Accept Refuse

EUR-Lex Access to European Union law

This document is an excerpt from the EUR-Lex website

Document 52016SC0392

COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT Accompanying the document Proposal for a Council Directive amending Directive 2016/112/EC, as regards rates of value added tax applied to books, newspapers and periodicals

SWD/2016/0392 final - 2016/0374 (CNS)

Brussels, 1.12.2016

SWD(2016) 392 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Proposal for a Council Directive

amending Directive 2016/112/EC, as regards rates of value added tax applied to books, newspapers and periodicals

{COM(2016) 758 final}
{SWD(2016) 393 final}


Contents

1.INTRODUCTION AND POLICY CONTEXT

1.1.Aim of the valued added tax (VAT)

1.2.Rules on VAT rates

1.3.Recent and upcoming reforms affecting the VAT treatment of publications

1.3.1.Switchover from an origin-based to a destination based VAT regime

1.3.2.Why act now? The reasons for a stepwise approach to adapt the rules on rates to the destination principle

2.WHAT IS THE PROBLEM AND WHY IS IT A PROBLEM?

2.1.Overview and problem tree

2.2.Problem drivers

2.2.1.Problem driver 1: E-publications and printed publications are substitutes to some extent

2.2.2.Problem driver 2: Member States want to use VAT for social objectives, including for publications

2.2.3.Problem driver 3: Member States have different VAT regimes for publications as regards the possibility to use super-reduced or zero-rates

2.3.Problem 1: The current directive does not allow Member States to apply to e-publications rates lower than the standard rate.

2.3.1.Views expressed in Open Public Consultation

2.3.2.What are the consequences of problem 1?

2.4.Problem 2: Member States are subject to unequal treatment in their ability to use lower rates for publications

3.WHY SHOULD THE EU ACT?

4.WHAT SHOULD BE ACHIEVED?

4.1.General objective 1: Propose a reform for VAT rates applied to e-publications as announced in the VAT Action Plan

4.2.General objective 2: Prepare the ground for the definitive VAT regime allowing for more freedom in setting VAT rates for Member States

4.3.Specific objective: Allow Member States to align VAT for e-publications with the rates currently in force for printed publications

4.4.Indications on policy objectives from stakeholder consultation

5.WHAT ARE THE VARIOUS OPTIONS TO ACHIEVE THE OBJECTIVES?

5.1.Baseline scenario: continuation of current rates applied with the possibility of aligning rates at the standard rate (no EU action)

5.2.Option 1: possibility of aligning VAT rates for publications at a reduced rate

5.3.Option 2: possibility of aligning VAT rates for publications at a reduced rate and at a super-reduced or zero rate

6.ANALYSIS OF IMPACTS

6.1.Baseline scenario

6.2.Option 1: possibility of aligning VAT rates for publications at a reduced rate

6.3.Option 2: possibility of aligning VAT rates for publications at a reduced rate and at a super-reduced or zero rate

7.HOW DO THE OPTIONS COMPARE?

7.1.Objectives

7.2.Impacts

7.3.Preferred option

8.MONITORING

9.ANNEXES

9.1.Procedural information

9.2.Stakeholder consultation

9.3.Who is affected by the initiative and how?

9.4.Analytical models used in preparing the impact assessment

Tables and Figures

Table 1: VAT rates applied to printed publications as at 1 August 2016    

Table 2: VAT rates applied to e-publications as at 1 August 2016    

Table 3: VAT rate differentials between formats    

Table 4: Market share of printed vs. e-books, non-EU and EU countries, 2013    

Table 5: Expenditure by sector and source of revenue    

Table 6 Budgetary impact - Option 1    

Table 7 Budgetary impacts - Option 2    

Figure 1 Problem tree    

Figure 2 Trends in e-book market shares, sales on Amazon.com    

Figure 3: Comparison of options with objectives and impacts    


1.INTRODUCTION AND POLICY CONTEXT

1.1.Aim of the valued added tax (VAT)

VAT is a general consumption tax with the primary aim of raising revenue for Member States’ budgets so as to finance the provision of public goods and social and other policies. Following the general consumption tax model, in the EU system one so-called 'standard' VAT rate is supposed to apply to the supply of most goods and services; however, most Member States have a long tradition of using VAT also as a social policy tool by applying lower reduced rates to certain goods and services deemed to be worthy of a preferential treatment.

1.2.Rules on VAT rates

The EU legal framework concerning VAT and in particular VAT rates is laid down in Council Directive 2006/112/EC of 28 November 2006 on the common system of value added tax 1 (hereinafter the "VAT Directive").

Member States must apply on supplies of goods 2 and supplies of services 3 a single standard rate of VAT, which may not be lower than 15%;

If they so wish, Member States may apply, in lieu of the standard rate, a reduced rate of no less than 5% 4 , but only if the goods or services belong to one of 24 specific categories of supplies listed in an annex of the VAT Directive (Annex III).

Member States may individually decide to restrict the application of the reduced rates to subsets of these 24 categories; however, they are not allowed to extend the scope of reduced rates beyond these categories. This has been confirmed by settled case law of the Court of Justice of the European Union (hereinafter the "CJEU"), which clarified that the reduced rate is exceptional 5 . As a default rule, the standard rate is applicable.

The VAT Directive explicitly excludes electronically supplied services 6 , including electronically supplied books, newspapers and periodicals
('e-publications') from the scope of reduced VAT rates. This rule dates back to 2002 and was intended to facilitate tax compliance for third-country B2C ('business-to-consumer) suppliers of electronic services, by relieving them of the burden of dealing with different rates (standard/reduced) in each Member State.

The above framework was intentionally designed around simple and general rules, but ended up being considerably complicated by a multitude of derogations granted during the negotiations which preceded the adoption of the VAT Directive of 1992, as well as in the Acts of Accession to the European Union. These derogations allow specific Member States to apply reduced rates lower than 5% ("super-reduced rates") to specific goods and services, or in some cases even to exempt them and deduct the VAT paid at the preceding stage (so-called "zero rates"). These derogations are temporary, in the sense that, although they do not expire on a specific date, they will expire upon the (as yet unscheduled) entry into force of the definitive regime.

Both the derogations and the contents of the list of goods and services eligible for reduced rates are largely the result of compromise decisions taken in the Council.

1.3.Recent and upcoming reforms affecting the VAT treatment of publications

1.3.1.Switchover from an origin-based to a destination based VAT regime 

The above-mentioned rules limiting Member States' autonomy in rate setting were directly linked to the long-term objective of introducing an origin-based VAT system. An origin-based VAT system defines the point of taxation as the place where the supplier is located; implying that the VAT rate applied is the one where the supplier is established. This objective required an approximation of VAT rates between Member States, because businesses established in Member States applying lower VAT rates had a competitive advantage over businesses established in other Member States, which would have had a direct distortionary effect on the Internal Market. Given the very different level of rates applied in Member States, as a first step, minimum rate levels were introduced in order to prevent further dispersion, although with extensive temporary derogations. It was hoped that over time VAT rates would converge across Member States, but despite several decades of effort no tangible progress was achieved.

On 12 January 2008, the Council adopted Council Directive 2008/8/EC aiming at changing the rules on VAT so as to ensure that VAT on services accrues to the country where consumption occurs, and to prevent distortions of competition between Member States operating different VAT rates.

Therefore already now, the vast majority of supplies of services are based on the destination principle, which has been implemented progressively since 2010. The last milestone of the implementation – and one with direct relevance for e-publications - was the implementation of destination-based taxation for e-services in 2015. Since 1 January 2015, all B2C supplies of telecommunications, broadcasting and electronic services, which include all e-publications, are taxable at the place where the customer resides and no longer where the supplier is located. Any Commission proposal allowing for reduced VAT rates on e-publications before 2015 could have created distortions of competition in the single market. At least it would have put pressure on Member States to cut their rates to match competitors, constituting a rationale for keeping the standard rate treatment.

Today also the vast majority of supplies of goods are taxed de facto where the customer resides, because the place of supply should be the place where the goods are located at the time when the supply takes place. However, in case of cross-border transactions (distance sales below a certain threshold, which lies between 35.000 euro and 100.000 euro depending on the Member State, and cross-border shopping) VAT is still charged according to the origin principle.

In the Action Plan on VAT 7 published on 7 April 2016 the Commission announced that it would present a legislative proposal to put in place a definitive VAT system 8 . This definitive system will rest upon the agreement of EU legislators that the VAT system should be based on the principle of taxation in the country of destination of the goods. This means that the taxation rules according to which the supplier of goods collects VAT from his customer will be extended to cross-border transactions. Together with the destination based taxation of services, the definitive regime will be fully based on the destination principle.

1.3.2.Why act now? The reasons for a stepwise approach to adapt the rules on rates to the destination principle

In the Action Plan on VAT the Commission, in line with the logic of the destination principle that removes the rationale for an approximation of VAT rates, proposed granting Member States greater freedom on rate-setting than what is allowed by current legislation. This aim should be achieved in two steps. The Commission intends to present a narrow proposal with regard to VAT rates for electronically supplied publications by the end of 2016 and in 2017 a general proposal on VAT rate setting linked to the proposal for the definitive VAT regime.

In this sense, the proposal to enable Member States to apply lower rates for electronically supplied publications is a first step in eliminating restrictions for Member States before a general proposal on VAT rate setting in 2017. While it could be argued that it would be better to deal with e-publications in the context of this general proposal, various reasons militate in favour of putting forward a targeted proposal on e-publications at this point in time.

The first reason is that the initiative for e-publications is tightly linked to the Commission's legislative proposal to modernise and simplify the VAT for cross-border e-commerce adopted at the same time as this proposal as a part of its Digital Single Market strategy (DSM), for which proposals are to be unveiled on 30 November 2016. The e-commerce proposal aims in particular at implementing an enhanced one-stop shop (OSS) to facilitate cross-border trade. The OSS currently in place for electronically supplied services should in future also cover tangible goods traded cross-border. Solving the rates issue is essential to the success of the DSM strategy that aims to take account of the realities of a digital economy and adapt current legislation accordingly. Leaving it out would disregard the objective of the DSM strategy to make businesses and EU legislation fit for the digital economy.

