatively easy to administer. The number of exemptions and deductions is minimal; and the system relies to a large degree on final withholding arrangements. With a three-brackets progressive rates structure, the current system attempts a high degree of progressivity. However, as shown further in the section the system is heavily skewed towards the lowest income distribution, with a largely flat effective tax rate structure that achieves only a minimal level of progressivity.

22. In 2017, PIT generated AMD 314 billion—equivalent of 6.1 percent of GDP. This performance is seemingly impressive if compared to PIT revenue figures of Armenia’s comparator group. Indeed, in 2016, the comparator countries collected a mere 3.5 percent of GDP in PIT (Table 3). However, when the revenue of the PIT is combined with social security taxes, Armenia’s PIT wedge appears very low when contrasted against the comparator group. In 2016, the comparator countries collected 12.4 percent of GDP—against Armenia’s 6.5 percent of GDP—in combined PIT and social security taxes (Figure 9). The bulk of PIT revenue—92 percent of total—is derived from the taxation of employment earnings, withheld at source. Investment income contributed a little over 3.2 percent of total PIT revenue in 2017.

Figure 9.
Figure 9.

Personal Income: Tax Wedge as percent of GDP

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF, World Economic Outlook, and OECD

23. Resident taxpayers are taxed on worldwide income; nonresidents are taxed on income received from Armenia sources only. A standard physical presence test is used to qualify for residence status, with physical presence of at least 183 days. It is not clear—at least in an unofficial English translation of the Tax Code provided to the Mission—whether the minimum period pertains to the calendar/tax year or any consecutive 12-month period.

24. The PIT system is schedular, with a progressive tax on employment income, and a flat tax on capital income. Wage income is taxed based on a three-rate progressive structure (Table 5). There is no personal allowance; the system is currently taxing those on or below the poverty line, and on the minimum wage. Capital income, such as dividend and interest, is taxed at a flat rate of 10 percent through final withholding. Rental income is taxed at a 10 percent, with additional 10 percent assessed if the rental income exceeds AMD 58,35 million.

Table 5.

Personal Income Taxation, July 2018

article image
Sources: Country Surveys International Bureau of Fiscal Documentation (July 2018), and the Tax Code of the Republic of Armenia (unofficial English translation).

25. There are two pillars in the Armenian pension system: a budget-financed defined benefit component and a “voluntary” pension fund component, in place since 2014. From July 1, 2018, contributions to the “Voluntary Pension Fund”—Pillar II—are mandatory for all employees born after January 1, 1974, including employees of the private sector.12 Originally, the employee contribution rate to their individual retirement accounts was set at 5 percent of income (up to monthly maximum of AMD 25,000) with government making a matching monthly contribution. In June 2018, as a temporary measure, the rates of contribution to the individual Pension Fund accounts were changed to 2.5 and 7.5 percent by employee and government respectively, to mitigate the impact on the disposal income of private sector employees13 that were mandated to join the pension system, effective from July 1, 2018.14 Linked to the current labor income, the Pension Fund plan is intended to provide the income replacement (consumption smoothing) benefits to the Armenian working population. While pension contributions are a burden on labor incomes, they are distinct from taxes as they will eventually lead to future pension income, entitling individuals to future benefits.

26. The distribution of employees and employment income is heavily skewed toward low- to middle-wage earners, with at least 79 percent of employed taxpayers earning less than the average monthly salary of AMD 194,00015 in 2017 (Figure 10). Cumulatively, over 65 percent of all employees are taxed under the first PIT bracket; they earn 32 percent of all employment income. The second bracket—between AMD 150,000 and AMD 2,000,000—contains about 35 percent of all employees and is responsible for the largest share of the taxable income—about 62 percent. Finally, a mere 0.3 percent of all employees fall into the top PIT bracket. Their income, 6 percent of all taxable employment income, attracts the highest tax rate—36 percent.

Figure 10.
Figure 10.

Distribution of Employees and Employee Income, 2017

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Sources: State Revenue Committee, IMF staff estimates

27. The PIT system can and should perform its distributional function to alleviate the tax burden for the poor and vulnerable sections of society. In Armenia, PIT is not performing this role as well as it could. In a context where about 30 percent of the population is poor (Figure 11), and the inequality is widening (evidenced by the rising levels of the Gini coefficient), authorities should be seeking to ensure that the already skewed income distribution (Figure 10 above) is not exacerbated by the tax system.

