ArticlePDF Available

Selectivity of State aid and progressive turnover taxes – Leaving the door (too) wide open?: Commission v. Poland

February 2022
Common Market Law Review 59(Issue 1):187-202

DOI:10.54648/COLA2022009

Authors:

Maciej Bernatt

University of Warsaw

Lukasz Grzejdziak

University of Łódź

In case C-562/19, Commission v. Poland, the Court of Justice confirmed its settled case law that the EU rules on State aid apply to tax measures which have not been harmonized under EU law. The Court also reaffirmed that tax measures must be reviewed under Article 107(1) TFEU using the three-element selectivity test. The ruling, however, brought important new developments. The case annotation discusses the shortcomings of the Court’s approach to the selectivity test, in particular in relation to the finding that a progressive tax rate must be considered a part of the reference system. It also explains why turnover is not necessarily – contrary to the Court’s view – a neutral criterion. Finally, it explains why the ruling opens the door for Member States to introduce such turnover tax systems which are de facto selective and may aim at pushing an anti-EU, patriotic economic agenda.

Content uploaded by Maciej Bernatt

Content may be subject to copyright.

COMMON MARKET LAW REVIEW

CONTENTS Vol. 59 No. 1 February 2022

Editors and publishers 1-2

Guest Editorial: EU emergency law and its impact on the EU legal

order, Bruno De Witte 3-18

Articles

E. Chiti, Managing the ecological transition of the EU: The European

Green Deal as a regulatory process 19-48

S. Peers, So close, yet so far: The EU/UK Trade and Cooperation

Agreement 49-80

H. van Kolfschooten, EU regulation of artificial intelligence: Chal-

lenges for patients’ rights 81-112

Case law

A. Court of Justice

Trust until it is too late! Mutual recognition of judgments and limita-

tions of judicial independence in a Member State: L and P,

A. Frackowiak-Adamska 113-150

Remedies for non-material damages: Striking out in a new direction?

Braathens, A. Wallerman Ghavanini 151-170

A (more) complete system of remedies: Effective judicial protection

of EU Member States in Czech Republic v. Commission,

N. Bacˇic´ Selanec 171-186

Selectivity of State aid and progressive turnover taxes – Leaving the

door (too) wide open?: Commission v. Poland, M. Bernatt and

Ł. Grzejdziak 187-202

B. National Courts

National security as an exception to EU data protection standards:

The judgment of the Conseil d’État in French Data Network

and others, A. Turmo 203-222

C. EFTA Court

On an equal footing. The EFTA Court’s ruling in the Norwegian

Social Security scandal: Criminal proceedings against

N, T. Bekkedal 223-238

Review essay:Je t’aime...moinonplus: Ten years of application of

the EU Charter of Fundamental Rights, X. Groussot and

G. Thor Pétursson 239-258

Book reviews 259-288

Survey of Literature 289-312

Aims

The Common Market Law Review is designed to function as a medium for the understanding and

implementation of European Union Law within the Member States and elsewhere, and for the

dissemination of legal thinking on European Union Law matters. It thus aims to meet the needs of

both the academic and the practitioner. For practical reasons, English is used as the language of

communication.

transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or

otherwise, without written permission from the publisher.

Permission to use this content must be obtained from the copyright owner. More information can

be found at: lrus.wolterskluwer.com/policies/permissions-reprints-and-licensing

Common Market Law Review is published bimonthly.

This journal is also available online. Online and individual subscription prices are available upon

request. Please contact our sales department for further information at +31(0)172 641562 or at

lnternational-sales@wolterskluwer.com.

Periodicals postage paid at Rahway, N.J. USPS no. 663–170.

U.S. Mailing Agent: Mercury Airfreight International Ltd., 365 Blair Road, Avenel, NJ 07001.

Published by Kluwer Law International B.V., P.O. Box 316, 2400 AH Alphen aan den Rijn,

The Netherlands

Printed on acid-free paper.

COMMON MARKET LAW REVIEW

Personal subscription prices at a substantially reduced rate as well as online subscription prices

are available upon request. Please contact our sales department for further information

at +31 172641562 or at International-sales@wolterskluwer.com.

Payments can be made by bank draft, personal cheque, international money order, or UNESCO

coupons.

All requests for further information

and specimen copies should be addressed to:

Kluwer Law International

P.O. Box 316

2400 AH Alphen aan den Rijn

The Netherlands

fax: +31 172641515

For Marketing Opportunities please contact International-marketing@wolterskluwer.com

Please visit the Common Market Law Review homepage at http://www.kluwerlawonline.com

for up-to-date information, tables of contents and to view a FREE online sample copy.

Consent to publish in this journal entails the author’s irrevocable and exclusive authorization

of the publisher to collect any sums or considerations for copying or reproduction payable by

third parties (as mentioned in Article 17, paragraph 2, of the Dutch Copyright Act of 1912 and

in the Royal Decree of 20 June 1974 (S.351) pursuant to Article 16b of the Dutch Copyright

Act of 1912) and/or to act in or out of court in connection herewith.

Microﬁlm and Microﬁche editions of this journal are available from University Microﬁlms

International, 300 North Zeeb Road, Ann Arbor, MI 48106, USA.

The Common Market Law Review is indexed/abstracted in Current Contents/Social &

Behavioral Sciences; Current Legal Sociology; Data Juridica; European Access; European

Legal Journals Index; IBZ-CD-ROM: IBZ-Online; IBZ-lnternational Bibliography of Peri-

odical literature on the Humanities and Social Sciences; Index to Foreign Legal Periodicals;

International Political Science Abstracts; The ISI Alerting Services; Legal Journals Index;

RAVE; Social Sciences Citation Index; Social Scisearch.

