Save

Sectoral Special Taxes in Hungary as Instruments of a Populist Fiscal Policy: A Legal Analysis

In: Review of Central and East European Law
Authors:
Marton Varju Research Professor, Centre for Social Sciences (ELKH), Institute for Legal Studies, Budapest, Hungary, varju.marton@tk.hu

Search for other papers by Marton Varju in
Current site
Google Scholar
PubMed
Close
and
Mónika Papp Research Fellow, Centre for Social Sciences (ELKH), Institute for Legal Studies, Budapest, Hungary
Lecturer in law, ELTE University, Faculty of Law, Budapest, Hungary, mpapp@ajk.elte.hu

Search for other papers by Mónika Papp in
Current site
Google Scholar
PubMed
Close
Open Access

Abstract

In order to meet a variety of locally relevant socio-economic expectations and needs, Hungarian tax policy in the last decade has applied a number of unconventional instruments. The government favored in particular the imposition of additional taxes on corporate taxpayers in specific sectors of the national economy. The taxation of turnover in addition to corporate profits has proved to be a particularly attractive idea. These measures pursue declared objectives which any national government that is faithful to its political mandate could endorse as its own. However, the design of these taxes and their related regulatory hiatuses, which have been subject to extensive scrutiny before the courts of the European Union, raise the possibility of their abusive and discriminatory use to the disadvantage of select individuals, in particular non-national corporate taxpayers. This element of contemporary Hungarian tax policy may well be considered as forming part of the government’s populist policy agenda.

1 Introduction

In the last ten years, many of the interventions by the Hungarian government in the national economy appear to have departed from what may be conceived as the policy mainstream. Markets have been restructured or foreclosed, market positions and opportunities have been fundamentally altered for incumbent and for prospective participants. These changes often had an evidently patriotic – possibly nativist – edge. Whilst pursuing objectives which a populist regime may favor, it is far from certain whether these market interventions can be classified as populist policy instruments per se. In tax policy, Hungary has become a determined supporter (and a beneficiary) of taxation which was introduced in addition to regular corporate income tax in specific sectors of the national economy. These so-called sectoral special taxes, which by now have become a standard feature of the Hungarian tax system, were declared in legislation to pursue seemingly objective fiscal policy aims which pertain to current challenges in public finances and to the rightful exercise of national taxing powers. However, appearances may be misleading in this case. The turnover-based special taxes – as the dominant form of sector-based additional corporate taxation in Hungary – enable taxation which keeps its actual objective concealed and targets certain taxpayers under obscure fiscal premises.

In this article, we examine whether the Hungarian sectoral special taxes – in particular the turnover-based additional corporate taxes – can be classified as populist fiscal policy instruments. In our analysis, which has obvious limitations as it was prepared by legal academics and not by fiscal policy experts, we focus on the design and the corresponding, previously indicated regulatory shortcomings of these taxes as potentially decisive characteristics. Our assessment starts with an overview of the tax measures at issue. Our aim is to identify their main features as instruments of revenue-side fiscal policy in one of the EU Member States. This is then followed by a scrutiny of analytical frameworks which may be potentially useful in assessing the populist nature of these taxes: economic and fiscal policy populism and comparison with the presumed global taxation mainstream. Considering briefly these frameworks enables us to propose a framework of assessment of our own. To complete our analysis, we move on to examining the legal developments before the courts of the European Union concerning the turnover-based special taxes favored by Hungary. The relevant judgments, the reach of which is confined by the limitations of EU law in reviewing and controlling such national tax policy choices, confirm in law the earlier discussed hiatuses in the legal design of the Hungarian taxes. The regulatory shortcomings leave the door open for their arbitrary application by the state to the disadvantage of certain targeted taxpayers, prioritizing the interests of the many over the interests and the rights of the few.

2 Sectoral Special Taxes in Hungary

Faced with a variety of social, economic and fiscal challenges, the Hungarian government has in the past decade introduced a curious mix of corporate taxes in addition to regular corporate income taxation. These measures, which were imposed on corporate taxpayers in specific sectors of the national economy, pursued a particular set of predominantly fiscal objectives. As revenue-raising taxes additional to corporate income taxation, they could legitimately be regarded as instruments which aimed at compensating the effects of cross-border tax competition (i.e., the comparatively low rate and the generous provision of allowances in corporate income taxation in Hungary) on public finances. Some of these taxes were crisis taxes the application of which was limited to the time necessary to recover public finances in a period of crisis. The taxes, which were imposed on corporate turnover, were declared to enable fairer taxation and a more proportionate and equitable distribution of the public burden among taxpayers. Some of the measures pursued alleged regulatory objectives, such as the protection of public health. Overall, the Hungarian sectoral special taxes seem to have been introduced to satisfy the considerable revenue hunger of the government as well as its similarly considerable redistributive desires.

2.1 Special Taxes in the Financial Services Sector

The legislation, which during the past 15 years has imposed and partially repealed a collection of special taxes and levies in the financial services sector, was introduced in 2006 when Hungarian public finances were already in a poor state. Act 2006:lix on improving the balance of public finances pursued the objective of securing extra revenues based on what it called the “solidarity” of those taxpayers that are capable of bearing tax burdens above the general public burden. Originally, it contained only the so-called “interest tax”: the levy imposed on credit institutions on their income from interest. The levy on credit institutions, which was based on the corrected value of the credit stock held, was introduced into the measure by Act 2010:cxxiii.1

A further levy was imposed on financial institutions in 2011 in connection with the financial reimbursement they received from the state as part of the settlement of non-performing foreign-currency consumer loans.2 Act 2010:xc regulated the special tax on financial institutions which was imposed under progressive tax rates. The credit institutions special tax, which was charged at 30% on profits before taxation, was introduced by Act 2010:cxxiii.3 It was deductible from the financial institutions special tax applicable to credit institutions. Further sector-specific burdens include the partially progressive tax on insurance providers,4 the financial transactions (payments) fee,5 the 2014 single-year special levy on credit institutions,6 and the special tax on investment service providers introduced in 2015.7

2.2 The 2008 Special Tax in the Energy and Public Utilities Sector

The additional income tax of energy providers and public utilities was introduced in the first year of the financial and economic crisis.8 It is imposed at 31% on the incomes of hydrocarbon extracting corporations, producers and traders of petroleum products, natural gas and electricity traders, electricity producers, natural gas and electricity universal service providers, natural gas and electricity distributors, public service waste management service providers, and public service water service providers.