Figure 11.
Figure 11.

Poverty Levels: Inequality

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: Statistical Committee, “Social Snapshot and Poverty in Armenia.

28. The system of collection and assessment of PIT in Armenia is dependent on a withholding regime. Under this arrangement, the obligation to withhold an amount of tax is imposed on independent third parties (withholding agents), e.g. employers and financial institutions. The tax withheld is final. Even when a taxpayer earns income from multiple sources, as long as the earned income is subject to taxation at source by tax agents, the taxpayer has no obligation to recalculate the tax due and submit a final income tax declaration form. As a result, the government under-collects the tax due on earned personal income, as each income source is treated independently of others for taxation purposes,16 creating unfairness along the way.

B. Analysis of Revenue and Distributional Impact of PIT Reforms

29. A major PIT system reform is envisioned in Armenia; several proposals for reform are currently evaluated. As discussed in Chapter I, the most important consideration for the current Armenian reform agenda is a holistic view of the tax system that recognizes the importance of revenue sustainability, and considers the tax reform in its entirety, within a comprehensive package of “give and take” tradeoffs. Such “rebalancing” view is propagated by the key stakeholders, including the MoF and SRC. It is also shared by the private sector representatives this mission had a chance to interview. The analysis of PIT reform outcomes that follows were guided by such comprehensive approach to the tax system (discussed in Chapter I).

30. Several objectives must guide the assessment of PIT reform options. Among them, progressivity and distributional fairness of the tax system take center stage due to the uniqueness of PIT as a tax instrument for delivering vertical and horizontal equity. Indeed, the tax reform initiatives in Armenia provide an opportunity for alleviating the tax burden for the poor and vulnerable sections of society. Other important considerations include discouragement of the informal economy, as improved perception of equity could reap benefit of stronger taxpayer morale and improved compliance. A consideration of labor market competitiveness relative to other countries in the region is also important in the context of Armenia due to historically significant levels of emigration of high income earners; i.e., the so-called “brain drain”.

31. Using employment income data for 2017, the mission built a simulation model to assess several employment taxation scenarios against revenue performance as well as the vertical equity and progressivity criteria.17 The key parameters defining five major simulated scenarios are summarized in Table 6. The current PIT regime, as of July 2018, with a progressive rate structure (23, 28 and 36 percent) was first compared with an alternative Scenario 1: a widely discussed option of taxing individual income under a two-rate progressive regime, with the income threshold at AMD 500,000 and rates of 20 and 25 percent applicable to incomes below and above the threshold (Table 6).

Table 6.

Alternative PIT Scenarios

article image
Source: IMF Staff estimates, based on data provided by SRC.

32. Next, a threshold, either in the form of a general tax credit, a basic deduction, or a zero-tax bracket was simulated to support tax progressivity and equity objectives by reducing or eliminating the tax burden on people with the lowest incomes. In Scenario 2, an alternative progressive structure that incorporates a tax credit equal to AMD 10,45018 was tested with rates identical to the prevailing statutory rates (23–28-36). Further, Scenario 3 is simulated, with personal allowance set at the level equal to the global poverty line; AMD 30,00019; the personal allowance is phased out once incomes reach the level of AMD 100,000. Under Scenario 4, a personal allowance equal to the prevailing minimum monthly wage—AMD 55,000—was offered, to be phased out at the income levels higher than AMD 100,000. Finally, a 4-tier structure with a tax-free threshold equal to AMD 55,000 (prevailing minimum monthly wage) and the current tax rates (0–23-28–36) was assessed in Scenario 5.

33. The revenue contraction under each of the five alternative scenarios was assessed, (Table 6), under static analysis, assuming no behavioral response by the taxpayers. Both, a percentage change, as well as an absolute AMD amount of yearly revenue change was evaluated in comparison with the prevailing PIT regime. Based on the simulation results, Scenario 5 that offers a tax-free threshold for all PIT taxpayers is the costliest option under consideration, with a total yearly revenue loss of AMD 87 billion (equivalent to 1.4 percent of GDP). The option with the least revenue contraction is Scenario 3 under which is a monthly personal allowance of AMD 30,000 is offered that is subsequently phased out at the level equal to AMD 100,000.