COMMON MARKET LAW REVIEW

Editors: Thomas Ackermann, Loïc Azoulai, Marise Cremona, Michael Dougan, Christophe

Hillion, Giorgio Monti, Niamh Nic Shuibhne, Ben Smulders, Stefaan Van den Bogaert

Advisory Board:

Ulf Bernitz, Stockholm

Kieran Bradley, Luxembourg

Alan Dashwood, Cambridge

Jacqueline Dutheil de la Rochère, Paris

Claus-Dieter Ehlermann, Brussels

Giorgio Gaja, Florence

Daniel Halberstam, Ann Arbor

Gerard Hogan, Luxembourg

Laurence Idot, Paris

Francis Jacobs, London

Jean-Paul Jacqué, Brussels

Pieter Jan Kuijper, Amsterdam

Miguel Poiares Maduro, Lisbon

Ulla Neergaard, Copenhagen

Siofra O’Leary, Strasbourg

Sacha Prechal, Luxembourg

Allan Rosas, Luxembourg

Wulf-Henning Roth, Bonn

Eleanor Sharpston, Luxembourg

Piet Jan Slot, Amsterdam

Christiaan W.A. Timmermans, Brussels

Ernö Várnáy, Debrecen

Armin von Bogdandy, Heidelberg

Joseph H.H. Weiler, New York

Jan A. Winter, Bloemendaal

Miroslaw Wyrzykowski, Warsaw

Managing Editor: Alison McDonnell

Common Market Law Review

Europa Instituut

Steenschuur 25

2311 ES Leiden

The Netherlands tel. + 31 71 5277549

e-mail: a.m.mcdonnell@law.leidenuniv.nl fax: + 31 71 5277600

Establishment and Aims

Editorial policy

Submission of manuscripts

Manuscripts should be submitted together with a covering letter to the Managing Editor. They

must be accompanied by written assurance that the article has not been published, submitted

or accepted elsewhere. The author will be notified of acceptance, rejection or need for revision

within three to nine weeks. Digital submissions are welcomed. Articles should preferably be

no longer than 28 pages (approx. 9,000 words). Annotations should be no longer than 10 pages

(approx. 3,000 words). Details concerning submission and the review process can be found on

the journal's website http://www.kluwerlawonline.com/toc.php?pubcode=COLA

The editors will consider for publication manuscripts by contributors from any country.

Articles will be subjected to a review procedure. The author should ensure that the

significance of the contribution will be apparent also to readers outside the specific

expertise. Special terms and abbreviations should be clearly defined in the text or notes.

Accepted manuscripts will be edited, if necessary, to improve the general effectiveness of

communication. If editing should be extensive, with a consequent danger of altering the

meaning, the manuscript will be returned to the author for approval before type is set.

The Common Market Law Review was established in 1963 in cooperation with the British

Institute of International and Comparative Law and the Europa Instituut of the University

of Leyden.The Common Market Law Review is designed to function as a medium for the

understanding and analysis of European Union Law, and for the dissemination of legal

thinking on all matters of European Union Law. It aims to meet the needs of both the

academic and the practitioner. For practical reasons, English is used as the language of

communication.

Selectivity of State aid and progressive turnover taxes – Leaving the

door (too) wide open?: Commission v. Poland

Case C-562/19 P, Commission v. Poland, Judgment of the Court of Justice

(Grand Chamber) of 16 March 2021, EU:C:2021:201

1. Introduction

The selectivity test is arguably the most complex part of any State aid

proceeding. This is particularly true as far as the assessment of national

taxation systems is concerned. The judgment annotated here offers important

developments regarding the understanding of the selectivity test, and thus

merits a closer scrutiny.

In Commission v. Poland, the Court confirmed its settled case law that the

EU rules on State aid apply to tax measures which have not been harmonized

under EU law. The Court also reaffirmed that tax measures must be reviewed

under Article 107(1)TFEU using the three-element selectivity test developed

in Paint Graphos and the subsequent case law.

The ruling, however, brought

important new developments. First, the Court’s reasoning sheds new light on

the relationship between the tax sovereignty of Member States and

competition. By its reasoning, based on Member States’ fiscal discretion, as

well as by limiting the possibility of considering the reference system as

inherently selective, the Court supported the broad concept of tax sovereignty

of Member States. Second, the Court clarified the test of selectivity of tax

measures, in particular in relation to the reference system and its application to

progressive taxation including turnover taxes. Third, the Court also specified

when the reference system is inherently selective, thus refining its Gibraltar

case law.

1. Joined Cases C-78-80/08, Paint Graphos et al., EU:C:2011:550. See also Case C-374/17,

Finanzamt B v. A-Brauerei, EU:C:2018:1024; Case C-233/16, Asociación Nacional de Grandes

Empresas de Distribución (ANGED) v. Generalitat de Catalunya, EU:C:2018:280; Joined

Cases C-234 & 235/16, ANGED v. Consejería de Economía y Hacienda del Principado de

Asturias i Consejo de Gobierno del Principado de Asturias, EU:C:2018:281; Joined Cases

C-236 & 237/16, ANGED v. Diputación General de Aragón, EU:C:2018:291; Case C-323/18,

Tesco-Global Áruházak Zrt. v. Nemzeti Adó- és Vámhivatal Fellebbviteli Igazgatósága,

EU:C:2020:140; Joined Cases C-106 & 107/09 P, Commission and Kingdom of Spain v.

Government of Gibraltar and United Kingdom, EU:C:2011:732; Joined Cases C-20 & 21/15 P,

Commission v. World Duty Free Group et al., EU:C:2016:981.