2.3 The Turnover-based Crisis Taxes Introduced for 2010–2013

The aim of the turnover-based taxes introduced in Act 2010:xciv – as declared in the measure – was to improve the balance of public finances, which had been undermined by the financial and economic crisis, and by the policies of previous governments. The taxes were claimed enable achieving this objective by imposing additional corporate taxation under progressive rates9 in commercial retail, telecommunications and the commercial activities of energy providers. The choice of the annual net turnover of taxpayers as the tax base was supposed to enable a differentiated treatment of tax subjects on the basis of their size and market position, in harmony with the general objective of the measure. The taxes were regulated as extraordinary crisis taxes which applied only in the 2010, 2011 and 2012 tax years.10

2.4 Other Special Taxes

Following the declared fiscal objective of ensuring the proportionate sharing of the public burden among taxpayers, Act 2012:clxviii imposed an extra tax on public utility service-cables and service-pipelines.11 The special tax imposed on telecommunications service providers was introduced by Act 2012:lvi with the declared purpose of securing revenues for “common social expenditures”. The tax is imposed on the duration of calls made, or on the number of messages sent by service users.

2.5 Turnover-based Special Taxes Introduced After 2014

After the (self-)repeal of the 2010 crisis tax package in 2013, the Hungarian government decided to introduce a new set of additional taxes based on corporate turnover. In the case of the advertising sector special tax regulated by Act 2014:xxii, the justification for the measure was found in the constitutional principle of ensuring the proportionate sharing of the public burden (taxation according to the financial ability (“ability to pay”) of taxpayers). The public health levy of the tobacco sector pursued the – in part regulatory – objective of securing revenues for the improvement of public healthcare services and of making the tobacco industry pay for the public health expenditure associated with the health risks of their products. As a further objective, Act 2014:cxiv referred to the proportionate sharing of the public burden among taxpayers.

The multiple times modified advertising sector special tax is imposed on the net turnover achieved from the analogue and the digital publication of advertisements for a consideration. The commissioning of advertising services is also taxed. The application of the tax as well as the relevant administrative provisions was suspended a number of times, most probably having regard to the ongoing investigation of the measure under EU state aid law and the ensuing judicial review procedure before EU courts.12 The tobacco sector public health levy, which was declared by Act 2014:cxiv to finance exclusively the public health expenditures of the budget, was imposed in three progressive bands on the annual net turnover of tobacco manufacturers and traders.13 Following the European Commission’s decision in 2016, which declared it incompatible with EU state aid law, the government abolished the levy and ordered the repayment of previously paid amounts.14

The food-chain inspection fee as introduced by Act 2011:clxvi and amended by Act 2014:lxxiv modifying the original 2008 legislation15 is a further turnover-based special levy in the Hungarian tax system. In principle, the levy is charged on the annual net turnover of shops trading with daily consumables for the purpose of financing the administration of food-chain inspections. The controversial 2014 modification of the fee introduced eight steeply progressive bands. These bands were abolished in the next year by Act 2015:clxxxii which reintroduced the single rate applicable before 2014.

2.6 The Special Taxes Introduced in 2020

In 2020, Hungary introduced a new set of special taxes. They were first regulated under covid-19 emergency powers with the aim of securing revenues for the public “Pandemic Fund”, which had been set up to finance the so-called “Action-plan for the protection of the national economy”.16 The credit institutions pandemic special tax, which was imposed on the balance sheet total of the second tax year preceding 2020, in practice meant the increasing for the 2020 tax year the highest tax band of the financial institutions special tax. According to Act 2020:xlvi, which replaced the original emergency regulation and implemented its provisions in the previously mentioned Act 2006:lix on the financial services sector special taxes, the pandemic special tax is an amendment introduced “with the purpose of mitigating the economic impacts of the corona virus pandemic”.

The re-introduced commercial retail tax covers both offline and online retail, the latter including the retail activities in Hungary of foreign-established providers. It is imposed in four progressive bands17 on the net turnover of the taxpayer engaged in commercial retail activities, plus the turnover achieved by the producer or the seller of the product from its sale by the taxpayer, and the rebates offered by the producer or the seller for the retailer. Act 2020:xlv, which now regulates the tax, does not mention the pandemic crisis among its objectives. Instead, it refers to the reinforcing of “consumption-turnover” taxes within the Hungarian tax system, the decreasing of income taxes and the public burdens on employment, and ensuring a more proportionate sharing of the public burden among taxpayers. As a further objective, the tax is imposed “having regard to” the negative environmental impact of commercial retail activities.

3 What Makes a Tax a Populist Policy Instrument?

The tax policy behind the Hungarian sectoral special taxes, which using diverse means seems to pursue the rather orthodox aim of maximizing public revenues, does not at first sight appear as radically unconventional. As introduced earlier, the Hungarian measures aim at raising additional revenues as well as making the national tax base more diverse and possibly more resilient in a volatile financial and fiscal environment. The policy pursued by Hungary focuses on consolidating public finances and securing financing for common expenditures in periods of crisis or in less extreme times. Furthermore, the turnover-based taxes introduced enable taxation which is more proportionate to the size and the market position of taxpayers. Together with the other sectoral special taxes, turnover taxes place additional taxation on taxpayers that – assumedly – have an ability to pay more into the public purse than other tax subjects. Nor does this tax policy emerge after a quick assessment of its objectives – despite the evident element of targeted surplus taxation in specific economic sectors affecting specific economic operators – as irresponsibly and unsustainably populist. In fact, it is rather unclear using which analytical framework could the Hungarian taxes be categorized as populist. This is what we examine next.

3.1 Economic and Fiscal Populism

Analyses of economic and fiscal populism provide perhaps the most evident framework for assessing the nature of Hungarian tax policy as manifested in the introduction of additional corporate taxes. Traditionally, populism in the domain of economic and fiscal policy has been associated with politics and policies which aim to address socio-economic underdevelopment, and to manage social conflicts which arise from (class) inequalities and unequal socio-economic opportunities.18 Such policies have been observed to rely on policy instruments which on the expenditure side of public finances pursue redistributive objectives, or to use instruments which aim at securing economic growth and development.19 More current, morally-charged generalist definitions of populism, which equate the notion with some kind of politically expressed moral (or other normative) cause for prioritizing the interests and needs of the “people” over those of the “elite”,20 mention other economic and fiscal policy tools. These include, for example, public expenditure on job creation and/or employment security, or active policies – which may require considerable public spending – on developing and improving public services available for the “people”.21 With their emphasis on the expenditure side of public finances, these definitional frameworks are not particularly useful for assessing and categorizing revenue-side fiscal instruments, such as the Hungarian special taxes. Although intensive redistributive policies normally assume the availability of public revenues for their financing, this circumstance alone cannot be used to classify revenue maximizing taxes as genuinely populist fiscal instruments.

The revenue-side of public finances has attracted its own conceptualizations of populism. Populist fiscal measures (taxes or tax rates) have been defined by contrasting them with mainstream, “orthodox” measures, and they are considered as referring to instruments – at least in the economics literature – which sustain “suboptimal and/or unsustainable fiscal policies”.22 This definition assumes some kind of a qualitative economic and/or fiscal policy assessment of national tax measures,23 which makes the conclusion as to the nature of the measure in question somewhat uncertain and possibly contestable.24 The different substantive policy perspectives taken in the course of the assessment also have an impact on the ultimate classification of national taxes.25 These complex conceptual-analytical frameworks focus on the short- or long-term (policy, economic, social etc.) failures or “bottlenecks”26 of government policy,27 or on the hiatuses and errors in policy detail or in strategy.28 Our admittedly limited, legally-focused analysis cannot undertake such assessments. Moreover, the apparent disinterest of these analyses in the issue of whether the measure itself – having regard to its objectives and design – can be classified as a populist instrument makes them of limited use for the purposes of this paper. For us, the most relevant lesson from the revenue-side populism discourse is that populist taxes may in principle be identified by contrasting them with orthodox, mainstream taxation solutions.