34. To allow for a possibility of an expanded tax base as a result of behavioral changes of the taxpayers, a potential base broadening element was incorporated into the model. These behavioral changes could be a reflection of two main tendencies: (1) the taxpayers incentivized to exit out of shadow into the formal tax regime due to reduced effective tax rates; and (2) better tax compliance due to improved tax administration efficiency. The last row of Table 6 shows by how much the current tax base must be expanded in order to neutralize revenue loss under each of the assessed scenarios. For example, under Scenario 1 (with two marginal rates: 20–25 percent), the current tax base must be broadened by 35.7 percent for a revenue-neutral outcome. Similarly, a base expansion of 62 percent is required to aim for revenue neutrality under Scenario 2.

35. More importantly, to support informed tax policy-making, the incidence analysis of the tax reform alternatives was conducted. Two questions may be asked: “who is winning the most under each of the reform scenarios? Does the reform aim to relieve the tax burden on the most vulnerable; or will it serve the interest of a relatively well-off income group?

36. Figure 12 and Table 7 show the reduction in tax liability for taxpayers at all income levels under Scenarios 1, 2 and 3 discussed above. Figure 12 plots the absolute AMD value of the reduced tax liability relative to the prevailing tax system. The heat-map in Table 7 presents the reduction in tax liability as a percentage change of the current tax liability. For example, under Scenario 2, the 263,683 poorest taxpayers whose monthly income is less than AMD 100,000 will see a 76.2 percent reduction of their tax due (first row of Table 7). Similarly, under Scenario 3, the taxpayers with incomes exceeding AMD 300,000 will see no changes in their tax liability; only the taxpayers at the lower end of the income distribution will see their tax dues to go down by more than half, with a significant positive impact on their disposable income.

Figure 12.
Figure 12.

Winners of PIT Reform

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF Staff estimates, based on data provided by SRC.
Table 7.

Incidence of the PIT Reform

article image
Source: IMF Staff estimates, based on data provided by SRC

37. It is immediately clear that Scenario 2 is the most beneficial for taxpayers at the low end of the income distribution. At the same time, as discussed above, it is the costlier option among the three under consideration. Scenario 1, which reflects a widely discussed, two-tier, 20–25 percent option involves a significant revenue downturn and is the least favorable for the poor, instead benefitting the relatively better-offs. The least costly option—the one preferred by this mission on the equity and distributional fairness grounds—is the option under Scenario 3. Under this scenario the entire revenue gain is distributed between the poorest income earners. The taxpayers with monthly income of up to AMD 100,000 benefit the most, followed by those with income below AMD 200,000.

Vertical Equity and Tax Progressivity

38. The concept of vertical equity refers to fairness across a group of people with different incomes (or levels of wealth). While most people accept the premise that richer people should pay more tax than poorer people, how much more tax they should pay is often highly contentious. The term “progressivity” refers to the rate at which taxes increase (as a proportion of income or wealth) as income or wealth rises. A “progressive” tax is one in which the effective tax rate increases as income (or wealth) increases. A “regressive” tax is one in which the effective tax rate decreases as income (or wealth) increases. A “proportional” tax is one in which the effective tax rate remains the same as income (or wealth) increases.

39. Simulation results further show that the proposed two-tier structure (Scenario 1) affects distributional equity of the tax system. Figure 13 displays tax concentration curves, showing the cumulative proportion of taxes against the cumulative proportion of income-receiving units (using pretax income as the classifier). According to this measure, a tax structure is judged to be progressive if the tax is more unequally distributed among taxpayers than is pretax income, thus resulting in a tax concentration curve which lies below the Lorenz curve. The redistribution is less favorable under the proposed two-tier rate structure (Scenario 1), as well as under a flat tax scenario (Scenario 4, the line closest to red-dotted line). The highest degree of progressivity is achieved when the tax credit is introduced (Scenario 2). Scenario 3, with tax-free income threshold of AMD 30,000 is less progressive than a tax credit option but, displays considerably higher degree of progressivity than the two-tier Scenario 1 or a flat tax Scenario 4.