2. Joined Cases C-106 & 107/09 P, Commission and Spain v. Government of Gibraltar and

UK.

Common Market Law Review 59: 187–202, 2022.

This annotation discusses the shortcomings of the Court’s approach to the

selectivity test, in particular in relation to the finding that a progressive tax rate

must be considered a part of the reference system. It also explains why

turnover is not necessarily – contrary to the Court’s view – a neutral criterion.

Finally, it explains why the ruling opens the door for Member States to

introduce such turnover tax systems which are de facto selective and may aim

at pushing an anti-EU, patriotic economic agenda.

This annotation does not cover the second ground of appeal, relating to the

Commission’s decision to initiate a formal investigation procedure against

Poland and to adopt a suspension injunction. The ECJ judged that the second

ground of appeal was based on a misinterpretation of the GC ruling, so the

judgment did not result in important modifications of the well-established

case law, nor any significant clarification.

The annotation is also relevant for the ECJ’s judgment (on the same day) in

the Hungarian advertisement tax case, as it concerns similar questions and

offers similar answers.

2. Factual and legal background

On 6 July 2016, the Polish Sejm adopted theAct on retail sales tax.

The Act

introduced a new progressive tax on turnover from the sale of goods to

consumers. The taxable persons were retailers of a various legal status

including natural persons, partnerships and companies. The tax base was the

amount of turnover from retail sales achieved in a given month in excess of

PLN 17 million (roughly EUR 3.75 million). The tax was supposed to be

payable on a monthly basis. The following tax rates were established:

– 0.8% on the part of an undertaking’s monthly turnover from retail

sales over PLN 17 million, but not exceeding PLN 170 million

(roughly EUR 37.5 million) and

– 1.4% on the part of an undertaking’s monthly turnover from retail

sales above PLN 170 million.

The tax was characterized by bracketed progressivity, meaning that different

tax rates were applied to different parts (brackets) of the turnover of all

undertakings.

3. Case C-596/19 P, European Commission v. Hungary, EU:C:2021:202, paras. 29–53.

Please note that this annotation does not cover the Commission’s second ground of appeal in the

Hungarian tax advertisement case alleging an infringement of Art. 107(1) TFEU in that the

General Court held that the mechanism for the partial deductibility of losses carried forward did

not constitute a selective advantage (paras. 54–69).

4. Journal of Laws 2020, Item 1293, consolidated text.

CML Rev. 2022188 Case law

According to the explanatory statement accompanying the Act, its main

objective was to increase tax revenues of the State budget allocated to the

financing of budgetary expenditure resulting from the implementation of the

support scheme for raising children “Family 500 plus”, as one of the sources

of its financing, and to obtain additional income to compensate for the

budgetary loss that arises as a result of the application of optimization

practices regarding corporate income tax.

The measure had not been notified to the Commission, but the Commission

has carefully observed the situation and, as early as February 2016, requested

the Polish Government for information regarding the planned introduction of

the retail tax. On 19 September 2016, the Commission adopted a decision to

initiate a formal investigation procedure against Poland and issued a

suspension injunction pursuant to Article 13(1) of the procedural Regulation.

On 30 June 2017, the Commission adopted a negative decision without

recovery, as the measure had not been implemented yet in practice. The

Commission considered the measure selective, as the tax tended to favour

certain retailers compared with others, depending on their relative turnover.

According to the Commission, the reference system covered retail tax but not

its progressive structure. The Commission identified three groups of

taxpayers, including one covering retailers not obliged to pay the tax due to

their turnover being below the first threshold. Then the Commission identified

average tax rates to be paid by a hypothetical member of each of these groups.

As these rates differed significantly, it was held that the tax measure resulted

in a derogation from the reference system, which was deemed to have a single

rate. The Commission’s decision was challenged by Poland before the General

Court. In its judgment, the GC found that the Commission had erred in

considering that the imposition of progressive tax on turnover generated by the

retail sale of goods gave rise to a selective advantage.

As a result, the GC

annulled the Commission’s decision.

A month and a half later, the GC annulled the Commission’s decision in the

Hungarian advertisement tax case,

holding that the Commission had erred in

classifying the tax measure as the selective advantage. This tax measure was

similar to the Polish one, for example with respect to bracketed progressivity.

5. Explanatory statement to the bill of Act on Retail Sales, p. 1, Sejm of 8th term, Sejm

paper no. 615 of 15 June 2016.

6. Council Regulation (EU) 2015/1589 of 13 July 2015 laying down detailed rules for the

application of Art. 108 [TFEU], O.J. 2015, L 248/9.

7. Joined Cases T-836/16 & T-624/17, Poland v. Commission, EU:T:2019:338.

8. Commission Decision (EU) 2017/329 of 4 Nov. 2016 on the measure SA.39235 (2015/C)

(ex 2015/NN) implemented by Hungary on the taxation of advertisement turnover, O.J. 2017, L

49/36.

9. Case T-20/17, Hungary v. European Commission, EU:T:2019:448.

Case C-562/19 P 189

The taxable amount was net turnover generated by the broadcasting or

publication of advertisements. Originally, six progressive tax rates had been

established. In 2015 Hungary changed the tax rates, which from then on were

as follows: 0% for the part of the taxable amount below HUF 100 million

(approximately EUR 280,000) and 5.3% for the part of the taxable amount

above the latter amount. The Commission, basing its reasoning on similar

arguments to those expressed in the Polish retail tax case, held that the

Hungarian measure granted a selective advantage to undertakings with a lower

turnover (small undertakings). The GC disagreed and annulled the

Commission’s decision.