The second reason is that there is a strong and urgent demand from Member States for an initiative on e-publications. Already in 2015, in a letter to the Commission, France, Germany, Poland and Italy demanded an amendment to the VAT directive allowing the option of applying a reduced rate not only to printed books, but as well to electronically supplied books. This request was shortly thereafter reiterated by the Council: in the Council conclusions on the 2016 Action Plan on VAT 9 , Member States welcomed "the intention of the Commission to present a proposal for increased flexibility for Member States, so that they could benefit from the existing reduced and zero rates in other Member States" and invited "the Commission to present a legislative proposal which integrates provisions concerning VAT rates for e-publications in the context of Digital Single Market initiatives by the end of 2016 […] Thus the Council conclusions represent a clear demand to enable Member States with the possibility to apply reduced, super-reduced and zero VAT rates for electronically supplied publications.

Furthermore, three Member States (Belgium, France and Italy) have unilaterally introduced super-reduced or zero rates for certain e-publications in advance of any proposal and only Belgium has returned to a standard rate subsequent to a request by the Commission.

Given the support and explicit request for this initiative on the part of Member States and the clear link with the DSM, it would appear artificial to postpone addressing the specific issue of e-publications to the presentation of the general proposal on rates, as the discussions on the general proposal in the Council are likely to last for several years, and a delay would be particularly harmful in the fast-moving setting of digital technology. The Commission has therefore taken the political decision to bring forward this specific proposal. There is little risk of introducing incoherence in the system, because the general rates proposal will go in the same direction as the proposal on rates, i.e. to grant Member States a greater room for manoeuvre in setting VAT rates. On the contrary, based on experience, discussing this narrow proposal in the Council is likely to provide useful lessons for the general proposal and to prepare the ground for it.

2.WHAT IS THE PROBLEM AND WHY IS IT A PROBLEM?

2.1.Overview and problem tree

This Impact Assessment argues that the main problems affecting the taxation of e-publications are two:

1. The current Directive does not allow Member States to apply to e-publications rates lower than the standard rate;

2. Member States are subject to unequal treatment in their ability to use lower rates for publications

The main drivers or causes for the above problems have been identified as follows:

1.the fact that e-publications are to some extent substitutes for print;

2.the fact that Member States want to use VAT for social objectives, including to decrease the price of printed materials;

3.and finally, that Member States have different VAT regimes for publications as regards the possibility to use super-reduced or zero-rates.

The most important consequences of these problems are identified as a cost disadvantage for digital publishing and the fact that those Member States that do not benefit from derogations are prevented from further lowering the tax rate on publications. Technological change, which has been leading to a growing diffusion of e-publications, and the switchover to the destination principle also influence the context. This causal chain is illustrated in Figure 1 below.

Figure 1 Problem tree

   

2.2.Problem drivers

2.2.1.Problem driver 1: E-publications and printed publications are substitutes to some extent

As highlighted by the results of the open public consultation, a strong current of stakeholders' opinion holds that the differences among the formats do not justify a higher VAT treatment for e-publications, apparently on the basis of the fact that the content is the most essential and defining characteristic of publications, and does not vary with the format.

Whereas a study 10 for the European Commission found that that printed and digital publications have different characteristics, there is nevertheless some degree of substitutability between traditional and digital publications.

This degree of substitutability constitutes for many Member States sufficient reason to request the alignment of the tax treatment of the digital and the printed format.

2.2.2.Problem driver 2: Member States want to use VAT for social objectives, including for publications

VAT is a general consumption tax with the primary aim of raising revenue for Member States’ budgets so as to finance the provision of public goods and social and other policies.

Under the general consumption tax model, a single 'standard' VAT rate is supposed to apply to the supply of most goods and services. In practice, however, most EU Member States have a long tradition of using VAT as a social policy tool by applying lower reduced rates to certain goods and services deemed to be worthy of a preferential treatment. This is apparent in the large number of categories for which Annex III allows the application of reduced rates (24) and in their extensive scope, despite a consensus on the part of economists that reduced VAT rates are a rather costly instrument.

The extent of the support for the application of reduced rates to publications is further confirmed by the fact that, although the derogations granting Member States added leeway on reduced, super-reduced and zero rates were explicitly meant to be temporary, practically none of them has ever been phased out.

While Member States usually apply a variety of rates to printed publications, depending on their nature (see table 1; only seven Member States apply a single rate), the reduced rate clearly dominates; only Bulgaria does not make use of the reduced rate option at all and applies the standard rate to all printed publications.

In the domain of e-publications, the multiple Member States initiatives to extend to them reduced rate treatment clearly shows that Member States intend to continue using VAT for social objectives. 11

2.2.3.Problem driver 3: Member States have different VAT regimes for publications as regards the possibility to use super-reduced or zero-rates

With regard to VAT rates, the VAT Directive in force treats differently both traditional printed publications and e-publications, and Member States.

As regards the distinction for the format, under the VAT Directive, printed publications are eligible for a reduced VAT rate, whereas e-publications must always be taxed at the standard VAT rate.

As regards however the second distinction Member States are not equally placed in the possibility to apply super-reduced or zero rates. Owing to the operation of derogations that were introduced as temporary measures but did not have an explicit expiry date, France, Italy, Spain and Luxembourg are allowed to apply rates lower than the 5% minimum for reduced rates ('super-reduced rates') while Belgium, Denmark, Ireland, Sweden and the United Kingdom even grant exemptions with a right to deduct VAT occurred at the preceding stage (so-called 'zero rate').

The rules on rates set in the VAT Directive have never been updated to reflect the adoption of the destination principle instead of the origin principle. This may be regarded as a serious flaw because, according to art. 113 of the Treaty, the VAT Directive should harmonise legislations (only) to the extent that such harmonisation is necessary to ensure the establishment and the functioning of the internal market and to avoid distortion of competition. Furthermore, as highlighted by optimal tax theory, the application of the destination principle leads to undistorted producer prices and creates international production efficiency. Restricting Member States in setting VAT rates was necessary to prepare an origin-based VAT system, but can no longer be justified in a VAT regime operating under the principle of taxation at the destination. Thus, there is no longer a justification why some Member States do not have the same freedom in setting VAT rates as others.

Destination-based taxation has been established for e-services including e-publications since 2015 and with regard to printed books, newspapers and periodicals, for which all Member States were always allowed to apply reduced rates (and for some even super-reduced and zero rates), differences in rate between Member States did not lead to any distortion of competition or hampered the functioning of the internal market. This is, because the destination principle applies to domestic supplies and the origin principle only applies to cross-border sales. Final consumers have little or no incentive to cross the border to purchase publications in order to benefit from lower VAT rates: there are language barriers, printed publications are of low value and fixed prices exist for most publications.

2.3.Problem 1: The current directive does not allow Member States to apply to e-publications rates lower than the standard rate.

The text of the VAT Directive in force treats differently traditional printed and e-publications with regard to VAT rates.

Under the VAT Directive, printed publications are eligible for a reduced VAT rate, whereas e-publications have to be taxed at the standard rate.

In addition, 9 Member States can apply super-reduced and zero rates to physical publications, whereas 19 Member States cannot do so.

Table 1: VAT rates applied to printed publications as at 1 August 2016    

Category

BE

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

CY

LV

Printed books

6

20

10

25

7

9

0

6

4

5.5

5

4

5

12

21

15

23

21

20

22

Books on other physical means of support

21

20

21

25

7

20

23

23

4

5.5

5

4

19

21

19

20

22

Printed newspapers

0

20

15

0

7

9

9

6

4

2.1

5

4

5

12

6

25

23

21

20

13

21

25

Printed periodicals

0

20

15

25

7

9

9

6

4

2.1

5

4

5

12

6

23

21

20

13

22

21

25

Number of rates

3

1

3

2

2

2

3

2

2

3

3

2

2

2

Category

LT

LU

HU

MT

NL

AT

PL

PT

RO

SI

SK

FI

SE

UK

Printed books

9

3

5

5

6

10

5

6

5

9.5

10

10

6

0

23

23

Books on other physical means of support

21

3

5

5

6

20

5

6

5

9.5

10

24

6

0

23

23

25

20

Printed newspapers

9

3

5

5

6

10

8

6

5

9.5

20

10

6

0

23

23

24

Printed periodicals

9

3

5

5

6

10

5

6

5

9.5

20

10

[ex]

0

23

23

24

6

8

Number of rates

2

1

1

1

1

2

3

2

1

1

2

3

3

2

   Source: Commission Services

In contrast, under the current text of the Directive, electronically supplied books, newspapers and periodicals must be taxed at the standard VAT rate. For VAT purposes, they are considered as electronically supplied services, to which the standard rate applies. Such rates are in practice markedly higher than the rates applied to printed publications (see table 2).

This rule is binding and cannot be adapted or interpreted. The Court of Justice of the European Union (CJEU) confirmed in its C-479/13 and C-502/13 judgements of 5 March 2015 that France and Luxembourg are not allowed to apply a reduced rate of VAT to the supply of electronic books. The Court found that the VAT Directive excludes any possibility of a reduced VAT rate being applied to ‘electronically supplied services' 12 .

Following the CJEU judgment, Luxembourg has aligned its VAT treatment of e-publications to the standard rate, whereas France still applies a reduced rate of 5.5% to e-books and since 2014 a super-reduced rate of 2.1% to the digital press. After extending the existing zero rates for paper newspapers and periodicals to the digital press in 2012, Belgium has, since 1 January 2016, reverted to applying the national standard rate to digital newspapers and periodicals. On that same date, however, Italy extended the application of its super-reduced rate of 4% to digital newspapers and periodicals, which was already applied to e-books since 1 January 2015.

Table 2: VAT rates applied to e-publications as at 1 August 2016

Category

BE

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

CY

LV

e-books

21

20

21

25

19

20

23

23

21

5.5

25

4

19

21

e-newspapers / e-periodicals

21

20

21

25

19

20

23

23

21

2.1

25

4

19

21

Category

LT

LU

HU

MT

NL

AT

PL

PT

RO

SI

SK

FI

SE

UK

e-books

21

17

27

18

21

20

23

23

20

22

20

24

25

20

e-newspapers / e-periodicals

21

17

27

18

21

20

23

23

20

22

20

24

25

20

   Source: Commission Services

To summarise, Member States are not allowed to lower VAT rates on e-publications and they cannot in particular align rates for e-publications with the rates currently in force for printed publications.