Figure 13.
Figure 13.

Distribution of Tax Burden under Alternative Tax Structures

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF Staff estimates.

A Word of Caution on a Flat Tax

40. During several stakeholder discussions, the mission received inquiries about possible introduction of a flat tax in Armenia as opposed to the progressive, multi-tier structures. Box 2 summarizes the key considerations in introducing a flat tax. To simulate impact of replacing the current progressive structure with a flat tax on labor income (while maintaining the existing structure for the taxation of capital income), the following two scenarios were simulated (Table 8). First, assuming no behavioral response to tax policy changes, a flat tax rate was simulated that would allow for revenue-neutrality of the tax reform. A flat rate of 25 percent was shown to ensure such outcome. Next, a scenario voiced during stakeholder discussions was tested—a flat tax of 20 percent. Revenue loss associated with this scenario is equal to AMD 53 billion under a static simulation. When a taxpayer behavioral response is simulated, to ensure revenue neutrality of the 20 percent flat tax regime, the personal income tax base must expand by 41.5 percent.

Table 8.

Flat Tax Scenarios

article image
Source: IMF Staff estimates, based on data provided by SRC.

41. Flat tax reforms are often associated with revenue losses. Among the Central and Eastern European countries that adopted flat-rate tax systems after Russia, the reforms generally seem to have caused a fall in PIT revenues that has not been fully offset either by changes in taxpayers’ behavior or by increases in other types of taxes. The introduction of a single personal income tax rate of 13 percent in the Russian Federation in 2001, was followed by an increase in real PIT revenues of about 26 percent in the first year after its introduction (Engelschalk and Loeprick, 2016). Yet, research suggests that the substantial increase in compliance was more the result of parallel efforts to strengthen tax administration (Ivanova, Keen, and Klemm, 2005). Similarly, tax revenue rose markedly in Georgia in the aftermath of the 2004 reform, which introduced a flat tax, but the revenue increase was most likely helped by draconian measures adopted by the government to reduce the inefficiency and corruption of tax administration.

Impacts of Introduction of a Flat Tax

The original flat tax (Hall-Rabushka, 1983) was a combination of a cash-flow tax on business income and a tax on workers’ income, both levied at the same, single rate (with a personal allowance available against the wage tax). However, most flat taxes that have been introduced use a looser definition as they refer only to personal income. The key impacts of the introduction of a flat tax system:

  • 1. Equity. So long as a flat tax has some basic exemption, the tax is progressive since the average rate of tax increases with the level of income. The more relevant question is whether it is more or less progressive than the tax scheme it replaces. The distributional impact of the flat tax reforms is commonly quite complex, and by no means unambiguously adverse for some of the least well-off.

  • 2. Work incentives. Marginal tax rates will fall for some and increase for other taxpayers, which might impact labor supply decisions. There is a vast empirical literature (Ivanova, Keen and Klemm, 2005) on the effects of tax reform on labor supply decisions. The broad consensus is that the effects of tax changes on the effort of primary workers are modest (reflecting offsetting, but perhaps large, income and substitution effects).

  • 3. Compliance, administration, and simplicity. There is indeed clearly an element of simplification in the flatness of a PIT, since this reduces the incentive to reallocate income across closely related individuals, makes withholding easier, and eases, for example, the need for income averaging for those with highly variable incomes. However, these effects could be nullified by the presence of the tax-free threshold, sometimes at quite high levels, resulting in two marginal tax rates. A recent study by the European Commission20 looking at the prevalence of undeclared work in the European Union finds that lower levels of undeclared work occur in Member States with higher per capita GDP, more modernized systems of government, higher levels of trust in authorities and lower levels of corruption, where social transfers are effective at reducing poverty, there are higher levels of public expenditure on labor market interventions to protect vulnerable groups. Further, the study finds no significant relationship between undeclared work and the implicit tax rate on labor.