3. Opinion of the Advocate General

In the Opinion of 15 October 2020, Advocate General Kokott proposed to

dismiss the appeal brought by the Commission. At the outset, she noted that

under recent case law on fundamental freedoms, national progressive taxation

systems based on turnover do not violate EU law.

Advocate General Kokott

was of the view that the same approach should be applicable within the State

aid area. She underlined that the amount of turnover constitutes a criterion of

differentiation that is neutral and that turnover constitutes a relevant indicator

of a taxable person’s ability to pay. Second, she stressed the importance of the

tax autonomy of Member States. According to her views, there is nothing in

EU law that prescribes a specific structure for the national taxes of the

Member States (except for harmonized taxes).The decision in this respect lies

within the competences of the national legislature. A progressive tax system

based on turnover is one of the legitimate solutions Member States may opt

for. In her view, a progressive tax system based on turnover can be challenged

only in exceptional circumstances, fitting within the narrow grounds

determined in the ECJ ruling in the Gibraltar case,

that is, in situations when

Member States abuse their tax sovereignty and act in an inconsistent manner.

In this respect, she noted that Gibraltar’s law on corporate tax “was intended

solely to circumvent the rules on State aid by using purportedly general

profit-based income taxation to establish very low taxation of certain

companies which were intended to generate income (offshore companies).”

Advocate General Kokott distinguished such a situation from the contested

Polish retail tax. In her view, the General Court rightly found that no such

10. Case C-75/18, Vodafone Magyarország, EU:C:2020:139, para 50, and Case C-323/18,

Tesco-Global Áruházak, EU:C:2020:140, para 70.

11. Joined Cases C-106 & 107/09 P, Government of Gibraltar.

12. Opinion in Case C-562/19 P, Commission v. Poland, EU:C:2020:834, para 43.

CML Rev. 2022190 Case law

inconsistency existed in the Polish tax on the retail sector. Confirming the

GC’s position, she underlined that the progressive structure of Polish law

resulted in heavier taxation of undertakings with a higher turnover and lower

taxation of undertakings with a lower turnover, following the redistributive

purpose associated with a progressive rate. Consequently, a selective

advantage could not be inferred solely from the progressive structure. Since

the Commission did not assert or demonstrate any other inconsistencies, the

Polish law could not be regarded as State aid.

4. Judgment of the Court

The ECJ dismissed the Commission’s appeal of the General Court ruling. The

Grand Chamber repeated its case law that the classification of a national

measure as “State aid” within the meaning of Article 107(1)TFEU requires all

of the following conditions to be fulfilled. First, there must be an intervention

by the State or through State resources. Second, the intervention must be liable

to affect trade between the Member States. Third, it must confer a selective

advantage on the recipient. Fourth, it must distort, or threaten to distort,

competition. In the case at stake, selectivity was the key issue. The Court

observed that in order to classify a national tax measure as “selective”, the

Commission must begin by identifying the reference system, or “normal” tax

system applicable in the Member State concerned. Thereafter, the

Commission must demonstrate that the contested tax measure is a derogation

from that reference system, insofar as it differentiates between operators who,

in the light of the objective pursued by that system, are in a comparable factual

and legal situation.

The central issue here was, therefore, for the Court to decide whether the

progressivity of rates provided for by the Polish retail tax system can be

deemed as an integral part of such a reference system, or whether, as the

Commission argued, the reference system relevant for the assessment of

Polish retail tax consists of a flat-rate turnover tax. When deciding on this

matter, the Court shared Advocate General Kokott’s view that in the light of

existing case law,

Member States are free to establish the system of taxation

which they deem most appropriate, meaning that the application of

progressive taxation falls within the discretion of each Member State.

According to the Court, Member States are free to opt for a progressive and

turnover-based system of taxation. In consequence, the Court disagreed with

the Commission and held that the progressivity of rates provided for by the

13. Judgment, para 31.

14. See cases mentioned supra note 1.

Case C-562/19 P 191

Polish retail tax system, and its turnover basis, are an integral part of the

reference system. At the same time, the Court did not rule out that the

characteristics constituting the tax, which include progressive tax rates, may,

in certain cases, reveal a manifestly discriminatory element; however, this is

for the Commission to demonstrate.

Building on Advocate General Kokott’s

arguments, the Court noted that Gibraltar’s law on corporate tax had been

configured according to manifestly discriminatory parameters, intended to

circumvent EU law on State aid. By contrast, the Court found that the

Commission had not established that the progressivity of the rates of Polish

retail tax was designed in a manifestly discriminatory manner, with the aim of

circumventing the requirements of EU law on State aid. In those

circumstances, the progressivity of the rates of the contested Polish tax

measure had been regarded by Courts as inherent to the reference system, in

the light of which the existence of a selective advantage had to be assessed.

With this in mind, the Court agreed with the GC that the Commission relied on

an incomplete and fictitious reference system.

5. Comments

As far as substance is concerned, the Court’s position in the annotated

judgment fully corresponds with its position in the Hungarian advertisement

tax case.

In particular, the interpretation of the notion of selectivity of State

aid in relation to a turnover-based progressive taxation system, is the same.

Therefore, the comments offered below apply also to this judgment.

5.1. Progressive taxation and a State aid reference system

In order to be classified as State aid, within the meaning of Article 107(1)

TFEU, a measure must favour certain undertakings or the production of

certain goods. Selectivity of tax measures mitigating the normal charges of

undertakings is assessed by means of the three-element test. First, it is

necessary to establish a system of reference, i.e. a normal tax system

applicable in the Member State concerned. Then, it must be assessed whether

a measure differentiates between operators who, in the light of the objective

pursued by that system, are in a comparable factual and legal situation.