2.3.1.Views expressed in Open Public Consultation

The lack of equal treatment between physical and electronically supplied publications being substitutes was described as a problem by a large majority of respondents who provided further explanations. Most of these respondents advocating an equal treatment draw a bigger picture and argued that Member States are unjustifiably restricted to lower VAT rates only for printed publications, with negative effects for digital learning, knowledge transfer and promoting culture in a digital economy. Member States have expressed the same view in public statements and four of them also in the consultation. The fifth Member States participating in the consultation supported the principle of extending greater flexibility to Member States on VAT rates and the ability of Member States to make a decision to apply a non-standard rate of VAT to e-publications, but did not accept that e-publications and physical books are equivalent and so should automatically be subject to the same rates. Finally, less than 1% of the respondents perceived the existing reduced rate for printed publications as a problem.

2.3.2.What are the consequences of problem 1?

The immediate consequences are higher VAT rates on e-publications compared to printed publications.

Table 3 shows the tax rate differential on e-publications. While its exact value depends on each Member States choices on rate levels and on the specific type of publication, the tax disadvantage is typically quite sizeable (10-20 percentage points) and can reach up to 25 percentage points.

Table 3: VAT rate differentials between formats

VAT rates 1/8/2016

BE

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

CY

LV

printed books

6

20

10

25

7

9

0

6

4

5.5

5

4

5

12

e-books

21

20

21

25

19

20

23

23

21

5.5

25

4

19

21

VAT rate differential

(percentage points)

15

0

11

0

12

11

23

17

17

0

20

0

14

9

 

BE

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

CY

LV

printed newspapers

0

20

15

0

7

9

9

6

4

2.1

5

4

5

12

e-newspapers

21

20

21

25

19

20

23

23

21

2.1

25

4

19

21

VAT rate differential

(percentage points)

21

0

6

25

12

11

14

17

17

0

20

0

14

9

VAT rates 1/8/2016

LT

LU

HU

MT

NL

AT

PL

PT

RO

SI

SK

FI

SE

UK

printed books

9

3

5

5

6

10

5

6

5

9.5

10

10

6

0

e-books

21

17

27

18

21

20

23

23

20

22

20

24

25

20

VAT rate differential

(percentage points)

12

14

22

13

15

10

18

17

15

12.5

10

14

19

20

 

LT

LU

HU

MT

NL

AT

PL

PT

RO

SI

SK

FI

SE

UK

printed newspapers

9

3

5

5

6

10

8

6

5

9.5

20

10

6

0

e-newspapers

21

17

27

18

21

20

23

23

20

22

20

24

25

20

VAT rate differential

(percentage points)

12

14

22

13

15

10

15

17

15

12.5

0

14

19

20

   Source: Commission Services

The current VAT treatment of e-publications is likely to have effects both on the supply and on the demand for e-publications.

Digitalisation alters the cost structure of the production process. For e-books (compared to print books), a large share of variable costs disappear (e.g. printing, physical logistics for distribution and returns, storage, return). Some costs are partly shifted to other sectors (e.g. promotion, which increasingly occurs through blogs and social media), while new costs appear (file conversion, cataloguing, security etc., and costs related to distributors). There are also costs that remain unchanged (e.g. authors). Overall, this results in lower costs of producing e-books, particularly marginal costs, while the issue is more complex for the overheads 13 .

The cost impact of a VAT differential of an order of magnitude such as listed in Table 3 is large enough to have a perceptible impact on publishers. While no statistics exist and costs vary considerably from one publication to another, an example can put the problem in perspective and give an idea of the usual order of magnitude of these costs. A 2010 report put typical publisher's production costs in the US at close to 9 dollars for a hardcover book priced at 26 dollars, and at 4.53 dollars for an e-book priced at 12.99 dollars 14 . In such an example, the differential cost factor for the publisher, introduced by VAT could reach up to 73% of e-book publishing costs (excluding overheads). While the actual impact will certainly vary from case to case, it is generally true that very few industries would be unaffected by a cost factor in the order of 10% of final prices. The marginal cost nature of VAT also speaks for an immediate impact on the market 15 .

For digital newspapers (as compared to print newspapers), editorial costs are generally higher as more content is added (extra photos or videos) and content is updated more frequently. Printing and delivery costs are no longer needed but there are additional costs, mainly technology associated costs from digital formatting and editing, development of apps and pay-walls and hosting. Another new and significant cost is the distribution fee paid to online retailers. Overall, the share of variable costs in the production of digital newspapers is significantly lower than for print newspapers, and unit costs can only decline in the future as circulation and turnover increases.

Digital periodicals (compared to print periodicals) show similar changes to that of digital newspapers. Staff costs are higher while paper, printing, binding and wholesale charges disappear and additional editorial and technological production costs and distribution fees are added. Overall, the industry has unit costs of production similar or slightly higher for digital products but, similarly to newspapers, such costs are likely to decline in the future as circulation and turnover increases.

In this context, to assess the consequences on the market, it needs to be discussed whether publishers affected by the tax would have the possibility of simply shifting it over to the consumer by increasing prices without suffering meaningful losses in terms of sales or profits (known in economics as the issue of 'tax incidence' or who effectively pays for the text). According to established economic theory, the market impact depends primarily on two variables, the elasticity of demand and supply.

In non-technical terms, the first variable measures the price sensitivity of consumers (if consumers are very price-sensitive, VAT will tend to be readily translated into price changes) while the second measures the reactivity of suppliers to sale prices, which is usually linked to the characteristics of the supply chain (a highly competitive market will tend to transfer quickly and fully any change in the VAT rate, in both directions, while the opposite tends to happen in a monopolistic market).

Consumers' price sensitivity

The sensitivity of consumers to increases in the price of publications, a.k.a. demand elasticity, has been investigated thoroughly in a study carried out for the Commission 16 . In terms of the own-price elasticity metric 17 , it was found to fall within the -3.2 to -4.5 range for both print and digital. This measure indicates a high sensitivity to price; the price elasticities (i.e. the market sensitivity to price) were found to vary somewhat by market segment, being higher for books than for periodicals and newspapers. These estimates are in line or slightly above previous estimates in the economic literature.

Characteristics of the supply chain

As concerns the effect on the supply side (elasticity of supply), the study found a more mixed picture, as some parts of the supply chain are competitive and some show a trend towards concentration, which would speak for an incomplete pass-on of the tax. Costs are not the only factor in price-setting, particularly in oligopolistic market segments where publishers can follow a strategy of pricing above marginal costs. 18 . An incomplete pass-on rate of, say, 50% would imply that roughly half of the tax leads to higher prices for e-publications and the remaining half to lower profit margins. Given also high demand elasticities, this leads to lower sales volumes 19 .

Overall, the study concluded that a pass-on rate of about one half is a reasonable estimate. This implies that higher VAT rates contribute, though not fully, to higher prices for e-publications, and vice versa. It also found that there could be some asymmetry in the reaction of the market to tax raises or cuts, but found evidence for this only in a specific market segment. The incomplete pass-on means that a higher VAT rate translates partly into higher prices for consumers and partly into a lower profit margin for publishers.

In addition, the value represented by the reduced rate VAT treatment not available to e-publications is quite substantial: it amounts to one billion euro annually in the EU-28. In case all Member States could align the rates for all publications at the lower level currently in place for printed publications, these VAT subsidies could increase to up to 3.7 billion euro annually in the longer term. 20 Of course, one could question whether all Member States would necessarily grant e-publications the same treatment as for printed publications; however, the breadth of Member States' support for an initiative to cut rates on e-publications and the fact that several of them introduced reduced and super-reduced rates even in contravention of the directive, suggests that many Member States would indeed introduce reduced rates, if they were given the legal possibility to do so.

Finally, although the VAT treatment is only one aspect and its impact cannot be isolated from that of many other factors influencing the development of e-publishing, it can be observed that the market for e-publications in the EU is less developed than in the US or in other advanced non-EU countries (see Table 4). Even in the most developed EU market, the UK, the market share of e-books is lower and for many Member States the market share is well below 5%.

Table 4: Market share of printed vs. e-books, non-EU and EU countries, 2013

Country

Printed books

E-books

US

79.5

20.5

Canada

Japan

83

85

17

15

UK

    86

    14

Germany

Austria

Spain

Italy

Slovenia

95

95

96

96.2

99

5

5

4

3.8

1

Note:    Figures are approximate owing to a series of methodological difficulties.

Sources    Global eBook 2015, A report on market trends and developments, Rüdiger Wischenbart et al,
Deloitte, 2015,
Print is alive and well – at least for books.
Available at: http://www2.deloitte.com/content/dam/Deloitte/global/Documents/Technology-Media-Telecommunications/gx-tmt-pred15-print-alive-and-well.pdf

2.4.Problem 2: Member States are subject to unequal treatment in their ability to use lower rates for publications 

The derogations to the VAT directive have the effect that Member States are not equal in their ability to use lower rates for publications. Several Member States have obtained the possibility to apply super-reduced and zero rates to publications, under provisions that were supposed to be temporary but were de facto never abolished. The several Member States who joined the EU from 2004 onwards were not in a position to negotiate such derogations. As a result, there is a substantial dichotomy in the room for manoeuvre of Member States on VAT rates for publications.

What are the consequences of problem 2?

Although these newer Member States have the ability to use the reduced rate, these differences in rates nevertheless have the effect of considerably constraining their room for manoeuvre. This is because Member States can only introduce up to two reduced rates which together have to cover all eligible goods and services. Thus, for these Member States extending further the support for publications would generally necessitate cutting VAT on a whole range of other supplies as well, with the inherent substantial budgetary cost, or restructure completely the allocation of goods and services between the two rates, which is politically extremely sensitive.