3.2 The Mainstream in Global and European Corporate Taxation

The global and the European mainstream in corporate taxation – provided that it can be identified – may serve as a potentially useful reference framework for the classification of the Hungarian special taxes as populist fiscal policy instruments. In (very) general terms, Western (core oecd country) taxation practices represent the policy mainstream.29 They have served as the globally dominant model – disseminated actively with the help of influential international organizations such as the oecd – towards which local tax policies have converged.30 The central priority is the increasing of government revenue, which is enabled most effectively by taxes that have a broad base (e.g., vat, personal and corporate income tax).31 Good tax policy entails taxation which is effective (is able to maximize revenues), simple, and is neutral towards taxpayers;32 distributive/redistributive issues – in particular, the fairness and equity of taxation33 – are for the expenditure side of public finances to address.34 Western tax policies – as they currently stand and are reflected in existing or developing international taxation standards – continue to provide the benchmark for assessing national practices, structures, or instruments of taxation.35

However, despite their significant influence, the recognition of Western taxation practices as the indisputable mainstream, to which other taxation solutions can be measured, is a contestable position. It is well-documented that the proliferation of the Western model has not led to uniform national tax policies across the globe.36 There is evidence of considerable local variation which is believed to have emerged from local economic, social and political circumstances and conditions.37 Furthermore, the problems of Western tax policy – in particular, undertaxation in corporate income taxation and the erosion of the national tax base – have led to individual states implementing alternative (unorthodox) tax solutions which promise available and sustainable revenues.38 Some states favored the introduction – in addition to regular corporate income taxation – of tax burdens, which show the characteristics of excise levies, such as “Robin Hood” taxes, or taxes that aim at equalizing or ensuring the fairness the tax burden borne by individual taxpayers.39 The undertaxation of internationally mobile taxpayers (multinationals) and the resulting unequal taxation of national and non-national corporate taxpayers emerged as a particularly pressing problem in the digital economy, which an increasing number of states addressed by implementing so-called digital equalization taxes.40 It is worth noting that the general objective of these unilateral national policies was revenue maximization, which we identified earlier as a mainstream taxation objective.41

Dealing with the challenges of corporate direct taxation in a globalized economic environment has not remained the prerogative of particularist national tax policies. Within the oecd, a consensus-based multilateral framework has been developed and adopted which aims at addressing and remedying national tax base erosion and the corresponding practices in international taxation. The oecd/G20 Base Erosion and Profit Shifting (beps) Framework42 was put in place pursuing a policy narrative of ensuring global taxation fairness and equal fiscal opportunities across the globe.43 The beps covers multiple action areas concerning corporate income taxation. As a significant beps tool, the beps Multilateral Convention/Multilateral Instrument,44 within its confined scope, targets the non-taxation or undertaxation of multinationals which have access to aggressive cross-border tax planning solutions enabling the shifting of corporate profits across tax jurisdictions. Even though the beps project explicitly recognized the local public and private harms of corporate tax avoidance,45 which may affect national economies differently or appear in different ways in different economies,46 it insisted on sustaining – with some reforms – the existing international tax framework.47 Neither unilateral solutions by states following their own needs, nor solutions outside of the international framework were supported ultimately.48 The beps – as assessed (critically) – further reinforces the revenue-focused Western approach and avoids a fundamental reform of the international tax system with a view to addressing the inequalities in global (tax) competition.49

With the problems of undertaxation and the erosion of the national tax base espoused by Western/oecd tax policy, the assessment whether a national tax policy belongs to the policy mainstream became a question of determining whether the tax measure is covered by the consensus-based global tax framework or not. In the beps process, the introduction of turnover-based taxes – as additional taxation to regular corporate income tax – to compensate cross-border undertaxation in the digital economy and to equalize the tax burden of different taxpayers was considered for a while as a solution which can be accommodated within the policy mainstream.50 While treating it outside of the consensus-based multilateral framework, the 2015 beps digital economy taxation report51 recognized the possible necessity of an extra digital equalization tax.52 It argued that the beps’ central reform, the introduction of the “significant economic presence” principle may not be able to secure in the digital economy the desired fair and equal attribution of profits among national tax jurisdictions. It admitted that the introduction of an equalization levy – as shown by the examples of similar national taxes – could ensure the equal tax treatment of foreign and national taxpayers and could enable states to collect equal revenues from both groups of digital taxpayers.53 However, the oecd – lacking global consensus on the issue – did not act upon the report’s recommendation,54 and left the introduction of an additional equalization tax to individual states acting unilaterally55 in national taxation competences.56

In the European Union, similar dynamics characterized common policy developments. On the one hand, the EU is a supporter of the consensus-based multilateral solutions of the oecd/G20 and remains dedicated to their implementation in Europe.57 It adopted the common corporate tax base (cctb) proposal,58 the common consolidated corporate tax base (ccctb) proposal,59 and the internal market tax avoidance practices directive.60 However, taxation problems in the digital economy have attracted particular, Member State-driven responses which left the EU with the task of coordinating and possibly consolidating these developments. In the 2017 Joint initiative political statement,61 a group of EU Member States called for the introduction of an “equalisation tax” on the turnover generated in Europe by digital companies established anywhere in the world. The proposed aim of this tax was to compensate the corporate income taxes unpaid by these companies after their activities in the Member States. The European Commission reacted to the joint initiative by explicitly recognizing in its policy proposal that Member States62 should be enabled to adopt measures in national competences for protecting the national tax base in the digital economy.63 The introduction of a national equalization tax, which would bring under the national tax jurisdiction “all untaxed or insufficiently taxed income generated from all internet-based business activities”,64 was recognized as one possible solution. The Commission’s legislative plans concerning fair taxation in the digital economy65 propose the taxation of “revenues” generated by digital services in a Member State irrespective of the place of establishment of their providers. However, despite the proposals the inter-state consensus required for the adoption of the necessary EU measures is still missing.66

Having regard to this earlier assessment, the circumstance that a corporate taxation solution is based on a multilateral/inter-Member State consensus does not seem decisive enough for characterizing it as a mainstream, or in the contrary case, an unorthodox, populist instrument. Political support across states for a multilateral tax response proposed with the purpose of addressing corporate undertaxation and the erosion of the national tax base is driven by complex considerations, such as ensuring fairness in international tax competition, or developing a long-term, sustainable international taxation framework. Therefore, falling outside the prevailing international consensus does not necessarily indicate that the main issue with the tax solution is its supposed unorthodoxy. Furthermore, additional special taxes, which expand or consolidate the national tax base or seek revenues from taxpayers that can be taxed more, pursue an objective – that of revenue maximization – which is definitely a mainstream objective. These assessments are unaffected by how unconventional these measures may seem when contrasted with regular corporate income taxation. On this basis, we need to seek the framework of our analysis elsewhere. We propose that the design of the tax (its base, its rates and its mode of imposition) as well as the consistency of the tax design with the declared (and assumed) taxation objectives need to be examined for reaching a more compelling conclusion regarding its supposed populist nature.