Finally, for measures fulfilling the second step of the test, and thus being

15. Judgment, para 42.

16. Case C-596/19 P, European Commission v. Hungary, paras. 32–53.

17. Joined Cases C-78-80/08, Paint Graphos et al., para 49; see also Case C-88/03, Portugal

v. Commission, EU:C:2006:511, para 56; Case C-374/17, Finanzamt B v. A-Brauerei, para 36;

CML Rev. 2022192 Case law

prima facie selective, it is required to assess whether such a derogation is

justified by the nature or the general scheme of the system.

The core element of the dispute in the Polish retail tax and Hungarian

advertisement tax cases, was the question of including the progressive nature

of the tax rate in the reference system. Before 2020, ECJ case law gave no clear

guidance on how to assess the selectivity of progressive tax systems.

Admittedly, it was clear that progressive rates of income tax constituted (and

still constitute) a general (i.e. non-selective) measure. However, the case law

did not address specifically the question at what stage of the selectivity of tax

measures test the general character of such measures is recognized, and, in

particular, whether a progressive tax rate should be considered as part of the

reference system. Moreover, progressive turnover taxes have not previously

been the subject of an ECJ ruling.

On the eve of the General Court rulings in the Polish retail tax and the

Hungarian advertisement tax cases, it seemed that a dominant view was that

progressive tax rates were not an element of the reference system. This

position was clearly stated by Advocate General Saugmandsgaard Øe in

A-Brauerei. Referring to the selectivity of progressive income tax, he pointed

out that: “[u]nder the reference framework method, the more favourable rates

constitute a differentiation which must be validated either by the lack of

comparability (second stage) or by the existence of a justification based on the

nature or overall structure of the system at issue (third stage).”

Similarly, the

Commission named “the progressive nature of income tax and its

redistributive purpose” as one of the bases to justify prima facie selective

measures by referring to the nature or general scheme of the system of

reference.

These views exclude progressive income tax rates from the

reference system, and recognize their general nature at the second or third

stage of the selectivity test.

In the annotated ruling, the Court took the opposite position, based on

arguments related to the tax autonomy of the Member States. In this regard, the

ECJ referred to two of its earlier preliminary rulings concerning Hungarian

Case C-233/16, ANGED, para 40; Joined Cases C-234 & 235/16, ANGED, para 33; Joined

Cases C-236 & 237/16, ANGED, para 28; Joined Cases C-20 & 21/15 P, Commission v. Wo rl d

Duty Free Group et al., para 57.

18. Joined Cases C-78-80/08, Paint Graphos et al., para 64; Case C-308/01, GIL Insurance,

EU:C:2004:252; Case C-143/99, Adria-Wien Pipeline, EU:C:2001:598, para 42; Case

C-374/17, Finanzamt B v. A-Brauerei, paras. 51 and 52; Joined Cases C-106 & 107/09 P,

Gibraltar, para 145.

19. Opinion of A.G. Saugmandsgaard Øe in Case C-374/17, Finanzamt B v. A-Brauerei,

EU:C:2018:741, para 66.

20. C/2016/2946, Commission Notice on the notion of State aid as referred to in Article

107(1) of the Treaty on the Functioning of the European Union, para 138, O.J. 2016, C

262/1–50.

Case C-562/19 P 193

progressive turnover taxes, that is, a retail sales tax and a special tax for

specific sectors of the economy.

In these rulings, the ECJ found that these

taxes did not infringe the principle of the freedom of establishment, as the

application of progressive taxation, in the absence of a harmonization of a

specific tax, falls within the scope of the discretion of the Member States to

establish their own taxation system. This also applies to setting turnover tax

rates.

The Court now applied this concept by analogy to the State aid rules.

Progressive taxation (regardless of whether it concerns turnover or income),

as well as the determination of the basis of assessment and the taxable event,

has been recognized as an element which is not selective as such. Commission

v. Poland may therefore be interpreted as a de facto exclusion of progressive

taxation from EU State aid rules. The only exception regards taxation systems

which “reveal a manifestly discriminatory element”.

In this respect, the judgment raises several doubts. The first refers to the

paucity of the Court’s argumentation, which is limited to a reference to the tax

autonomy of the Member States and the neutral nature of the turnover

criterion. In general, the Court emphasized that this autonomy was limited by

EU law, including the State aid rules. However, by framing this autonomy in a

broad manner, the ECJ gave priority to Member States’ tax sovereignty over

EU State aid rules and, thus, competition rules. Moreover, the Court did not

explain at all why it departed from the methodology of the ANGED ruling,

where tax autonomy was analysed within the second stage of the selectivity

test and not the first one (i.e. the determination of the reference system).

Second, by introducing the subjective element of the selectivity assessment

into the first stage of the three-step selectivity test, the Court distorted the

logical and clear structure of the test. The establishment of the reference

system, and the identification of a derogation from it, should be objective in

nature and limited to the establishment that law differentiates between entities

which are in a comparable situation. The assessment of subjective factors,

such as the taxpayer’s ability to pay the tax, should take place at the subsequent

two stages of the test, covering the determination of whether the derogation

from the reference system causes differentiated treatment of undertakings in a

similar factual and legal situation in the light of the intrinsic objective of the

reference system.