The application of the destination principle prevents a direct distortion to the internal market, but there is a fairness problem, because there is no longer an objective justification for this difference in treatment. Although it is hard to document the consequences of a perceived unfairness in the system, the fairness issue pops up frequently in informal discussions between the Commission and Member States and with regard to the application of reduced VAT rates it has resulted in various infringement procedures against Member States that wanted to extent the scope of reduced rates to products that could benefits from a reduced rate in one or more Member States based on a derogation, but that were included in Annex III of the VAT Directive. Hence, addressing the fairness problem satisfactorily is an important prerequisite to ensure the success of any legislative proposal in the field of publications and to prepare the ground for a more general reform of rates, as highlighted by the Council conclusions stating that Member States should benefit from the existing reduced and zero rates in other Member States 21 .

3.WHY SHOULD THE EU ACT?

Under the subsidiarity principle, the EU should act only when Member States (or even sub-national entities) cannot effectively address the problem.

This is here the case, because Member States are obliged to apply the VAT Directive correctly and the VAT Directive prevents Member States from granting electronically supplied publications the same VAT rates that currently apply to printed publications. This was confirmed by the CJEU in its C-479/13 and C-502/13 judgements of 5 March 2015. The Court has clearly stated that Member States cannot apply a reduced rate to e-books under the current VAT Directive and therefore, a mere interpretation of the VAT Directive, or soft-law initiatives, cannot remedy the situation.

Thus, an EU-level legislative initiative to amend the VAT Directive is indispensable in order to tackle the problem.

The proposal constitutes an early application of the envisaged definitive VAT regime for the publication market in advance of a general VAT rate proposal in 2017 without any risk of distortion in the internal market.

Such an advanced targeted proposal is a political choice giving the particular demand by Member States and the uncertainty when a general VAT rate proposal would be adopted. As such a general proposal can only be adopted by the Council once the definitive VAT regime has been agreed on, it can take several years until the envisaged general VAT rate reform enters into force. The Commission is already now confronted with Member States that unilaterally apply lower rates than what is allowed under the current VAT Directive and strong Member States' requests that require a timely solution. It is also justified to advance on this political priority area linked to the DSM strategy and to update the VAT Directive recognising technological progress that continues and also justifies requires a timely reaction by the Commission.

Moreover, the outcome in the Council negotiations on this proposal will test the waters ahead of a general VAT rate proposal providing indications on the degree of subsidiarity supported by the Council and as well about how the Member States will in practise deal with increased power in VAT rate setting.

4.WHAT SHOULD BE ACHIEVED?

4.1.General objective 1: Propose a reform for VAT rates applied to e-publications as announced in the VAT Action Plan

In the Action Plan on VAT the Commission committed to present a proposal on VAT rates applied to e-books and electronic newspapers in 2016, which cannot benefit from reduced rates available for physical publications and to address this issue in the context of the Digital Single Market strategy.

4.2.General objective 2: Prepare the ground for the definitive VAT regime allowing for more freedom in setting VAT rates for Member States

The definitive VAT regime based on the destination principle will allow for greater diversity in VAT rates and will abolish unjustified restrictions on Member States in setting VAT rates. This general objective is linked to the proposal on VAT rates to be presented in 2017, which will include the possibility for all Member States of applying super-reduced and zero rates that are currently in place as derogations only for some Member States.

4.3.Specific objective: Allow Member States to align VAT for e-publications with the rates currently in force for printed publications

This specific objective prepares for the definitive VAT regime in an area where the destination principle has already been implemented. Member States have a long tradition of applying lower VAT rates to books, newspapers and periodical, but are forced by current rules to restrict these lower rates to physical publications. Member States should be allowed to apply the same VAT rates, currently in force for physical publications, to e-publications.

The specific objective aims to achieve a timely solution on which all Member States can agree. Past experience has shown that a stand-alone proposal should not be subject to years of discussion and negotiations in the Council and could enter into force in 2017, whereas it could take several years to agree on a general reform of rules for VAT rates for the definitive VAT regime.

4.4.Indications on policy objectives from stakeholder consultation

The above objectives tally well with the indications expressed by stakeholders. In particular, nearly all respondents to the open public consultations pointed out the need to allow Member States to align VAT rates for publications reading and elaborated on this point by highlighting a variety of societal aims of publications, e.g. education, science, culture and democratic society. The main theme emerging from the open public consultations was that lower VAT rates for e-publications will benefit individuals and society and that the social, economic and cultural dimensions of reading are closely connected. Some of the respondents added that Member States should therefore be allowed to further promote all audio-visual or digital works, such as educational games. However, the large majority of respondents advocated a proposal targeted at publications.

5.WHAT ARE THE VARIOUS OPTIONS TO ACHIEVE THE OBJECTIVES?

5.1.Baseline scenario: continuation of current rates applied with the possibility of aligning rates at the standard rate (no EU action)

The current VAT Directive does allow for aligning the VAT rates of printed and electronically supplied publications – but only at the standard VAT rate. However, all Member States (except Bulgaria) have a long tradition of granting reduced or even lower rates to printed books, newspapers and periodicals. It is completely unrealistic that Member States will abolish the long-existing reduced rate.

Accordingly, the baseline scenario takes into account the VAT rates presently applied and assumes that Member States will continue to tax printed publications at reduced rates. It also assumes that Member States would be enabled by the general VAT rate proposal in 2017 to apply reduced, super-reduced and zero VAT rates to electronically supplied publications. However, such a general VAT rate proposal would be subject to legislative negotiations and the entry into force of a definitive VAT regime and it is therefore assumed that it exceeds the time horizon of this Impact Assessment of 2021.

Furthermore, the detailed arrangements of the future definitive VAT, the implementation of the extension of the OSS to simplify e-commerce of physical publications, would not have an impact on the publication market, because they exceed the time horizon of this impact assessment. The general taxation rules as described in chapter 1 have already been adapted to the destination based principle and would remain in the definitive VAT regime.

Consequently, the base line scenario takes only into account the VAT rates presently applied and the current need for France and Italy to apply a standard rate to all electronically supplied publications.

5.2.Option 1: possibility of aligning VAT rates for publications at a reduced rate 

This option would allow all Member States the possibility to apply a reduced rate to all publications including e-publications. This increased flexibility would however not oblige Member States to lower the VAT rates for e-publications. The existing super-reduced rates and zero rates for certain printed publications in some Member States would remain.

Technically, this option consists in adapting the provisions of Annex III in the VAT Directive by eliminating the references to "all physical means of support" and to specific formats for printed publications (brochures, leaflets and similar printed matter, children’s picture, drawing or colouring books, music printed or in manuscript form, maps and hydrographic or similar charts). This would require a legislative proposal to amend the VAT Directive.

Striking out the references to format and referring in general to books, newspapers and periodicals is a requirement to cover as well e-publications and Member States would be able to continue to restrict the application of reduced rates to certain books, newspapers and periodicals, e.g. by excluding specific formats or content, as they already do it now with regard to printed publications 22 . This solution would also allow Member States to continue to apply a reduced rate for audio books, audio newspapers and periodicals for people with sight loss.

The supply of pure music and video content would continue to be taxed at the standard VAT rate, as would publications that predominantly consist of music and video content. Member States would then have the option of introducing specific provisions in their national VAT law in order to determine how to apply the general provision in the Directive. The current VAT rules already contain similar general provisions with regard to advertising; publications predominantly devoted to advertising cannot benefit from a reduced rate, and Member States have the discretion to specify the term "predominantly" in their national VAT law.

The option does not put forward any EU level definition of what is a "book", a "newspaper" and "periodical". This choice is based on the following considerations, which respect in particular the subsidiarity principle:

1. The choice is in line with Annex III of the VAT Directive, which only lists the categories of goods or services eligible for reduced VAT rates without providing definitions, whereby Member States are free to restrict the scope of each category.

2. A specific definition of printed publications has never existed at EU level, but some Member States have introduced national definitions restricting the scope of Annex III. Introducing specific definitions at EU level would on the one hand restrict those Member States that currently have not introduced specific definition of a book, newspaper or periodical in their national law and might on the other hand oblige some Member States to change their existing national definitions. In general, national specifications may more accurately reflect cultural and market differences between Member States.

3. E-publications are evolving and any specific definition of what is a book, newspaper or periodical risks being outdated within a short time. Given the unanimity requirement for EU legislation in tax matters, Member States are generally able to adapt the rules to future needs in a more timely fashion than the EU could.

In the future, in case a Member State would like to discuss a common EU wide interpretation of the terms book, newspaper and periodical, such a discussion could always take place in the VAT Committee, which could also issue guidelines in this respect.

Finally, the application of the destination principle ensures that there is no risk for the functioning of the internal market in case Member States apply different definitions of electronically supplied books, newspapers and periodicals including audio and video content.

5.3.Option 2: possibility of aligning VAT rates for publications at a reduced rate and at a super-reduced or zero rate

Option 2 is an extended version of option 1; besides the freedom to apply a reduced rate to e-publications, Member States would also be able to apply super-reduced or zero rates to all publications.

This extension to option 1 is necessary in order to enable all 28 Member States to align the VAT rates for e-publications with the rates they currently apply to printed publications and not only 19 as under option 1.

The possibility to allow for super-reduced and zero rates would be incorporated as a general provision for all Member States. Introducing a general provision only to specific Member States in the VAT Directive, e.g. to those can currently derogate from the general provisions for printed publications, is legally not allowed. Hence, the only possibility to allow those Member States applying a super-reduced or zero rate to printed publications to align the VAT rates for electronically supplied publications with these rates is a general provisions treating all Member States equally.

It should be noted that the extensions of option 2 compared to option 1 is fully in line with those two options for a general VAT rates reform outlined in the VAT Action Plan. Both reform options 23 in the Action Plan include the extension of derogations (such as the zero rates and the super-reduced rates on publications) to all Member States.

6.ANALYSIS OF IMPACTS

The assessment has been carried out against the impacts on

individual Member States’ budgets (VAT rates applied in the Member States have no impact on own resources calculations and therefore the EU budget);

consumers;

the publishing industry (including impact on SMEs);

the printing industry and retail sector;

the environment;

the simplicity of the VAT system (with impacts on public administrations, third countries and international trade).