3.3 The Regulation of the Hungarian Special Taxes: Faulty by Design?

As noted earlier, the sectoral special taxes introduced by Hungary pursue – as set forth in the relevant legislation – fairly unconventional taxation objectives.67 They all aim to secure extra revenues for the budget (i.e., maximize revenues) either in times of a crisis or in more normal times. Being sector-oriented, they are set to achieve this objective by imposing taxes in sectors of the national economy where further taxable revenues are suspected. The turnover-based special taxes – additionally – pursue the declared aim of making taxation fairer by distributing the tax burden among taxpayers more proportionately. In general, the special taxes aim to increase the internal diversity and (thus) the resilience of the national tax base.

However, the design of these taxes – in particular that of turnover taxes – does not support the prevalence and the genuineness of these objectives in an entirely convincing manner. Turnover-based taxes tax fundamentally corporate size and the market position of the corporate taxpayer, and not – as claimed by the legislature – a financial source available with the taxpayer for further taxation. The turnover achieved by corporate taxpayers – on its own – does not indicate their actual capacity for (additional) taxation, which latter also depends on the costs they incur in the course of their operation. Therefore, it is not guaranteed that the tax burden imposed will correspond to the actual financial capacity of the targeted taxpayer, which also has the significant consequence that the measure cannot be assumed to serve its declared objective of taxing taxpayers more proportionately to their “ability to pay”. Furthermore, when taxpayers with different actual abilities to pay (additional) taxes are taxed on the basis of their turnover, taxation will be discriminatory, possibly arbitrarily imposed on the targeted taxpayers. In effect, these corporate taxpayers are compelled to bear an additional share of the public burden without tax regulation ensuring that they objectively – and not only assumedly – have the necessary financial capacity.68 There are thus problems with the clarity and the predictability of turnover taxation,69 which also raise doubts about whether the declared objectives of these measures are the objectives actually pursued by them.

As a further issue affecting the credibility of the Hungarian taxes, addressing the undertaxation of internationally mobile corporate taxpayers (multinationals) in the national tax jurisdiction, which seems like an evident aim for additional corporate taxation, was not recognized explicitly as their objective. Nor was it considered and mentioned by the national legislature whether the extra taxation imposed aimed at compensating the effects of international tax competition on regular corporate income taxation in Hungary and on the revenues generated by it, as manifested in particular in the comparatively low rate set for taxation or in the generous tax allowances offered. The silence of Hungarian policy on this matter is disconcerting as it would be difficult to deny that – especially in the case of the additional taxation of turnover – the aim of the surplus corporate taxation is to compensate for the taxes lost under corporate income tax. Turnover-based additional taxes are also suitable instruments for equalizing the tax burden between taxpayers which can and which cannot escape the national tax jurisdiction (i.e., internationally mobile taxpayers and their less mobile national competitors) as such taxes are regarded as capable of securing the collection of revenues due in the national tax jurisdiction.70 They have been characterized as enabling the assertion of national taxing powers – with the help of a broad nexus rule – over internationally mobile, non-resident taxpayers, and improving the neutrality of taxation “by restoring the level playing field” between the earlier mentioned two cohorts of taxpayers.71 The concealment of these potential objectives of taxation is in itself a serious regulatory hiatus. The possible rationale of this approach – that otherwise Hungary would have had to establish accurately, and not simply assume, that there are taxpayers in the national tax system which are undertaxed and that Hungary was unwilling to provide that clarification as it would have reduced its discretion in taxing whoever it may want to tax and how – is even more problematic.

If designed and regulated differently, the Hungarian taxes could in principle be assessed as pursuing objective aims pertaining to the more general fiscal objectives of halting the erosion of the national tax base and ensuring a fairer distribution of the public burden. However, there is evidence that the turnover-based special taxes operate instead – as permitted by their design – as so-called “pure-revenue” taxes72 aiming simply to raise extra revenues from deliberately targeted corporate taxpayers. The declared objectives of the Hungarian measures cannot be used to justify such taxation. In the absence of other officially declared objectives, such as the compensation of undertaxation (equalization), the Hungarian additional taxes therefore appear as arbitrarily imposed in addition the corporate income tax burdens the targeted taxpayers already have to bear. Such taxation may in fact entail the violation of the taxation principle raised in support of these measures – the principle of the proportionate sharing of the public burden among taxpayers – as the targeted taxpayers are taxed according to their (wrongly) assumed taxability and not according to their actual and accurately established financial capacity for taxation. Undeniably, the sectoral special taxes secure funding for expenditures in the interest of the public, but this comes at the cost of collecting revenues from certain targeted taxpayers in an unlawfully discriminatory manner. As it turns out, the taxpayers that felt discriminated against as a result of considerable extra taxation were nationals of other Member States competing in the Hungarian market. The disadvantages allegedly suffered by them opened a way for legal challenges against the Hungarian taxes under EU law.

4 The Hungarian Special Taxes Before the Courts of the European Union

The Hungarian turnover-based special taxes have been subject to scrutiny before the courts of the European Union under rules of EU law which were introduced to prohibit discrimination in a national economy against non-national economic operators and other individuals (the EU fundamental economic freedoms) and to prohibit the distortion of competition by the state to the benefit of select economic operators and to the disadvantage of their – often non-national – competitors (EU state aid law). The judgments delivered in these cases were unable to establish the violation of any of these legal prohibitions by the Hungarian measures. However, the gaps and mistakes that can be found in the courts’ reasoning revealed – in our assessment – that the additional taxation of corporate turnover in Hungary is fundamentally inconsistent with its objectives declared in legislation. Focusing on the deference owed under EU law to Member State fiscal policy choices, the rulings allowed the national legislature to rely on assumptions – instead of precise and objectively determined evidence on the actual taxable capacity of taxpayers – when regulating taxation obligations, which meant that Hungary could continue to conceal its aims with the targeted additional taxation of larger size corporate taxpayers.