Only at this point does the possibility of the assessment of

factors such as the “neutrality” of the differentiating criterion and the payment

capacity of taxpayers arise. Such elements could also be reviewed during the

21. Case C-75/18, Vodafone, para 50, and Case C-323/18, Tesco, para 69.

22. Case C-75/18, Vodafone, para 50, and Case C-323/18, Tesco, para 70.

23. Joined Cases C-234 & 235/16, ANGED, para 43; Joined Cases C-236 & 237/16,

ANGED, para 38; Case C-233/16, ANGED, para 50.

24. Joined Cases C-78-80/08, Paint Graphos et al., para 61.

CML Rev. 2022194 Case law

third stage of the test within factors justifying the differentiation between

undertakings on the grounds of the nature or general scheme of the reference

system. It is more logical to apply such a test structure, as it allows for an easy

identification of the reference system, and eventual deviations from it, which

in turn establishes the basis for the analysis of their causes. As suggested by

the Court’s case law from before the year 2020, the determination of the

reference system did not and should not require an assessment of the reasons

for establishing a specific tax solution.

Still, this approach was abandoned

by the Court in the annotated ruling without sufficient supporting arguments

being offered.

5.2. Turnover as a non-neutral criterion

In Commission v. Poland, the Court referred to the argument of Member

States’ fiscal discretion, as well as limiting the possibility of considering the

reference system as inherently selective. By doing so, it supported the broad

concept of tax sovereignty of EU Member States. Recognizing Member

States’discretion in such a broad way led to the lack of review of the effects of

progressive turnover taxes, such as the Polish retail tax, on competition. The

application of such tax leads to taxpayers with a lower turnover being put in a

disproportionately better competitive position than those with a high turnover.

The Court’s argument that an undertaking’s ability to pay directly depends on

its turnover is disputable. It goes without saying that such ability is determined

directly by the income of the taxpayers, and therefore on their profitability or

the value of the margin charged. An undertaking with a monthly turnover of

EUR 100 million operating at a loss does not have a greater ability to pay the

tax than an undertaking with a turnover of only a tenth of that, but with a

profitability of 10 percent. Already in the case of a flat-rate turnover tax, the

former undertaking would have to pay ten times more tax. Further

differentiation of the amount of the tax levied, as a result of the application of

a progressive tax scale, leads to a systemic and permanent differentiation of

the conditions for conducting business activity between undertakings with the

highest turnover (usually large) and other undertakings. Contrary to the ECJ

reasoning, it is difficult to consider the turnover criterion as neutral, related to

the taxpayer’s ability to pay, or, stricto sensu, redistributive.

The Court’s reasoning disregards market effects of the measure and, instead,

contrary to well-settled case law regarding the objective nature of State aid and

25. See e.g. ibid., paras. 49 and 50.

Case C-562/19 P 195

selectivity,

it refers to criteria of a rather formal nature: the construction of

the measure and the tax autonomy of the Member States.As the Commission

rightly pointed out in its decision, progressive turnover tax with bracketed

progressivity

does not significantly differ in its effects from turnover tax

with global progressivity, in which different tax rates are applied to the whole

turnover of undertakings depending on the size of their turnover, and which

can be considered de jure selective.

The differences between such taxes are limited to two elements. First, in the

case of a tax with bracketed progressivity, the effective tax rate would change

as the taxable turnover increases, while in the case of the tax with global

progressivity, taxpayers in each of a particular tax group would be subject to a

fixed rate. The gradual distribution of the tax burden should not be considered

as an obstacle to the recognition that certain undertakings are in a better

situation than others, but merely that the advantage increases gradually, and is

inversely proportional to this burden. The second difference refers to the same

taxation rules to be applied to all potential taxpayers in the case of a tax based

on bracketed progressivity, as opposed to globally progressive tax. The

application of the same taxation rules, however, according to settled case law

of the ECJ, constitutes only a circumstance determining the de jure

non-selectivity of a measure, but does not exclude its selectivity de facto.

These differences can hardly be considered as justifying a different treatment

of the two taxes.

The amount of turnover is commonly considered as a basic criterion

determining the size of an undertaking. It is one of the three criteria, apart

from staff headcount and the balance sheet total, for classifying undertakings

as SMEs.

Under well-established case law, the size of an undertaking has

26. As the ECJ held in its already classic ruling in Case 173/73, Italian Republic v.

Commission, EU:C:1974:71, para 13: “Accordingly, article [107] does not distinguish between

the measures of State intervention concerned by reference to their causes or aims but defines

them in relation to their effects.”

27. In case of bracketed progression, the higher rate applies only to the excess tax base

above the threshold which triggers a new rate, whereas in global progression, the higher rates

apply to the entire turnover.

28. See the Commission’s State Aid Notice, cited supra note 20, para 121, explaining that

“de jure selectivity results directly from the legal criteria for granting a measure that is formally

reserved for certain undertakings only (for instance: those having a certain size)”. Poland

defended the legality of its retail tax system by arguing that contrary to turnover progression,

bracketed progression does not lead to differentiated taxation between entities that are in a

comparable situation.

29. Joined Cases C-78 to 80/08, Paint Graphos et al., para 52.

30. Annex I to Commission Regulation (EU) 651/2014 of 17 June 2014 declaring certain

categories of aid compatible with the internal market in application of Arts. 107 and 108 of the

Treaty, O.J. 2014, L 187/1–78.

CML Rev. 2022196 Case law

normally been considered as one of the selectivity criteria.