The above listed impacts cover the main economic, social and environmental impacts and have been assessed against the present VAT system and the predicted market developments as defined under the baseline scenario.

6.1.Baseline scenario

Expenditure on printed books has declined steadily since 2008 and the overall downward trend is expected to continue in the coming years (Table 4). Similar trends are observed in newspapers and periodicals. Demand in both market segments contracted sharply from 2008 to 2009 and decreased gradually thereafter, albeit with a flatter downward trend for periodicals.

E-Books

E-books currently represent a small share of the total European book industry. In 2010, e-books had a market share in turnover of around 2%. In 2012, the share was close to 5% due to rapid growth between 2010 and 2012. According to the Federation of European Publishers, growth has slowed down in 2013 and 2014 and the market share of e-books remained at around 5%.

Overall, the increase of market share for e-books slowed down significantly in 2014, or even stalled, as was the case in France and Spain. In most markets, genre fiction as well as self-published titles is the main terrain for e-books. However, in some countries, notably in Germany and the Netherlands, e-books have become popular also for front-list bestsellers of general fiction 24 .

Until 2017 a further increase in the market share of 1 to 2 percentage points is expected.

Table 5: Expenditure by sector and source of revenue

Note: excludes BG, LV, LT, EE, SK, MT, CY and LU. Source: PwC Global Entertainment and Media Outlook: 2013-2017, www.pwc.com/outlook

Newspapers 25

Online news displayed on websites and digital newspapers have become a more important part of the European newspaper industry during the past 10-15 years, as the sector tries to adapt to the strong impact from the shift from paper-based to online reading, which has undermined the traditional business model (print circulation in Europe fell by 20% in the five years to 2014 26 ). Newspapers are reorienting their business strategy towards the digital sector: In 2011 and 2012 the turnover share of digital newspapers (without digital advertising revenues) of total market turnover was estimated to be 0.5% and 1.2% respectively, whereas the share is expected to increase to 5% in 2017. If digital advertising revenues are taken into account, the share is estimated to be approximately 11% in 2017.

Periodicals 27

Digital periodicals currently represent a small part of the total market. In the EU market, the turnover share of digital periodicals (without digital advertising turnover) was estimated to account for 0.4% of the total market turnover in 2011 and 1.35% in 2012. There are expectations of further growth to approximately 8% in 2017. If digital advertising revenues are included, the digital market share was about 5% in 2012 and is expected to be more than 15% in 2017. Therefore, the digital market is growing for periodicals and the market share for digital periodicals is expected to be larger than that of digital newspapers in the future.

Budgetary impact on Member States

Existing VAT reduced rates for the purchase of printed books, newspapers and periodicals would remain in place. The total VAT revenue for all books, newspapers and periodicals is expected to remain constant, because the traditional way of disseminating publications is expected to continue on its slow downward trend, whereas a growth of the market for electronically supplied publications taxed at the standard rate is anticipated. Whereas the price of new e-publications would be lower compared to the printed publications that disappear and therefore the VAT revenue, this would be compensated for by higher VAT rate applied to e-publications.

The situation is different in France and Italy, where the standard VAT rate would need to be implemented for all e-publications. Currently France and Italy apply a super-reduced rate to e-newspaper and e-periodicals. And with regard to e-books, Italy currently applies a super-reduced rate and France a reduced rate. The increase in VAT revenue for France and Italy could amount to around 1 billion euro annually in the medium term.

Impact on consumers

There would be no impact on the prices for printed publications and electronic publications and hence no impact for consumers except in France and Italy where the rate is assumed to be raised to comply with the VAT Directive. The gradual migration from print to digital would continue.

Impact on the publishing industry

The market for printed publications should continue to decrease gradually and the market for electronically supplied publications should continue to grow gradually, resulting in a further slight increase in the market share of electronically supplied publications.

Impact on the printing and retail sector

Printed publications would continue to be taxed at lower rates compared to electronically supplied publications and no impact is assumed on the printing industry that generates turnover from printing not only on paper, but also on plastics and textiles apart from the fact that printed publications would continue to decrease gradually.

No trend change is assumed in the baseline scenario with regard to brick-and-mortar bookstores and newsstands. These businesses must adapt their business models to the ongoing challenge from online sales, as highlighted by the significant decrease in the number of bookstore and newsstands during the last decade, for example by expanding their offer of other products and services, and this adaptation process is projected to continue. 28  

Environmental impact

No significant environmental impact is assumed in the baseline scenario. Migration from printed publications to electronically supplied publications will continue, but less than under the outlined options, where a positive environmental impact is expected.

Impact on the simplicity of the VAT system

The current VAT rules would continue to apply and neither the cost of VAT compliance (for businesses) nor the cost of VAT collection (for all Member States except France and Italy) would change.

6.2.Option 1: possibility of aligning VAT rates for publications at a reduced rate

Option 1 would allow all Member States to apply a reduced VAT rate to
e-publications. While it cannot be anticipated with any certainty whether Member States would lower VAT rates for e-publications, the impact assessment is carried out under the technical
assumption that Member States would align the VAT rates for electronically supplied publications with the VAT rates currently in place for printed publications as far as they could:

18 Member States would align the VAT rates for e-publications and printed publications at the reduced rate

Bulgaria would keep the standard rate for all publications.

The other 9 Member States are not able to align VAT rates, because they apply super-reduced and zero rates to certain printed publications. All of these 9 Member States would keep the super-reduced and zero rates for certain printed publications and 8 of them would apply a reduced rate to e-publications to reduce the VAT rate differential between formats. Denmark, which applies a zero rate to printed newspapers, has no reduced rates and would therefore continue to apply the standard rate to e-newspapers publication.

It should be stressed that by construction, this is a maximalist assumption, as it is likely that not all Member States will choose to lower the rates on e-publications. The impact of the measure would in this case be smaller. From the public consultation it could e.g. be presumed that there is currently no intention in the United Kingdom to lower VAT rates for certain e-publications.

Stakeholder consultation

A 94% share of respondents of the open public consultation agreed that Member States should be allowed to apply a reduced VAT rate to e-books and 88% of the respondents agreed that Member States should be allowed to apply a reduced VAT rate to e-newspapers and e-periodicals, too. The shares were higher among readers, researchers and libraries (98%) compared to the printing industry (73%) and the publishing industry (83%).

Budgetary impact on Member States

VAT revenue on electronically supplied publications would decrease in all Member States except Bulgaria. Given the present market share of e-publications, the decrease in VAT revenue is limited, although it would gradually increase as the market share of electronically supplied publications grows over time.

A maximum immediate decrease of VAT revenues of 5% can be expected based on the fact that the market share of e-publications is less than 5% in all Member States except the UK, but the decrease could amount up to 15% in some Member States in the long term by an increased market share of e-publications.

The situation is different in France and Italy that already apply a super-reduced rate to e-newspaper and e-periodicals. These two countries would have to increase their VAT rates. Whereas Italy also had to replace the currently applied super-reduced rate to e-books, France could continue to apply a reduced rate to e-books.

Table 6 Budgetary impact - Option 1

Impact on VAT revenues 29  

Books

Newspapers, Periodicals

Decrease of maximum 15%

BE, CZ, DE, EE, IE, EL, ES, HR, CY, LV, LT, LU, HU, MT, NL, AT, PL, PT, RO, SI, SK, FI, SE, UK

BE, CZ, DE, EE, IE, EL, ES, HR, CY, LV, LT, LU, HU, MT, NL, AT, PL, PT, RO, SI, SK, FI, SE, UK

No impact

BG, DK, FR 

BG, DK 

Increase of maximum 5%

IT

FR, IT

Source: Commission Services own calculation

Depending on the future market share of e-publications, which could be 20% in the medium term, there could be a loss in VAT revenue for all 28 Member States together of 540 million euro annually on the basis of current market shares or a maximum of 2.2 billion euro annually in the medium term. The increase in VAT revenue for France and Italy can be estimated to amount to a maximum of 300 million euro per year in the medium term.

Overall, under this option a decrease in VAT revenue of around 2 billion euro annually could be assumed in the medium term, which represents 0.2% of the total VAT revenue in EU-28.

The impact is calculated on the basis of household final consumption expenditure in Europe on books, newspapers and stationery. The calculation relies on several simplified assumptions, which are the following:

The first assumption relates to the additional VAT loss arising from non-deductible input VAT by exempt sectors and public bodies. This is estimated at 20% of the VAT paid by households.

Expenditure on physical and digital books, newspapers and stationery amounted to EUR 91 billion in 2014. By deducting 10% to exclude miscellaneous printed matter and stationery and drawing materials, we estimate expenditure on books, newspapers and periodicals at approximately 82 billion. These figures are gross of VAT. The bulk of household expenditure originates from Germany (28%), France (16%), the United Kingdom (15%), Italy (10%), Spain (5%), Poland (4%), the Netherlands (4%), Belgium (2%), Austria (2%), Finland (2%), Portugal (2%) and Sweden (2%). The share of the other Member States is much more limited and ranges from 1.4% in Denmark to 0.1% in Malta.

Expenditure for books, newspapers and periodicals has constantly decreased during the last decade, but remained stable in 2014 for the first time. On the basis of this stabilisation, we assume for the sake of modelling that total expenditure will remain roughly constant in the future. This simplified assumption also take into account a potential increase in the number of e-publications due to lower VAT rates. Whereas, the number should increase, the expenditure should remain constant as new e-publications are expected to be less expensive than the printed publications that disappear from the market.

With regard to e-publications, their market share of currently 5% is expected to grow to 20% in the medium term and an average VAT rate differential between the standard and the reduced VAT rate of 13.5 percentage points is taken into account.

Impact on consumers

Assuming an overall pass-on rate of 50%, consumers would benefit from a reduction in prices for e-publications of nearly 7% in all Member States except France and Italy, where consumers would face an increase of prices of 1.7 percentage points in France for the digital press and of 3 percentage points on all e-publications in Italy. Overall, the reduction of prices for consumers should amount to 1 billion euro per year in the medium term.

As mentioned earlier, the assumption of a pass-on of 50% is based on the assessment of the Commission study on e-publications 30 . A VAT rate cut does not necessarily translate into lower prices, as some businesses could opt to keep final sale prices constant. The degree by which any VAT cut is passed on to consumers generally depends on several factors and may well be different from one type of e-publication to another.