As well-known, the crisis taxes introduced in 2010–13 were challenged before the EU Court of Justice for violating the internal market fundamental economic freedoms. The central claim was that additional taxation adjusted to turnover had led to discrimination on the basis of nationality, as non-national corporate taxpayers, which tended to be larger than their national competitors, were handed the higher (extra) tax bills. As a forerunner to these cases, the judgment in Hervis made it clear that the Court was prepared to establish discrimination against non-nationals when that claim is supported by clear and unambiguous evidence. In Hervis, the so-called “aggregation rule” regulated in the 2010 commercial retail tax, which determined the taxable amount as the consolidated turnover of the company group of which the taxpayer was a member, provided such evidence of discrimination and thus of the violation of the freedom of establishment.73 The compatibility of the tax itself with EU law was examined in Tesco-Global, which was decided by the Court together with the challenge mounted against the 2010 telecommunications special tax in Vodafone Magyarország.74

In both cases, the claim of direct discrimination was dismissed lacking evidence to that effect.75 Indirect discrimination, which was claimed on account of the higher taxation of non-native taxpayers, was rejected by the Court following a rather curious line of reasoning. It argued that the higher tax burden of these taxpayers was only a random development – and not a sufficiently certain consequence of the regulation of the tax – which emerged from the specific structure of the relevant Hungarian market at the given time, where the largest undertakings happened to be non-nationals.76 Furthermore, the evidence presented to establish the correlation between high turnover and the resulting high taxation, and the nationality of the taxpayer was rejected as unconvincing.77 The judgments felt prepared to rule that the progressive taxation of corporate turnover pursued the aim of taxing taxpayers in a “neutral” manner according to their financial and taxable capacity, which is a choice that the Member States may legitimately make within their fiscal policy autonomy.78

The rulings left the analysis of the Hungarian taxes incomplete. The scrutiny of indirect discrimination did not touch upon the issue of market access, in particular the impact of the tax on the access of new participants from another Member State to the Hungarian market as compared to the position of their incumbent national competitors. As raised by commentators, entering into another national market and securing a position there usually entails high investment and operational costs. This has the evident consequence that the turnover achieved by incoming operators may not reflect the same ability to pay additional taxes as the (similar) turnover of domestic undertakings.79 This mismatch between the turnover of incoming corporate taxpayers and their actual taxability under a tax which is imposed in addition to regular corporate income taxation raises the further issue of whether the declared objective of the Hungarian taxes in fact corresponds with the objective actually pursued by them. This question was not investigated by the Court, which was happy to accept the aims that were stated in the underlying national legislation as genuine and valid.80 With this omission, the judgments arguably permitted that the untransparent, possibly arbitrary choices of the Hungarian government would go unchecked in law.

As mentioned earlier, the judgments observed that the taxation of corporate turnover enables the neutral taxation of taxpayers which matches their financial capacity for additional taxation. However, the Court refused to consider in detail whether this is the actual case and whether this is the case for every taxpayer affected. Its reasoning recognized the fiscal objective declared in the Hungarian legislation, and went on to declare on that basis – as had been suggested by the Advocate General in her opinions81 – that there was a reasonable (albeit neither complete, nor fully accurate) and therefore legally sufficient connection between the turnover achieved by the taxpayer and its taxable capacity. Although the review of national tax measures under the EU’s fundamental economic freedoms is necessarily confined by the (direct) tax policy autonomy enjoyed by the Member States, such use of generalized assumptions concerning taxability in cases dealing with claims of discriminatory higher taxation is – in our assessment – an erroneous judicial approach. The Court simply ignored the possibility that the turnover achieved by the taxpayer may not provide an indication of its actual “ability to pay”. This hiatus in the Court’s decision reveals a more fundamental problem with the Hungarian taxes. When taxable capacity is only assumed but not established objectively, taxation – even if it is additional taxation – cannot claim to pursue the objective of ensuring a more proportionate sharing of the public burden among taxpayers.82 Rather, it burdens taxpayers as a potentially punitive, excise-type levy which is imposed arbitrarily on targeted taxpayers.

The 2014 advertising sector special tax was challenged under EU state aid rules.83 In its negative decision, the Commission found that the lower rate of taxation applied to low-turnover (smaller) undertakings under the tax’s progressive rates provided a selective economic advantage disadvantaging in competition their high-turnover (larger) competitors.84 The General Court disagreed with the Commission. Its judgment, which under the selectivity requirement of state aid law failed to consider in true detail the consistency of the design of the tax with the declared objectives of taxation, held that the Commission had erred when it concluded that the Hungarian tax pursued exclusively the aim of securing the collection of additional revenues for the state budget. In the General Court’s assessment, the objective suggested by the Commission was overly general and did not correspond with the Hungarian tax’s fiscal objective, which – as declared by the Hungarian legislature – was to redistribute the tax burden among taxpayers.85 The judgment also accepted that the design of the tax was consistent with this objective. In this regard, it argued that since turnover represents the capacity of taxpayers for taxation, the progressive taxation of turnover can be used to impose a higher tax burden on taxpayers with a larger financial capacity.86 The General Court held that such a form of differentiated taxation is based on objective differences among taxpayers, and it is thus neither selective (discriminatory), nor arbitrary.87 It concluded that differentiation among taxpayers based on size (turnover) corresponds with the objective of fairer taxation.88

In appeal, the Court of Justice upheld the first instance ruling. First, its judgment recognized the autonomy of the Member States in developing their national tax system as well as their freedom to introduce surplus corporate taxes which “take account of the ability to pay of taxable persons”.89 Then, it confirmed that the discriminatory potential (selectivity) of such additional taxation needs to be assessed against tax itself, and not with reference to the regular system of corporate income taxation which it was introduced to supplement.90 With deference to national tax policy choices thus established, the Court established that turnover is a relevant indicator of taxability, that it differentiates between taxpayers in a neutral way, and that additional corporate taxation based on turnover is a choice which the Member States may legitimately make.91 This part of the judgment also suggested that the tax burden does not have to be precisely adjusted to the actual financial capacity of taxpayers in order to pass the legal hurdle of EU state aid law.92 The Court closed its reasoning by pointing out that there was no evidence presented in the case that would have suggested that the Hungarian tax had been regulated – with the intention of circumventing EU obligations – on the basis of manifestly discriminatory parameters.93

Both rulings endorsed unquestioningly the objective declared in national legislation for the Hungarian tax, and relied on that objective directly in their reasoning, without subjecting it to any form of legal scrutiny, even though EU state aid law provides for EU courts the necessary legal means.94 As a further problem, the judgments found no issue with the fact that taxation was based on an assumption – and not on objective evidence that would cover the taxation situation of every relevant taxpayer – that turnover indicates the capacity95 of taxpayers for (additional) taxation.96 As raised already in connection with the 2010–13 special taxes, there are identifiable instances – which are relevant for the application of EU internal market law and its core prohibition of discrimination – when turnover does not express taxable capacity, and taxes imposed on similar turnovers may entail – without objective justification – very different actual tax burdens for the affected taxpayers.97 Such tax design enables discriminatory tax treatment and the selective provision of advantages in competition having regard to the size and the market position of corporate taxpayers,98 which should have been accepted by the Court as evidence of manifest discrimination violating EU law. The gap between turnover and taxability also means that the Hungarian tax cannot be claimed to serve the objective of distributing the tax burden more proportionately among taxpayers with different abilities to pay (extra) taxes. There are much better fitting objectives for additional turnover taxation, such as equalizing the tax burden of undertaxed internationally mobile taxpayers. However, since the Hungarian measures do not refer to such an objective, the tax burden they impose is left unexplained and unjustified.