  • 4. Automatic stabilization. The common argument that a flat tax weakens an automatic stabilization of the economy, upon which increasing reliance is generally placed in coping with shocks, might not be correct. The level of the threshold amount under the flat tax, below which incomes are not taxed, turns out to be crucial: if there is no threshold, then METR falls and the stabilizer weakens. However, with some taxpayers excluded from tax, the marginal tax rate applied under the flat tax will need to be higher than would otherwise be the case in order to raise the same revenue as the pre-reform tax. This tends to increase built-in stability.

42. The general concern of a flat tax is that it raises income inequality since it does not take into account the ability-to-pay considerations. Analysis of flat tax proposals suggests that they should be discouraged for Armenia as they clearly benefit the taxpayers at the higher end of the distribution at the expense of those at the lower end (Figure 14). In fact, a revenue-neutral 25 percent flat tax increases the tax burden on the poor, turning the better-offs into clear winners. The distributional fairness concern is compounded even further if there are attempts to substitute the revenue losses from the PIT reform with a one percentage point increase of the standard VAT rate (from the current 20 to 21 percent). Given the regressivity of VAT, this tax rebalancing has a more adverse impact on lower income cohorts vs. higher income households. The authorities are therefore invited to analyze the distributional impact of the VAT rate increase by utilizing household expenditure survey data as further explained in Appendix 3.

Figure 14.
Figure 14.

Winners and Losers: Flat Tax

Citation: IMF Staff Country Reports 2019, 031; 10.5089/9781484396674.002.A001

Source: IMF Staff estimates, based on data provided by SRC

Horizontal Equity and Tax Planning

43. Horizontal equity means that different taxpayers with the same income or assets pay the same amount of tax. The concept is simple and generally non-controversial in principle. But in practice, there can be significant dispute over what “same income” (or same consumption or same assets) means. A tax system that includes exemptions or special rules to treat economically similar taxpayers differently will not achieve horizontal equity. The multitude of tax regimes (see Chapter III), creates arbitrage opportunities across different types of income. The resulting tax burdens are widely divergent, possibly encouraging tax planning opportunities. Another important difference between the employee and self-employed regime rests in the requirements for payment of pension fund contributions. Currently, employed workers’ wages are subject to proportional payments of 2.5 and 7.5 percent21 of gross wage value for pension fund contributions payable by employees and government respectively; while self-employed individuals are not part of the system. This differentiation creates distortions. Considerations could be given to harmonizing the requirements for payment of pension contributions for self-employed individuals with those for employed workers. Further analytical work is however required to show the divergence of tax burdens across different types of individual taxpayers.

Final Reporting of Personal Income

44. It is important for Armenia to implement the final reporting of tax. It can be implemented through the provision of pre-filled tax returns to individuals. Such final tax reporting mechanisms have evolved to become a significant (and for some, transformational) element of revenue authorities’ e-services strategy worldwide. This is especially the case for the PIT, with almost half of revenue administrations using elements of this method. Pre-filing entails the use by revenue bodies of information held by them (e.g., taxpayer identity information, elements of taxpayer history, and third-party reports of income and deductions) to populate fields within tax returns which are then made available to taxpayers for verification. The completeness of the return sent to taxpayers is contingent on the range of third party reports that can be used by revenue bodies.

45. A pre-filled return can be forwarded to taxpayers in paper form or in electronic form. Electronically, taxpayers can access their pre-filled return via the Internet and, if required, advise of any necessary adjustments. Following the processing of confirmations and adjustments by the revenue body, final notices of assessment should be mailed to taxpayers, along with any refunds of tax owing to taxpayers. Under a ‘silent acceptance’ practice, taxpayers are not required to confirm that a return is correct in all aspects; instead, this is deemed to be the case if the revenue body receives no advice from the taxpayer after a prescribed period of time. Taxpayers are subsequently issued with an official notice advising their final assessment details.22

46. Very importantly in the context of Armenia, individuals must be subject to audit. Currently, the law allows audit of companies but not individuals (natural persons). Undeclared tax liabilities may result in high penalties and underreporting of a tax liability of at least AMD 2 million is treated as a criminal offence under the Criminal Code punishable by penalty or even imprisonment. However, taxpayers report their income through a self-assessment system and cannot be subjected to a tax audit, rendering such penalties ineffective. With sufficient information collection mechanisms in place, audits by tax authorities should act as an incentive for increased compliance and tax collection from individuals. The 2016 TA Report provided in the Supplementary Analysis and Legal Drafting Guide considerable background and draft legislative language as to how to broaden the audit to individuals, require self-declaration by all individuals, introduce measures against income splitting, and individuals avoiding wage taxation by incorporating themselves as independent contractors. The same concerns are still being raised by policy makers, suggesting that the 2016 TA Mission’s recommendations have been forgotten.