It is, therefore,

legitimate to ask why a measure which results in the differentiation of the

competitive situation of low- and high-turnover retailers, has been considered

neutral and redistributive, while awarding an advantage specifically to all

SMEs in a Member State is normally treated as selective? Indeed, such a

redistributive character could be attributed to progressive income taxes, where

a direct relationship can be established between a tax burden and an

undertaking’s ability to pay a tax measured by its income. The redistributive

nature of progressive turnover taxes, in case of which such a direct link does

not exist, might at best be understood broadly and in an abstract manner – in

the context of a generally worse market position of smaller undertakings (with

lower turnover) in comparison with larger ones. However, the adoption of such

a broad concept of redistribution could justify almost every preferential

treatment of small market players and lead to its classification as a general

measure.

The acceptance by the ECJ of the broad concept of the tax autonomy of

Member States within the area of State aid law opens the possibility for

Member States to structure tax measures in such a way as to arbitrarily

differentiate between certain groups of undertakings without being caught by

Article 107(1) TFEU. Instead of de jure selective instruments taxing only the

largest players in a given sector, Member States could adopt measures with

almost the same market effect by introducing a progressive turnover tax, at

rates fixed on such a level as to cover exactly the same undertakings, though

such tax would still classify as a general measure.

5.3. Limiting the limits of the Gibraltar ruling

One of the key aspects of Commission v. Poland is a narrow interpretation of

the judgment in Gibraltar.

The ECJ held that Gibraltar can be relied on to

find that a given measure is selective only when the “tax system had been

configured according to manifestly discriminatory parameters intended to

circumvent EU law on State aid”.

The Opinion of Advocate General Kokott

suggests that this takes place only once the tax system is built so it is internally

inconsistent, with the objective to favour certain undertakings. Indeed,

Gibraltar’s tax system presented an obvious example of such a situation.

31. See e.g. Joined Cases T-92 & 103/00, Territorio Histórico de Álava – Diputación Foral

de Álava, Ramondín, SA and Ramondín Cápsulas, SA v. Commission, EU:T:2002:61, para 40.

See also the Commission’s State Aid Notice, cited supra note 20, para 121.

32. Joined Cases C-106 & 107/09 P, Government of Gibraltar.

33. Judgment, para 43.

Case C-562/19 P 197

The question is whether other tax systems could be considered selective if

they are internally consistent as far as their legal text is concerned, but in

practice are meant to favour certain undertakings (de facto selectivity). In the

light of the Court’s present interpretation of Gibraltar, the answer seems to be

negative. Such an approach is problematic, however, as it limits the

Commission’s room for manoeuvre in challenging tax systems which are in

practice discriminatory.A very high burden of proof for the Commission for

the future is another consequence of the present judgment. Indeed, in this

ruling (and also in the Hungarian advertisement judgment), the ECJ held that

the Commission had failed to demonstrate that the Polish (and Hungarian) tax

system was built according to manifestly discriminatory parameters. The

Commission failed to persuade the ECJ that the assertion by the Polish

authorities that the progressive structure of the tax helped preserve small-scale

retailers against large format retail, is evidence that those authorities were

seeking to influence the structure of competition in the market. The ECJ

arrived at its conclusion despite the fact that the contested retail tax was

formulated in such a way that higher tax rates apply principally to

foreign-owned retail chains, while keeping domestic retailers within the zero

or first tax rate.

This judgment may encourage other Member States to

follow suit, while leaving little room for the Commission’s intervention under

Article 107(1) TFEU.

5.4. Reinforcement of the anti-EU market changes in CEE Member States

Commission v. Poland raises concerns as to its compliance with the

ideological foundations of EU State aid rules, whose primary objective is to

34. There is no detailed data available regarding the payers of the retail sales tax in Poland.

However, the information provided by the Ministry of Finance shows that in Jan. 2021, the tax

was paid only by 104 taxpayers, of which only 17 were in the group subject to the highest tax

rate. The total amount of the tax paid in Jan. 2021 was PLN 197,377,427.22 (roughly EUR 44

million), of which as much as 80% was paid by the top 10 market players, see <www.

wiadomoscihandlowe.pl/artykul/ujawniamy-wplywy-z-podatku-od-sprzedazy-detalicznej-za-

styczen-nasz-news> (all websites last visited 4 Dec. 2021). It is difficult to determine precisely,

on the basis of the available data, which undertakings belong to this group. According to the list

prepared by Business Insider (see <businessinsider.com.pl/finanse/handel/10-najwiekszych-

firm-handlowych-w-polsce-dane-mf/g742d54>), based on data from the Ministry of Finance

on the revenues of firms in Poland obtained in 2019, the top 10 included only 2 undertakings

controlled by Polish shareholders – Eurocash S.A. and Dino Polska S.A. The former, however,

is not subject to the retail sales tax, as it sells through a network of franchise stores, and, pur-

suant to the provisions of the Act on Retail Sales, it is the franchisees and not the franchisors

that are subject to the tax. Thus, the top 10 largest retailers in Poland are almost entirely sub-

sidiaries of foreign companies. For similar controversies related to the Hungarian advertise-

ment tax, see “RTL Hurt by Hungary’s Advertising Tax”, Financial Times, available at <www.

ft.com/content/cd03780a-6b42-11e4-be68-00144feabdc0>.

CML Rev. 2022198 Case law

prevent Member States from applying measures distorting the Internal

Market.

The specificity of the economies of the Eastern European EU

Member States is manifested in the strong presence of subsidiaries of foreign

investors among the largest market actors in many sectors of the economy,

including retail and media. In many sectors, undertakings with the highest

turnover and subsidiaries of foreign undertakings constitute the same group.

Both Polish and Hungarian turnover taxes have been constructed to weaken

the competitive position of undertakings with foreign capital and favour

smaller market actors of typically national origin.