5 Conclusions

In this article, we sought answer to the question of whether the additional sector-specific corporate taxes imposed by Hungary over the last decade and a half can be classified as instruments of a populist fiscal policy. For this purpose, we examined their objectives as declared in national legislation as well as their design as seemingly unconventional revenue-side fiscal instruments. Our analysis found that on the revenue-side of public finances it is fairly difficult to classify a measure as populist. Corporate taxes – both regular and extra – tend to pursue the objective of revenue maximization which is a traditional, mainstream taxation objective. The protection of the national tax base is another key objective pursued which now forms part of the tax policy mainstream worldwide and in Europe. Populism is more likely to be found on the expenditure-side of the budget. Nevertheless, we proposed and established, partly on the basis of the developments before the courts of the EU concerning Hungarian turnover-based extra taxes that the regulation and the design of the Hungarian measures raise issues which may be assessed as lending them a populist character. The Hungarian turnover taxes keep their genuine taxation objectives concealed, and their imposition, which is based on a broad legislative assumption of additional taxability, contains elements of arbitrariness. The extra tax burden they impose is not guaranteed in law to match the actual financial capacity of the targeted larger sized, predominantly non-national taxpayers, which enables taxation that aims to serve the interest of the many in the local population at the expense of a few unfortunate, non-native corporate tax subjects.

Acknowledgements

This research has received funding from the National Research, Development and Innovation Fund of Hungary (project no. K129245) as well as the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 822590. Any dissemination of results here presented reflects only the authors’ view. The Agency is not responsible for any use that may be made of the information it contains.

1

Repealed from 2017 by Act 2016:lxvi.

2

Act 2011:lxxv. Foreign-currency consumer loans constituted a critical social pandemic during and after the global financial and economic crisis affecting tens of thousands of households that faced unsustainable private financial burdens when the Hungarian currency collapsed.

3

It was repealed from 2019 by Act 2018:xli.

4

Act 2012:cii on insurance tax. Its objectives were the simplification of taxes on insurance and the securing of revenues for “common social expenditures”.

5

Act 2012:cxiv.

6

Act 2013:cc.

7

Act 2014:lxxiv.

8

Act 2008:lxvii. The separate tax on energy products was introduced under EU harmonization obligations based on Directive 2003/96/ec (oj L83/51). The original implementing act (2003:lxxxviii) was repealed in 2016, and the energy tax now forms part of the general excise duties framework (Act 2016:lxviii on excise duty).

9

For commercial retail, the tax rate was determined in three progressive bands with reference to the annual net turnover of the taxpayer: 0% on the proportion of the taxable amount under 500 million huf, 0.1% on the proportion between 500 million and 30 billion huf, 0.4% on the proportion between 30 and 100 billion huf, and 2.5% on the proportion that exceeds 100 billion huf. For telecommunications, the tax rate was 0% on the proportion of the taxable amount under 100 million huf, 2.5% on the proportion between 100 million and 5 billion huf, and 6.5% on the proportion that exceeds 5 billion huf. For energy providers, the tax rate was 1.05% of the annual net turnover.

10

The act regulated its own repeal after three years.

11

The tax covered water-supply, sewage and drainage pipelines, natural gas and district-heating supply pipelines, electricity and telecommunications cables. In order to support the development of telecommunications infrastructure, the owners of telecommunications cables were provided tax reductions in four progressive bands. A further tax reduction was offered for the development of high-speed telecommunications cable networks.

12

See Act 2017:xlvii and Act 2019:lxxiii.

13

Act 2015:lxxxi extended the application of the levy to the tax years after 2014.

14

Act 2016:cxxxv. Commission Decision (EU) 2016/1846 of 4 July 2016 on the measure sa.41187 (2015/C) (ex 2015/nn) implemented by Hungary on the health contribution of tobacco industry businesses, oj L 282/43.

15

Act 2008:xlvi.

16

Government regulation 108/2020 (iv. 14.) on the pandemic special tax of credit institutions and Government regulation 109/2020 (iv. 14.) on the commercial retail tax. As another measure, the government acting under emergency powers diverted the remaining part of the motor vehicle tax (40%) from local authorities to the central budget (Government regulation 92/2020 (iv. 6.).

17

They are identical to the tax bands applied for the 2010 commercial retail special tax.

18

See Paul W. Drake, “Conclusion: Requiem for Populism?,” in Michael L. Conniff (ed.), Latin American Populism in Comparative Perspective (University of New Mexico Press, Albuquerque, 1982), 217–245, 218; Michael L. Conniff, “Introduction: Toward a Comparative Definition of Populism,” Michael L. Conniff (ed.), Latin American Populism in Comparative Perspective (University of New Mexico Press, Albuquerque, US, 1982), 3–29, 5.

19

For a critical account of economic populism, which focuses on its (inevitable) failures, see Rudiger Dornbusch and Sebastian Edwards, “Macroeconomic Populism,” 32(2) Journal of Development Economics (1990), 247–277.

20

See, for instance, Jan-Werner Müller, “Was ist Populismus?,” 7(2) Zeitschrift für Politische Theorie (2016), 187–201, at 187; Cas Mudde, “The Populist Zeitgeist,” 39(4) Government and Opposition (2004), 541–563, at 542.

21

On the complex, cyclical nature of the process, where these are only possible outputs depending on the actual state of power relations between the political and the economic elite and the “people”, see Timothy Besley and Torsten Persson, “The Rise of Identity Politics: Policy, Political Organization, and Nationalist Dynamics,” Working Paper, London School of Economics (April 2021), available at https://www.lse.ac.uk/economics/Assets/Documents/personal-pages/tim-besley/working-papers/the-rise-of-identity-politics.pdf.

22

Jess Benhabib and Andrés Velasco, “On the Economics of Fiscal Populism in an Open Economy,” Federal Reserve Bank of Minneapolis Discussion Paper 97 (1995), 2.

23

For instance, policy failure in national fiscal and economic policy, as manifested, for example, in the so-called Latin American “populist policy cycle”, may serve as an indication of a populist as opposed to a mainstream fiscal policy, Jeffrey D. Sachs, “Social Conflict and Populist Policies in Latin America,” National Bureau of Economic Research Working Paper No. 2897 (1989), 5.

24

Benhabib and Velasco, op.cit. note 22, 2.

25

For instance, from a redistribution perspective a tax, which is condemned by others as a failed and unorthodox, can be regarded as a successful measure, because it may enable the financing of public services to the benefit of society. See also Sachs, op.cit. note 23, 6.

26

Usually, measured against the often inevitable short-term benefits and improvements.

27

See Rudiger Dornbusch and Sebastian Edwards, “The Macroeconomics of Populism,” in Rudiger Dornbusch and Sebastian Edwards (eds.), The Macroeconomics of Populism in Latin America (University of Chicago Press, Chicago, US, 1991), 7 –13, 9 and 13. See also the overview in Petar Stankov, The Political Economy of Populism (Routledge, Abingdon, 2021), at 62–63.