Recommendations

  • Maintain the current progressive structure of the PIT to ensure distributional fairness. Avoid introducing a flat PIT.

  • To improve progressivity of the PIT system while minimizing the revenue impact of the reform, consider introduction of a personal allowance at the level equivalent to the global poverty line—around AMD 30,000—to be phased out at incomes higher than AMD100,000.

  • Consider harmonizing the requirements for payment of pension contributions for self-employed individuals with those for employed workers.

  • Ensure horizontal fairness by aligning effective tax burdens on identical income earned by taxpayers in different sectors (regardless of being organized as natural or legal persons).

  • Communicate to taxpayers the benefit of better compliance, as additional revenue from the expanded tax base will be used for a gradual decrease in tax burden for all income groups.

III. Corporate Income Tax, Incentives, and Presumptive Taxes

A. Corporate Income Tax (CIT)

47. The Armenian authorities are looking for ways to reinvigorating investment in the country that would translate into more jobs and economic growth. Achieving a broader corporate tax base and thereby being able to raise the same revenues at lower, more uniform rates through elimination of tax preferences is feasible. The mission reviewed alternative policy options that may be superior to tax incentives in terms of leveling the playing field for all investors, while offering greater transparency, administrative simplicity, and cost effectiveness. Based on a corporate tax micro-simulation analysis, alternative tax policy scenarios were simulated that could provide sounder and more sustainable attractiveness to investments. This could be indeed achieved in a revenue neutral fashion.

The Current Investment Incentive Regime Provides Attractive Options

  • Allowance for wages of disabled employees—companies employing disabled can deduct 150 percent of the wages and other payments paid to such employees.

  • Incentives for free economic zones—the annual income of a resident company or operator in a free economic zone, is exempt from tax—being an unlimited tax holiday.

  • Government-approved projects—resident companies involved in business projects (excluding projects in the field of trade and finance) that are approved by governmental decree are granted a CIT exemption equal to 100 percent of the salary paid for newly established jobs. The exemption cannot, however, exceed 30 percent of the corporate tax payable for the current tax period. The exemption applies for a period of five full reporting years following governmental approval. Companies conducting governmental-approved construction and installation activities exclusively outside of Armenia are subject to a CIT rate of 5 percent. In addition, income paid to Armenian resident employees of such companies is taxed at a reduced rate of 13 percent. These rates apply from June 13, 2015.

  • Large exporters—companies or groups of companies that (1) exclusively export goods and services; (2) do not carry out business activities in the field of metal mining and the processing and sale of precious minerals and excisable goods; (3) receive the sales proceeds in foreign currency on bank accounts held in Armenia; and (4) have their business plans approved by the government, are taxed at the following CIT rates: 5 percent if the annual exports of the company or the group of companies exceed AMD 40 billion; and 2 percent if the annual exports of the company or the group of companies exceed AMD 50 billion (on the whole amount exceeding the threshold). This incentive lapses next year.

  • Information technology (IT)—resident companies involved in IT projects certified according to the Law of the Republic of Armenia "on IT sector state support" are granted a CIT exemption for revenues from IT activities. The exemption applies for a period of certificate validity. With effect from April 8, 2017, newly registered IT companies with up to 30 employees are exempt from CIT for a period of 5 years. The salaries paid to the employees of these companies are taxed at a flat rate of 10 percent.

  • Renewable energy production—licensed resident companies involved in production of electric energy from renewable sources are granted a CIT exemption for revenues from electric energy sales. The exemption applies to sales to licensed energy distribution companies.