The price setting behaviour of publishers is determined by the "willingness to pay" of consumers. The economic literature on the sensitivity to price of publications suggests that overall, books have a high price elasticity, typically exceeding unity, but with large differences between and within market segments; sensitivity to price is typically lower for newspapers, while magazines fall somewhat in between. The overall high price elasticity shows that suppliers have an incentive to pass on VAT reductions to consumers, which will then benefit from lower prices. In case demand elasticity is low (e.g. because readers want to read one specific publication which they are unwilling to substitute), the degree of VAT pass-on will be more limited. There is no data available on which pass on of VAT for publications could be estimated. On the basis of the study, an overall pass-on of 50% is taken as a working assumption to illustrate the possible impacts of a VAT cut. Based on the assumption of a pass-on of 50%, Consumers would benefit from a reduction in e-publications prices by nearly 7% (given a VAT rate differential of 13.5 percentage points) amounting to a maximum of around 1 billion Euro annually.

The above results from the Commission study are borne out by the replies to the open public consultation. According to respondents, e-books prices would go down, either by the amount the VAT is reduced (45%), or by less than that amount the VAT is reduced (39%) and only 8% expect no changes in prices. Similarly, prices for e-newspapers and e-periodicals are expected to decline, either by less than the amount the VAT is reduced (42%) or by the amount the VAT is reduced (32%) and only 12% assume that prices will remain the same. While a plurality of respondents expects the VAT cut to be fully passed on, an equal share expects a less than full or zero pass-on.

The expected decline in prices will result in a corresponding increase in sales volumes for electronically supplied publications, but the extent to which lower prices for electronically supplied publications will increase sales will also depend on the diffusion of suitable electronic reading devices: consumers with access to such devices will react more quickly to digital price changes, while those without access to electronic devices may be reluctant to change over to e-reading irrespective of price cuts.

This is likely to impact differently on e-newspapers and e-books; while large-screen smartphones, tablets and consumers may be sufficient and practical for most consumers of e-newspapers and of some periodicals, e-books are typically read on specialised e-readers, whose diffusion is currently substantially lower. Recent data is available for France, where only 28% of consumer buying e-books possess an e-book reader, but 87% of them possess a smartphone. However, a decline in prices for e-books should spur an increase in ownership of e-readers, as consumers recoup more quickly the cost of the device, particularly in those Member States where VAT on printed publications goes up, and there are recent indications that a certain share of users is not averse to reading on smartphones 31 .

Respondents to the open public consultation believe that in the event of reduced prices for e-publications, consumers would buy roughly the same amount of printed books, but buy more e-books (46%) and some consumers would switch from printed to e-newspapers/e-periodicals (40%).

Impact on the publishing industry

According to the latest Eurostat publication 32 , publishing activities overall generated 0.5% of the gross value added of the total economy in 2013 and the total share in the non-financial business economy was 1% in 2011. The 90 000 enterprises in 2011, of which 99% are SMEs, employed 900 000 persons and around 60% of the persons employed in the publishing industry are employed by the SMEs. The number of employees has slowly decreased over the last decade.

When comparing the turnover of enterprises in the publishing sector, Eurostat reports a turnover of 28 billion euro for book publishing and a turnover of 73 billion euro for publishing newspapers and periodicals in 2011 33 .

However, the Eurostat data and the underlying classification (NACE Rev.2) do not distinguish between printed or digital publications; only for e-books a general market share is communicated by the Federation of European Publishers. Therefore, no quantitative estimate of the impact of VAT rate changes on the publishing industry can be made on the basis of official or administrative data 34 .

Nevertheless, the qualitative analysis of an extension of reduced rates to electronically supplied publications with regard to Member States' budgets suggest that net profit margins should increase by nearly 7 percentage points (given a VAT rate differential of 13.5 percentage points) equivalent to around 1 billion euro per year under the assumptions that the other 50% of the VAT decrease is passed on to the consumers by lower prices. 35 As a consequence, not only the turnover of digital products could increase as a price effect, but there could as well be limited substitution from print to digital. In particular the demand for paperbacks, which is more price sensitive, will see an increased competition from e-books, whose prices are already now lower. 36  

Impacts on individual businesses would depend on their product mix: those mainly selling e-publications would clearly benefit. In addition, in the e-books segment, there are some indications that a higher market share for e-publications could benefit micro- and small enterprises, as these can be efficiently marketed also by smaller businesses. Recent market estimates indicate that on Amazon, the major US retailer, the market share of at least some smaller-scale publishers ('indie' publishers, single-author publishers) is on the increase, whereas the share of 'Big Five' publishers has been on a prolonged slump (see figure below).

Figure 2 Trends in e-book market shares, sales on Amazon.com

Source: February 2016 Author Earnings report, available at http://authorearnings.com/report/february-2016-author-earnings-report/

In the newspapers segment, given current attempts by many newspapers to move from a 'free online' to a paid-subscription business model, it can be assumed that a lower VAT would reduce the current critical pressures on profitability and help maintain diversity and pluralism in the sector 37 ; obviously however the newspaper segment would suffer in the Member States with VAT rate increases.

In terms of jobs, the creation of new opportunities through the growing of electronically supplied publications sector are expected to be limited. Given that e-publishing requires a lower minimum efficient scale of production, this could encourage an increase in the number of viable businesses in the publishing industry, with positive effects for diversity, for a varied cultural offer, and presumably for employment in the creative sector. On the other hand, the overall decreased profitability will have a negative impact on the industry that could result in further job losses. The size of the expected decline in profitability cannot be quantified.

The impact of VAT rate changes in a destination-based VAT system on competitiveness and innovation appear negligible. The only potential impact could derive from the assumption that the digital publishing sector is more vibrant than the traditional publishing sector, which might spill over into limited gains in competitiveness.

Any additional costs that could emerge from the introduction of new VAT rates for e-publications for adapting accounting and sales software are explained and discussed under the distinct criterion "simplicity of the VAT system".

Impact on the printing industry and retail sector

Despite the lag in the availability of official statistics, the Commission services estimate that the printing industry in Europe generates a turnover of around 88 billion euro from printing on paper, plastics, or textiles, comprising 120 000 companies employing over 750 000 persons.

The shrinking of the printing sector has already caused job losses in the printing industry itself and in the distribution sector, which mainly consists of very small enterprises (around 85% of them employ fewer than 20 people) and there is considerable reform ongoing in particular in respect to print like consolidating operations, collaborating and outsourcing that put pressure on SMEs. Therefore, a tax reduction would unlikely be a reason for further job losses and consumers might even substitute physical by electronic publications because for reasons unrelated to the price (i.e. if considered as easier to use). 38 In addition, the restructuring in the sector has rendered the printing industry in some way less dependent from printing on paper. Therefore lower rates for e-publications would not have any significant impact on the printing industry and the current trend is supposed to continue with further job losses, which can however not be quantified in the absence of reliable time-series. There could only be a limited impact for the sector by lower VAT rates for e-publications in case substitution of printed by digital publications would be motivated by price differences. However, both products have different features and the preferred choice of most consumers cannot be influenced by lower prices for e-publications of 5 to 10%.

As for the retail sector, it has faced significant pressures during the last decade due to the increase of distance sales (e-commerce). Online bookshops have increased, whereas brick-and-mortar bookstores have had to adapt their business models, e.g. by opening cafes and offering new ideas such as subscription services in which readers are regularly sent books to suit their interests and tastes. In the US, the world's most advanced market for e-books, large retailers have been especially hardly hit: Barnes & Noble has closed hundreds of stores, while Borders went out of business in 2011. In the EU, the situation varies; in some Member States, fixed book prices have helped independent bookstore to partially resist the increased competition by consumers buying books online. In Ireland and the UK, more than one quarter of independent bookstores have closed, but their number has remained fairly constant in France in the last decade 39 . A limited migration from print to digital would add further pressure on bookstores, but this development is already on-going and the major challenge for bookstores remains competition from e-commerce.

Nevertheless, there is evidence from the US that it is possible for bookstores to successfully meet this challenge. Independent bookstores have rebounded strongly there: data published by the American Booksellers Association indicate that the number of member stores has increased by 27% since 2009 40 , and turnover in the first half 2016 was by 6.1% on the previous year 41 . Given uncertainties on how to estimate how successful these adaptation strategies will be, this impact assessment does not contain any projection of the impact on the retail sector. When taking into account online sales, which are more and more offered by brick-and-mortar stores and the possibility of extending this e-commerce to the supply of e-publications, there should be only be a limited negative impact on these businesses by lower VAT rates for e-publications. The absence of reliable data for Europe on the retail sector does not allow any further quantitative assessment.

The open public consultation also did not seem to reveal grave concerns about the impact on printing industry as, despite the possibility of enhanced price competition with e-books, 55% of the respondents indicated that there should be no negative impacts for the industry. Some respondents even mentioned that traditional booksellers are increasingly present on the e-book market and that a VAT reduction together a simplified taxation of bundles could help booksellers to enhance their e-book offer to final consumers.

The printing industry itself provided 22 answers to the public consultations and the majority of these respondents do not see any impact on the printing industry in case of reduced prices for e-publications (68% with regard to e-newspapers/e-periodicals and 50% with regard to e-books) and only 18% expect a reduced demand for print. The difference among sectors might be due to already high penetration of digital publications in the newspaper market.

Environmental impact

Electronically supplied publications, just like any digital product, might be seen as ecological products since dematerialisation of publications saves from the production of physical support (no paper), and any handling and distribution (warehousing, shops, transport). However, from an environmental point of view, it is not that clear whether e-reading is really environmentally friendly. Indeed, production of electronic devices itself is energy-consuming, uses raw materials and eventually contributes to increasing electronic waste, which is harder to recycle than paper. Hence, one has to balance the ecological benefits from producing, transporting, storing and disposing of physical publications with those from producing, operating and disposing of e-readers. In this respect, to assess the ecological footprint it is useful to distinguish between e-books, which are typically read on dedicated readers, from e-newspapers, which are often read on devices that the consumer would most likely have bought anyway (smartphones, computers, and tablets).