28

Sachs, op.cit. note 23, 6.

29

See, for instance, the oecd Model Tax Convention on Income and on Capital, http://www.oecd.org/ctp/model-tax-convention-on-income-and-on-capital-full-version-9a5b369e-en.htm.

30

For the original assessment of the state of play, see Nicholas Kaldor, “Will Underdeveloped Countries Learn to Tax?,” 41 Foreign Affairs (1963), 410–419. For a recent assessment of developments and the contemporary situation, see Philipp Genschel and Laura Seelkopf, “Did They Learn to Tax? Taxation Trends Outside the oecd,” 23(2) Review of International Political Economy (2016), 316–344.

31

Ibid., 318 (and the literature cited). The abolition of taxes on trade and their replacement with these forms of taxation were another crucial element of the Western model.

32

See, for instance, the so-called Ottawa Taxation Framework, http://www.oecd.org/ctp/consumption/1923256.pdf.

33

See, in this regard, Richard Abel Musgrave, “Horizontal Equity, Once More,” 43(2) National Tax Journal (1990), 113–122, and “Progressive Taxation, Equity, and Tax Design,” in Joel Slemrod (ed.), Tax Progressivity and Income Inequality (Cambridge University Press, Cambridge, UK, 1996), 341–356.

34

Genschel and Seelkopf, op.cit. note 30, 317.

35

Ibid., and Laura Seelkopf, Hanna Lierse and Carina Schmitt, “Trade Liberalization and the Global Expansion of Modern Taxes,” 23(3) Review of International Political Economy (2016), 208–231.

36

Genschel and Seelkopf, op.cit. note 30, 340.

37

For instance, cross-border (capital) tax competition has led to developing and transition economies changing their corporate income tax regimes. The flat rate taxation thus introduced is supposed to preserve their competitiveness as regards other national economies. See Duane Swank, “The New Political Economy of Taxation in the Developing World,” 23(3) Review of International Political Economy (2016), 185–207.

38

See, in this regard, the critical analysis of post-crisis global tax policy responses in Allison Christians, “Taxation in a Time of Crisis: Policy Leadership from the oecd to the G20,” 5(1) Northwestern Journal of Law and Social Policy (2010), 19–40.

39

See Christoph Jescheck, “The Substantive Scope of Tax Treaties in a Post-beps World: Article 2 oecd mc (Taxes Covered) and the Rise of New Taxes,” 45(5) Intertax (2017), 382–390.

40

See Georg Kofler and Julia Sinnig, “Equalization Taxes and the EU’s Digital Services Tax,” in Werner Haslehner, Georg Kofler, Katerina Pantazatou and Alexander Rust (eds.), Tax and the Digital Economy: Challenges and Proposals for Reform (Wolters Kluwer, Alphen aan den Rijn, Netherlands, 2019), 101–146, at 112–113. As a further particular taxation development focusing on the equitable distribution of the public burden, progressive taxation has re-emerged. See Julian Limberg, “What’s Fair? Preferences for Tax Progressivity in the Wake of the Financial Crisis,” 40(2) Journal of Public Policy (2018), 171–193, “Tax the Rich? The Financial Crisis, Fiscal Fairness, and Progressive Income Taxation,” 11(3) European Political Science Review (2019), 319–336, and “Banking Crises and the Modern Tax State,” Socio-Economic Review (2020), mwz055, and Laura Seelkopf and Hanna Lierse, “Democracy and the Global Spread of Progressive Taxes,” 20(2) Global Social Policy (2020), 165–191. Essentially, the aim is to implement some kind of “horizontal equity” in taxation. On the latter concept and its difficult reception in the taxation community, see Brian Galle, “Tax Fairness,” 65(4) Washington & Lee Law Review (2008), 1323–1379.

41

On revenue maximization as a state objective, see Margaret Levi, Of Rule and Revenue (University of California Press, Berkeley, US, 1988).

42

oecd/G20 Inclusive Framework on Base Erosion and Profit Shifting, available at https://www.oecd.org/tax/beps/.

43

The ideas of fairness and equal opportunities concerned the ability of national jurisdictions to tax, and not the (substantive) fairness of taxation among different taxpayers.

44

Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting, available at https://www.oecd.org/tax/beps/multilateral-convention-to-implement-tax-treaty-related-measures-to-prevent-BEPS.pdf.

45

See, for instance, Action Plan on Base Erosion and Profit Shifting (oecd Publishing, Paris, 2013).

46

The use and the relevance of non-mainstream unilateral tax solutions depend on local circumstances, Kofler and Sinnig, op.cit. note 40, 113 (and the literature cited).

47

oecd Action Plan, op.cit. note 45, 10–11.

48

The possibility of unilateral action which is more stringent that the beps Framework was, however, accepted. See Addressing Base Erosion and Profit Shifting (oecd Publishing, Paris, 2013), 51.

49

Christians, op.cit. note 38, 27–28, and Tim Büttner and Matthias Thiemann, “Breaking Regime Stability? The Politicization of Expertise in the oecd/G20 Process on beps and the Potential Transformation of International Taxation,” 7(1) Accounting, Economics, and Law: A Convivium (2017), 1–16 (and the literature cited).

50

Kofler and Sinnig, op.cit. note 40, 114.

52

Ibid., para. 302 (such a levy is able to tax the value created in the given national economy).

53

Ibid.

54

See, Statement by the oecd/G20 Inclusive Framework on beps on the Two-Pillar Approach to Address the Tax Challenges Arising from the Digitalisation of the Economy, available at https://www.oecd.org/tax/beps/statement-by-the-oecd-g20-inclusive-framework-on-beps-january-2020.pdf, and the 2018 oecd Interim Report on the Tax Challenges of Digitalisation, available at https://www.oecd.org/ctp/tax-challenges-arising-from-digitalisation-interim-report-9789264293083-en.htm, para. 346.

55

The interim report noted that states will “secure an appropriate tax base” in the national tax jurisdiction in their own way(s), either by taxing profit or “some other equivalent factor” in measures addressed predominantly to multinationals, ibid. The different taxation routes followed are: alternative applications of the permanent establishment threshold, withholding taxes, specific regimes to deal with multinationals, and turnover taxes, ibid., para. 347.

56

Kofler and Sinnig, op.cit. note 40, 106.

57

See Mario Tenore, “Trends and facts in European tax integration: harmonization and coordination,” in Pasquale Pistone (ed.), European Tax Integration: Law, Policy and Politics (ibfd Publishing, Amsterdam, Netherlands, 2018), 3–20, at 4.

58

Proposal for a Council Directive on a Common Corporate Tax Base, com(2016) 685 final.

59

Proposal for a Council Directive on a Common Consolidated Corporate Tax Base (ccctb), com(2016) 683 final.