What may be considered

as a normal or neutral tax policy in many Western EU Member States, is an act

of “economic patriotism” in Poland, Hungary and other Central and Eastern

European EU Member States. In practice, the ECJ ruling opens the floodgate

through which such measures will have a chance to flow. If one forgets for a

moment the elegant concept of the three-element test of selectivity of tax

measures, and instead focuses exclusively on the objectives assigned to the

State aid rules on the one hand, and on the effects of the progressive turnover

taxes on the other, the conclusion can be drawn that the measure reviewed by

the ECJ is not very far from the most evident measures EU State aid rules aim

to eliminate. Favouring companies with national capital against subsidiaries

of foreign undertakings directly conflicts with the internal market objective,

which is a paramount rationale for EU State aid control. Moreover, placing a

tax burden exclusively on the biggest market actors may create a competitive

imbalance causing changes in the market structure, which do not result from

the operation of market mechanisms but from the State intervention.

Therefore, judgment in Commission v. Poland (as well as the parallel ruling

in the Hungarian advertisement tax case) raises yet another question about the

sufficiency of the legal response of EU institutions to the developments of

Central European Member States (particularly in Hungary and Poland), which

35. Nicolaides, “What should State aid control protect? A proposal for the next generation

of State aid rules”, 40 ECLR (2019), 276–283, at 276; Hofmann, “State aid review in a

multi-level system: Motivations for aid, why control it, and the evolution of State aid law in the

EU” in Hoffman and Micheau (Eds.), State Aid Law of the European Union (OUP, 2016), at p.

9; Rusche, Micheau, Piffaut and Van de Casteele, “State Aid” in Faull and Nikpay (Eds.), The

EU Law of Competition (OUP, 2014), at p. 1925; de Cecco, State Aid and the European

Economic Constitution (Hart Publishing, 2013), at p. 38.

36. Mr Tadeusz Kos´cin´ski, Polish Minister of Finance, expressly confirmed that the objec-

tive of the retail tax was to strengthen the market position of Polish stores against their foreign

counterparts: “the retail tax is aimed at leveling the playing field between large discounters, the

costs of which are relatively lowerthan those of Polish stores. If we do nothing, then in 5–10 years

we will have very few Polish stores, and the stores will not be so cheap, because the competition

will not work.” (see “Minister finansów o podatku handlowym: to wyrównanie szans dla małych

sklepów. Teraz pod lupa ecommerce”, available at <www.wiadomoscihandlowe.pl/artykul/

minister-finansow-o-podatku-handlowym-to-wyrownanie-szans-dla-malych-sklepow-teraz-

pod-lupa-ecommerce>. For additional confirmation see also supra note 34.

Case C-562/19 P 199

arguably go against EU values and may violate EU law. While the weakening

of the rule of law, especially as far as the independence of the national

judiciary is concerned, has been subject to intense debate, and the

infringement proceedings launched against Poland had some (although

limited) effects in practice,

changes in the economies of these Member

States, which go against EU values such as open markets, non-discrimination,

and competition, often followed a narrow, technical approach, and so were

subject to a limited scrutiny by the EU institutions. This is true both for the

approach to national legal reforms undermining the reach of EU competition

law,

and bringing risks to fundamental freedoms and leading to the

fragmentation of the EU Single Market.

Arguably, this judgment is a good

example of such a process in the area of State aid.As described more precisely

above, the ECJ accords a wide margin of discretion to Member States. At the

same time, in practice, the Court limits its checks to a formal review of

national legislation, without taking a more nuanced approach in reviewing

what the national legislation actually aims to achieve and what it may entail.

The problem with such an approach is that the national governments of

Hungary and Poland are known for their narrow legalistic approach and the

instrumental use of law that may, in practice, lead to results contrary to the

values on which the EU is founded.

6. Conclusions

By adopting a wide scope for the concept of tax autonomy of Member States,

and tightening the standard of proof required to establish the discriminatory

character of a national tax system, the ECJ effectively limited the scope of

application of EU State aid rules in the area of taxation. The ruling is thus a

step towards increasing the scope of the tax autonomy of Member States at the

expense of the EU Single Market. It opens the door for Member States to

arbitrarily shape their progressive tax systems in such a way as to differentiate

the economic situation of certain market players, and thus distort competition.

37. In this respect see e.g. Scheppele, Kochenov and Grabowska-Moroz, “EU values are

law, after all: Enforcing EU values through systemic infringement actions by the European

Commission and the Member States of the European Union”, 39 YEL (2020), 3–121.

38. See Bernatt, Populism andAntitrust: The Illiberal Influence of Populist Government on

the Competition Law System (Cambridge University Press, 2022).

39. See Varju and Papp, “The crisis, national economic particularism and EU Law: What

can we learn from the Hungarian case?”, 53 CML Rev. (2016), 1647–1674.

40. For a discussion on narrow legalism in the context of democratic backsliding see

Scheppele, “Autocratic Legalism”, 85 University of Chicago Law Review (2018), 545–583, and

Bernatt and Ziółkowski, “Statutory anti-constitutionalism”, 28 Washington International Law

Journal (2019), 485–525.

CML Rev. 2022200 Case law

We expect that the governments of certain Member States will gladly take

advantage of this opportunity.

Maciej Bernatt and Łukasz Grzejdziak

* Respectively:Associate Professor of Law and the Director of the Centre forAntitrust and

Regulatory Studies, Faculty of Management, University of Warsaw; ORCID:

0000-0001-8943-3600; Assistant Professor, University of Łódz´, Faculty of Law and

Administration. Neither author has anything to disclose within the meaning of ASCOLA

Transparency and Disclosure Declaration, <ascola.org/declaration-of-ethics/>.

Case C-562/19 P 201