E-books:

Existing research has highlighted that the environmental benefits from e-books depends strongly on the ecological footprint of the e-readers themselves and, crucially, on the number of physical books they offset. In 2010, Apple conducted an environmental report for its iPad tablet, which estimated total lifetime impact of 130 kg of CO2, 58% of which incurred in at production, 11% at transport, and 30% in customer use; disposal and recycling was assessed to contribute only 1%. Newer dedicated e-readers using e-ink technology undoubtedly result in lower operating energy use, as the approximately 3W consumption of an iPad has been markedly improved upon in subsequent e-readers. However, for the Kindle, a Cleantech estimate indicated a lifetime carbon footprint of 168 kg CO2, similar to the Apple estimate 42 . This compares to an estimate of only about 4 kg CO2 per average printed book. A rough calculation thus would suggest that an e-reader would start producing environmental benefits once it replaces 30 to 40 printed books. However, there are a number of uncertainties about the accuracy of carbon footprint calculations, as there could be an underestimation bias by producers. A recent estimate by an environmental group concluded that at 1 to 30 printed books offset per e-reader, printed books have a lower impact, at 60 and over, e-readers have a lower impact, and between these extremes, the impact is uncertain. 43 A trend towards greater use of multi-purpose devices blurs the picture further.

Recent statistics on readership and use of electronic devices exist for the U.S. A Pew Research Centre for 2013 found that the average American reads 12 books per year 44 , which would imply that the average reader would have keep the same device for five years without replacing it, before starting to accrue an environmental benefit. However, a few people read many more books than average, as highlighted by the difference between the median and the average number of books read annually (the median being only 5); a shift to e-reading by these strong users is obviously more likely to be environmentally beneficial. However, only 28% of readers read 11 or more books a year, compared with 32% of Americans that own an e-reader, suggesting that many, perhaps most, strong readers already own a device 45 . Moreover, most owners of e-readers also continue reading printed books; only 5% of respondents reported having read at least one e-book but no printed books in the previous year.

Obtaining a precise estimate of the books actually offset is complicated because readers' behaviour may change as a result of using the reader; they might, for example, purchase more e-books than they would purchase physical books.

On the other hand, to the extent that a cut in the VAT rate on e-books would merely stimulate the replacement of printed with e-books, without boosting the sales of e-readers, the environmental impact could be positive, given that e-readers use little energy.

Overall, the available evidence does not seem to indicate any sizeable ecological benefit from e-books.

E-newspapers, e-periodicals:

The situation is different for e-newspapers. A positive environmental impact is generally assumed for e-newspapers because they are typically accessed on reading devices that consumers already possess, such as smartphones, tablets and computers. Given their limited energy use compared to traditional publications, the environmental impact is unambiguously positive. However, the Commission is not aware of any quantification of this impact.

Impact on the simplicity of the VAT system

Allowing for a reduced rate for e-publications lead to a certain increase in complexity because the current rule, whereby all electronic services are taxed at the standard rate, is the simplest possible arrangement for rates.

Large businesses active in many countries would have to deal with an increased number of VAT rates and to respect possible national specifications defining in which circumstances a reduced rate could be applied to e-publications. For small suppliers engaged on the domestic market only, the impact on compliance costs would be limited.

From the point of view of the EU cross-border supplier, given that it has to charge the VAT rate applied in the Member State where the customer resides, the costs of VAT compliance depend on how clear and user-friendly are the national rules established by the Member States. Currently, determining the applicable rate for e-publications is straightforward, as suppliers just need to look up the standard rate in the country. 46 Reduced rates for e-publications will, however, oblige suppliers to verify, if their services are eligible for a reduced rate in the country of the customer and under which conditions in order to establish the correct VAT rate in each Member State, where their customers reside. Suppliers established in third countries that electronically supply publications to EU customers will also have to comply with the VAT rules in each Member States.

Whereas businesses already engaging in cross-border sales of physical books would have accounting and software solutions that allow for an adaptation to e-publications, businesses when wanting to start engaging in selling e-publications cross-border would face additional costs for implementing such solutions. However, while selling physical goods still involves above a certain threshold (between 35.000 and 100.000 euro depending on the Member States) registering in all Member States, where customers are located, suppliers of e-publications can since 2015 use a One-Stop Shop (OSS) and continue to register for VAT and pay VAT in only one Member State, which they can choose. This facilitates the market entry for smaller suppliers and reduces the administrative burden by 95% compared to a registration in all Member States. 47 Any additional costs for software are a general requirement in a destination based VAT regime and are negligible in comparison to the reduction of administrative costs by using the OSS. In addition, businesses have a degree of flexibility about how to identify the location of their customers. They are not obliged to use IP-addresses and could as well use e.g. the IBAN number of the customers' bank accounts.

This conclusion found support in the open public consultation. A large majority of businesses that answered to the open public consultation do not expect an increase in administrative burden and only very few respondents highlighted that the increased complexity could complicate cross-border sales for smaller platforms and that only big companies could afford software solutions automatically locating customer and calculating VAT accordingly.

On the other hand businesses nearly unanimously highlighted in the open public consultation that aligning the rates for printed publications and e-publications would reduce administrative burden, because the taxation of bundles (a supply consisting of a printed book together with a code to download an electronic version of the same book or a printed newspaper with access to the corresponding e-newspaper) would be simplified. The need to apply two different tax rates under a single bar code was seen as a severe obstacle in selling bundles due to the need for specific software solutions and thus in promoting e-publications.

Whilst such claims by the concerned sector are justified to a certain extent, it is important to mention that the taxation of bundles falls under the sole responsibility of the Member States. The taxation of bundles is not regulated at EU level in the VAT Directive and therefore there are no rules that prohibit Member States from granting a reduced rate to certain bundles under the current VAT rates. However, if Member States applied a reduced rate to printed publications and e-publications, this would indeed provide supplier with increased certainty in case of supplying bundles. 48 .

With regard to the VAT Directive and the common system of VAT in the EU level, extending reduced rates to electronically supplied publications will complicate the common system. By breaking the simpler current principle by which the standard rate is applied to all e-services, Member States would need to distinguish
e-publications from other e-services and would at least need to exclude from the reduced VAT rate any books, newspapers or periodicals that predominantly consist of music and video content. The result, 28 national definitions, will add complexity to the VAT system and could trigger discussions on the national definitions in each Member States about extending the reduced rate treatment to all kinds of digital news services, other electronic services related to publishing or even internet providers.

Third countries are not directly affected by the proposed changes to the VAT Directive, as the VAT Directive only applies to the EU territory,

6.3.Option 2: possibility of aligning VAT rates for publications at a reduced rate and at a super-reduced or zero rate

Option 2 aims at allowing all Member States to align VAT rates for e-publications with the VAT rates currently applied to printed publications. In order to achieve this aim a provision of granting to Member States the possibility to apply super-reduced and zero rates to e-publications has to be introduced in the VAT Directive. As the provision should not introduce a preferential treatment for e-publications over printed publications in the VAT Directive, the provision needs to cover printed publications as well.

Of the 45% of the respondents of the open public consultation that voted against abolishing super-reduced and zero rates, the large majority (90%) agreed to allow all Member States to apply super-reduced or zero rates to all publications.

By introducing a general provision allowing Member States to apply super-reduced and zero rates to publications, such a provision would replace the 10 existing derogations with regard to super-reduced and zero rates for printed publications, which would need to expire once the definitive VAT regime enters into force. While it cannot be known whether Member States would use this freedom to lower VAT rates for e-publications and possibly printed publications, the impact assessment is carried out under the assumptions that Member States would align the VAT rates for electronically supplied publications with the VAT rates currently in place for printed publications, i.e. 26 Member States would apply a reduced VAT rate to e-publications and that 9 Member States would in addition apply super-reduced and zero rates to certain e-publications. Bulgaria would continue to apply the standard rate and Denmark would apply the standard rate to e-books and a zero rate to most newspapers and periodicals.

It should be stressed, as done under option 1, that the impact of the measure would in fact be smaller.

Budgetary impact on Member States

VAT revenue on electronically supplied publications would decrease in all Member States except Bulgaria. In line with the present market share of e-publications the decrease in VAT revenue would be limited, although it would constantly increase with an increasing market share of electronically supplied publications.

The biggest impact would be expected for the UK, which has the biggest market for electronically supplied publications in the EU and where the VAT rates for electronically supplied publications would be lowered from 20% to 0% for most publications.

According to the "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)" a maximum immediate decrease of 5% in VAT revenue can be expected based on the fact that the market share of e-publications is less than 5% in all Member States except the UK, but the decrease could amount up to 15% in most Member States in the medium term by an increased market share of e-publications.

No change is assumed for France and Italy that already apply reduced or super-reduced rate to e-publications, as well as for Bulgaria and for Denmark with regard to books.

Table 7 Budgetary impacts - Option 2

Impact on VAT revenues 49  

Books

Newspapers, Periodicals

Decrease of more
than 15%

BE, ES, HU, IE, PT, UK

DK

Decrease of maximum 15%

DE, EL, HR, CY, LU, MT, NL, AT, PL, RO, SI, SK, FI, SE

HR, CY, LU, MT, RO, UK

Decrease of maximum 5%

CZ, EE, LV, LT

BE, CZ, DE, EE, IE, EL, ES, LV, LT, HU, NL, AT, PL, PT, SI, SK, FI, SE

No impact

BG, DK, FR, IT 

BG, FR, IT 

Source: Commission Services calculation based on the "Economic study on publications on all physical means of support and electronic publications in the context of VAT"

There could be a VAT loss for all 28 Member States together of a maximum of 2.2 billion euro per year in the medium term due to the introduction of reduced rates and a further decrease in VAT revenues would occur due to the introduction of super-reduced and zero rates, which could amount to 2.5 billion euro per year in the medium term. Overall, the total VAT loss with regard to e-publication could amount to 4.7 billion per year in the medium term, which represents 0.5% of the total VAT revenue in EU-28.

Impact on consumers

Assuming an overall pass-on of 50% like under option 1, consumers could benefit from a reduction in prices for e-publications of more than 10% on an average in
EU-28 amounting to around to 2.3 billion euro per year (half of the VAT cuts).