60

Council Directive (EU) 2016/1164 of 12 July 2016 laying down rules against tax avoidance practices that directly affect the functioning of the internal market, oj L 193/1. The policy proposals put forward by the Commission in 2020 – partly as a reaction to the covid-19 pandemic and its consequences – do not include non-conventional taxation components. See Commission communication – Fair and simple taxation supporting the recovery strategy, com(2020) 312 final, and Commission communication – Tax Good Governance in the EU and beyond, com(2020) 313 final.

61

Political Statement: Joint Initiative on the Taxation of Companies Operating in the Digital Economy, available at https://www.consilium.europa.eu/media/37181/council-conclusions-on-digital-taxation-nov-2017.pdf.

62

As an “immediate, supplementary and short-term” solution.

63

Commission communication – A Fair and Efficient Tax System in the European Union for the Digital Single Market, com(2017) 547 final, 9–10.

64

Ibid., 10.

65

Proposal for a Council Directive laying down rules relating to the corporate taxation of a significant digital presence, com(2018) 147 final, and Proposal for a Council Directive on the common system of a digital services tax on revenues resulting from the provision of certain digital services, com(2018) 148 final.

66

Kofler and Sinnig, op.cit. note 40, 113. Some Member States already have in place a digital economy special tax, others are less enthusiastic about the introduction of an equalization tax, and it seems that at EU level the common approach supported will be the eventual implementation of the oecd/G20 multilateral framework, see https://www.consilium.europa.eu/hu/policies/digital-taxation/.

67

For an overview analysis, see Dániel Deák, “Legislating Unorthodox Taxes: The Hungarian Experience,” 36(3) Society and Economy (2014), 339–368. For an analysis focused on turnover taxes, see Dániel Deák, “Szankcionálható-e az árbevételre vetített progresszív adó?,” lxiv(9–10) Külgazdaság (2020), 89–116.

68

The taxation of turnover seems to rest on a mere assumption of taxability having regard to the taxpayer’s size.

69

See, in this regard, Christoph Jeschek, “Debate: Taxes on Digital Services and the Substantive Scope of Application of Tax Treaties: Pushing the Boundaries of Article 2 of the oecd Model?,” 46(6/7) Intertax (2018), 573–578.

70

Kofler and Sinnig, op.cit. note 40, 114 (practically, they work as “special excise taxes”). Such solutions are attractive for national tax administrations because they offer a relatively uncomplicated taxation response, which avoids the problem of setting rules for profit attribution and for the allocation of tax jurisdiction, ibid.

71

Interim Report, op.cit. note 54, paras 359 and 361.

72

Dario Stevanato, “Are Turnover-based Taxes a Suitable Way to Target Business Profits?,” 59(11) European Taxation (2019) 538–546, at 544, and Julia Sinnig, “Turnover Taxes Under State Aid Spotlight,” 59(2/3) European Taxation (2019) 106–112, at 109.

73

ecj, Case C-385/12, Hervis (2014) EU:C:2014:47, paras 38–39.

74

The state aid claim in the cases was found inadmissible by the Court, ecj, Case C-75/18, Vodafone (2020) EU:C:2020:139, paras 28–31; ecj, Case C-323/18, Tesco-Global (2020) EU:C:2020:140, paras 40–43.

75

ecj, Case C-75/18, Vodafone, para. 44; ecj, Case C-323/18, Tesco, para. 64.

76

ecj, Case C-75/18, Vodafone, para. 52; ecj, Case C-323/18, Tesco, para. 71.

77

See, in this regard, ecj, Case C-75/18, Vodafone, paras 45–48; ecj, Case C-323/18, Tesco, paras 65–68.

78

ecj, Case C-75/18, Vodafone, paras 51–52; ecj, Case C-323/18, Tesco, paras 70–71.

79

See Phedon Nicolaides, “Has an Economic Myth Become a Legal Fact? The Case of Turnover Taxes”, https://www.lexxion.eu/en/stateaidpost/has-an-economic-myth-become-a-legal-fact-the-case-of-turnover-taxes/, and Ruth Mason and Leopoldo Parada, “The Illegality of Digital Services Taxes Under EU Law: Size Matters,” 92 Tax Notes International (2018), 1183–1197.

80

Parada argued that without investigating the true intent of the Hungarian policy-maker, the legitimacy of the taxes cannot be established. Leopoldo Parada, “How the Vodafone Magyarország Opinion Affects EU Debate on Turnover-based Digital Taxes,” 95 Tax Notes International (2019), 399–407.

81

Opinion of ag Kokott in ecj, Case C-75/18, Vodafone Magyarország (2020) EU:C:2019:492, para. 100 and Opinion of ag Kokott in ecj, Case C-323/18, Tesco-Global (2020) EU:C:2019:567, para. 102.

82

See, in this regard, Stevanato, op.cit. note 72, 544, and Sinnig, op.cit. note 72, 109.

83

Its administrative provisions regulating cross-border compliance were successfully challenged under the fundamental freedoms in ecj, Case C-482/18, Google Ireland Limited (2020) EU:C:2020:141.

84

Commission Decision (EU) 2017/329 of 4 November 2016 on the measure sa.39235 (2015/C) (ex 2015/nn) implemented by Hungary on the taxation of advertisement turnover, oj L 49/36.

85

ecj, Case T-20/17, Hungary v Commission (2019) EU:T:2019:448, paras 87–90. See also the identically worded judgment concerning the Polish commercial retail tax in ecj, Joined Cases T-836/16 and T-624/17, Poland v Commission (2019) EU:T:2019:338.

86

ecj, Case T-20/17, Hungary v Commission, para. 89.

87

Ibid., paras 92, 103, 105.

88

Ibid., para. 110.

89

ecj, Case C-596/19 P, Commission v Hungary (2019) EU:C:2021:202, paras. 43–44 and 46.

90

Ibid., paras 45–46.

91

Ibid., paras 46–47.

92

Ibid.

93

Ibid., paras 49–50.

94

See also Szudoczky and Károlyi observing that the Hungarian legislature did not provide much help in determining the genuine aim of the advertisement tax. Rita Szudoczky and Balázs Károlyi, “The Troubled Story of the Hungarian Advertisement Tax: How (Not) to Design a Progressive Turnover Tax,” 48(1) Intertax (2020) 46–66, at 49, 52 and 54.

95

The Court explicitly recognized that turnover is “merely a relative indicator of ability to pay”.

96

See, in this regard, the criticism in Phedon Nicolaides, “Multi-rate Turnover Taxes and State Aid: A Prelude to Taxes on Company Size,” 18(3) European State Aid Law Quarterly (2019), 226–238, at 236. See also Niels Tack, “The 2019 Autumn Conference of the European State Aid Law Institute,” 19(2) European State Aid Law Quarterly (2020), 101–104, at 104.

97

Sinnig, op.cit. note 72, 110, and Szudoczky and Károlyi, op.cit. note 94, 53.

98

See Stevanato, op.cit. note 72, 540.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1068 225 14
PDF Views & Downloads 1327 307 16