Micro-simulation Modeling of Corporate Tax Returns

48. A micro simulation analysis (Box 3) of Armenia’s CIT returns was conducted based on the micro-, firm-level data provided by the SRC. Alternative CIT policy choices and their impact in terms of revenue generation were analyzed. More specifically, one of the key questions was: Subject to tax revenue-neutrality, how far could the corporate tax rate be lowered if all tax incentives were to be repealed? In other words, what is the tax rate that could be applied uniformly to a broadened tax base, conditional on the revenue generated by the new system being at least equivalent to the one generated by the current system?

49. The micro-analysis of the CIT system shows that it features a narrow tax base, non-uniform effective tax rates, and tax exemptions (Table 9). The narrow base is the result of a generous set of tax allowances. To this narrow base, different nominal tax rates apply to different investments within the corporate sector depending on the economic activity, sector, and size. It results in effective rates significantly being different from the statutory tax rate (e.g., an effective rate of 12 percent in the construction sector, or an effective rate of 17.4 percent in the transportation and storage sector). The resulting CIT is further reduced by profit tax exemptions.

CIT Micro-Simulation Model

  • The “micro” in micro-simulation simply means that the data from individual tax returns, not aggregate statistics, are used in the analysis. The basic unit of micro-analysis is a single taxpayer. The “simulation” in micro-simulation means that the analysis essentially imitates (simulates) the filing of tax declarations by taxpayers under alternative tax law scenarios. It replicates the calculations made by each taxpayer to minimize tax liability or maximize after-tax income, consistent with the tax law being simulated. Subsequently, the results for all corporate taxpayers (or select groups of corporate taxpayers) are aggregated to determine the overall (or distributional) effect of a proposed policy choice. The utility of micro-simulation in evaluating alternative tax policy proposals is in its capacity to address simultaneous interactions among alternative tax bases and tax rate parameters.

  • The database used for the micro-simulation analysis of CIT regime in Armenia is based on the entire population of the corporate tax returns filed with the SRC in 2017. In its final format, the data from the SRC is received in Excel format, in which each row is a taxpayer and each column is a variable describing the taxpayer’s tax declaration variables (e.g. income from various sources, expenses, additions and adjustments, allowances, etc.), as well as data identifiers (e.g. for sector and location). For security and confidentially

  • purposes all individual corporate taxpayer identifications were removed; each taxpayer is listed by a generic name: Taxpayer 1, 2, etc.

  • The initial SRC file contained 14,384 taxpayers and 42 variables. From 14,384 database entries, 3,997 were empty, leaving a total of 10,387 non-empty fields. Further, each of the CIT taxpayers can represent more than one sectoral activity. For example, a single taxpayer can have activities in the Construction and Manufacturing sectors at the same time. The database received from the SRC represents a total of 18,142 sectoral activities.

  • The dataset was tested for consistency to identify irregular entries, outliers and errors. The inconsistent entries have been edited, resulting in a new clean dataset.

  • To compute corporate income taxes under alternative tax policy scenarios, the information contained within the dataset is processed consistent with what a taxpayer would do when filing his/her actual corporate tax declaration in real life.

Table 9.

Microsimulation of CIT Returns: Tax Expenditures

article image
Source: IMF Staff, based on data provided by SRC.

50. The main “beneficiaries” of the CIT privileges are easily observable in Table 9.23 These are agriculture, forestry and fishing, mining and quarrying, construction, wholesale and retail trade, as well as the financial and insurance activities, and education. By far, the agricultural sector—including its multi-billion corporations—enjoys the lion share of all preferential treatment in Armenia. Indeed, agriculture attracts 61 percent of total tax-base narrowing tax incentives, an effective rate of 17.9 percent, as well as 73.9 percent of all profit tax exemptions.

51. Based on the microsimulation model, several alternative tax law scenarios were simulated (Table 10). The first “what if” simulation—Scenario 1—preserves all preferential treatment where it has been granted, including base-eroding tax allowances, non-uniform tax rates, and profit tax exemptions. It simulates a single percentage point reduction of the statutory tax rate, applicable to standard taxpayers (those who apply the standard 20 percent prevailing under the current law). The simulation illustrates the cost of a rate reduction by one percentage point is equal to AMD 5.3 billion.

Table 10.

Microsimulation Scenarios

article image
Source: IMF Staff, based on data provided by SRC.