An expected decline in prices would result in some increase in sales volumes for
electronically supplied publications in all Member States, because all Member States could align VAT rates for e-publications with the lower VAT rates in place for printed publications. The decrease in prices for electronically supplied publications would be higher than under option 1 and could encourage some more consumers to buy these e-publications, but again depending on certain other aspects as outlined under option 1.

Impact on the publishing industry

By extending the reduced, super-reduced and zero rates to electronically supplied publications, an increase in the profitability of digital products is expected and estimated at around 10% or around 2.3 billion euro annually (half of the VAT cut). The increase in the turnover of digital products would also be influenced by migration from print to digital that reduces the consumption of printed books and periodicals at the same time.

According to the "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)" that assessed this option the total turnover of the publishing industry could slightly decrease by a maximum of 5% across all Member States as the loss in sales from print in absolute terms would be slightly higher than the gains from e-publications in absolute terms. These estimates therefore include the assumptions of a migration phenomenon from print to digital that reduces the consumption of printed books and periodicals at the same time. In particular demand for paperback books, which is more price sensitive, will see an increased competition from e-books whose prices are already slightly lower. 50  

The study estimates that a majority of Member States could experience a decrease between 5% and 15% in the book market and between 1% and 5% in the newspaper and periodical market.

The benefit of increased profits for the publishing industry as outlined in option 1 apply also to this option and would be even more pronounced, because of an even higher increase of the profitability .

Impact on the printing industry and retail sector

The impact is identical to option 1.

Environmental impact

The impact is identical to option 1.

Impact on the simplicity of the VAT system

As under option 1 there is a negative impact deriving from the departure from the principle of standard rate treatment for all electronic services.

The main challenge with regard to options 2 lies in the fact that option 2 introduces the possibility to apply new super-reduced and zero rates to e-publications. If Member States decide to introduce zero rates for e-publications but fail to introduce clear provisions in their national legislation to exclude predominant audio-visual content from the scope of these rates, the partial supply of zero rated internet content could be established and the complexity of taxing the digital economy would increase.

Allowing Member States more freedom in setting VAT rates, in particular super-reduced and zero rates, is a political choice to give preference to subsidiarity over the possible risk of adding complexity to the VAT system.

In this regard also option 2 is also a temporary solution, because the list of goods and services eligible for super-reduced and zero rates would for the moment only encompass publications. Whereas the list of goods and services eligible for reduced rates has been established during two decades of discussions in the Council, this constitutes a temporary unequal treatment of goods and services before the general VAT rate proposal in 2017, which would involve discussions regarding which goods and services such a treatment should be allowed and a general discussion on the scope of such a list.

7.HOW DO THE OPTIONS COMPARE?

7.1.Objectives

Option 1 does not fully achieve the specific objective of allowing all Member States to align the rates for e-publications with the rates they currently apply to printed publications, as only 19 Member States could do so.

Option 2 fulfils the specific objective and allows all 28 Member States to align their VAT rates for printed publications and e-publications at the rates currently in place.

7.2.Impacts

Apart from budgetary impacts, no option has significant economy-wide impacts beyond the publishing and printing/distribution sector (and their customers). The impacts are limited to publishing activities that generated 0.5% of the gross value added of the total business economy (around 1% of the non-financial business economy) in 2013 and which even include advertising newspapers, which either free of charge or taxed at the standard rate now and under both options.

It is impossible to estimate the impact of a change in the VAT Directive with regard to VAT rates without making assumptions on whether Member States would change the current VAT rates in place. The VAT Directive does not impose any VAT rates. The VAT Directive only determines minimum rates and limits the application of any reduced rates in order to harmonise the common system of VAT in the European Union, which is essential for an origin based definitive VAT regime as it was initially foreseen. Given that it is the Member States' responsibility to determine VAT rates, it is not sure how many and which Member States will implement any rate cuts. Thus, for the purpose of this Impact Assessment the technical assumption that all Member States except Bulgaria will lower VAT rates constitutes a maximum impact scenario.

When comparing the two options, the only difference lies in the possibility to apply super-reduced and zero rates to publications under option 2. As stated before, the general VAT proposal in 2017 would include such a possibility and option 2 is therefore more in line with the Actions Plan on VAT. However, the adoption of a general VAT rate proposal will take time and option 2 would introduce a temporary unequal treatment between publications and all other goods and services (including food), for which there is no general super-reduced and zero rate available.

With regard to revenue erosion it should be noted that Member States have until now been protected by certain restrictions for Member States in setting VAT rates. This would be less the case in a definitive VAT regime granting Member States more freedom in setting VAT rates. Member States would take more responsibility to determine the VAT revenue that should be generated. Member States are surely aware that VAT subsidies in form of lower VAT rates can hardly be designed in a revenue neutral way. In case consumers bought more publications because of lower VAT rates, such VAT revenue could well be increased, but at the same time, more VAT at mostly the standard rate can be deducted by the suppliers for their inputs. So, the more products are taxed at a reduced or super-reduced VAT, the more VAT at the standard rate is deducted and this has the tendency to decrease VAT revenues even further.

Budgetary impact vs. baseline

Under the baseline scenario VAT revenues would need to increase for France and Italy. Under option 1, most Member States would see a small decrease in VAT revenues, because they can apply reduced VAT rates to e-publications. Under option 2 VAT revenues could further decrease because of the applications of super-reduced and zero VAT rates for e-publications.

Baseline scenario: positive impact (+ 1 billion euro per year)

Option 1: negative impact (- 3 billion euro per year compared to baseline)    

Option 2: negative impact (- 5.7 billion euro per year compared to baseline)

Consumers

Not taking into the account the impact on consumers in France and Italy under option 1, consumers will benefit from reduced prices for e-publications, which are estimated to amount to half of any VAT cuts

Option 1: positive impact (++)

Option 2: positive impact (++)

Publishing industry

Under each option the publishing industry would benefit from the fact that all of their supplies could benefit from a VAT reduced rate. Compared to the baseline scenario an overall increase in profitability amounting to half of any VAT cuts is assumed for both options.

Option 1: positive impact (++)

Option 2: positive impact (++)

Looking at SMEs in the publishing industry, the impact assessment has shown that with regard to digital book publishing SMEs already now benefit from simplified market access and from the fact that e-publishing requires a lower minimum efficient scale of production.

Option 1: positive impact (++)

Option 2: positive impact (++)

Printing and retail sector

Under each option, the printing industry would be challenged by lower VAT rates for e-publications and a potential substitution of printed publications by e-publications.

Option 1: limited negative impact (-)

Option 2: limited negative impact (-)

Environmental impact

The environmental impact of all options mirrors somewhat the impacts with regard to the printing industry. In particular a substitution of printed newspapers by electronically supplied newspapers has a positive environmental impact, whereas the environmental impact with regard to e-books depends on the future ability of readers to use multi-purpose devices.

Option 1: positive impact (++)

Option 2: positive impact (++)

Simplicity of VAT system

Both options allow Member States to align the VAT rates for all publications at the level of reduced rates and provide suppliers with legal certainty when supplying bundles, but break with a simple and clear rule to tax all e-services at a standard rate. This would complicate the rules in comparison to the baseline scenario and has a negative impact in both options. However, under option 2 the VAT system is further complicated by the introduction of a general provision allowing Member States to apply super-reduced and zero rates, which do for the moment not exist for the supply of any other goods or services.

Option 1: limited negative impact (-)

Option 2: negative impact (--)

Overall assessment (objectives and impacts)

Effectiveness:

Option 1 only allows 19 Member States to align the VAT rates for e-publications with the current VAT rates for printed publications and option 2 allows all 28 Member States to align VAT rates for e-publications with the current VAT rates for printed publications.

Coherence:

Both options are coherent with the subsidiarity approach outlined in the Action Plan on VAT, providing Member States with more flexibility in setting VAT rates in a destination based VAT regime. Option 1 establishes a more limited solution, which could be reinforced later upon entry into force of a more general rates proposal, while option 2 aims at providing already now full freedom in setting VAT rates in one specific area. Given the unanimity principle in taxation matter option 2 would test the waters ahead of the introduction of a general reform and would stimulate reflection in the Council on how derogations (super-reduced and zero rates) that need to expire in a definitive VAT regime should be replaced by general provisions for all Member States.

Efficiency:

The impacts of the options depend solely on Member States' decisions. Member States could always apply a standard VAT rate to the supply of all goods and services, but they have a long tradition of granting VAT subsidies to certain sectors for the benefit of the final consumer. Under the assumptions made in this Impact Assessment report the impacts of option 1 are more positive than for option 2 because of the budgetary impact and the impact on the simplicity of the VAT system. Nevertheless, these impacts are the direct effect of a political choice to give Member States full (or at least more) flexibility and responsibility in setting VAT rates, which is foreseen for the definitive VAT regime as announced in the VAT Action Plan.

In line with the subsidiarity principle, this impact assessment prioritizes freedom of choice for Member States, as these will at any rate retain a full ability to safeguard revenue by not lowering the VAT rate on e-publications, and thus can control to what extent the budgetary cost is proportionate to the objective. To assess whether revenue or other considerations should prevail, is at the end of the day up to each Member State. As a result, both options are preferred over the baseline scenario.

Both options constitute a political choice to respond to the urgent request expressed by many Member States to be enabled to align VAT rates for all types of publications. They only differ by the extent by which flexibility is granted. In comparison to option 1, option 2 fully achieves the objective of rate alignment in all Member States, while advancing further on granting full rate flexibility for the supply of a specific product (publications), ahead of a general VAT rate proposal.

Option 1: limited positive impact (+)

Option 2: limited positive impact (+)



Figure 3: Comparison of options with objectives and impacts

OPTIONS

Baseline scenario

TIMELY CHANGES TO THE RULES ON VAT RATES

Option 1

Reduced rate for all publications

Option 2

Reduced, super-reduced and zero rates for all publications

OBJECTIVE ADRESSED

Allow Member States to align VAT rates for publications

No

Partly

(19 Member States)

Fully

(28 Member States)

IMPACTS

Budgetary impact

+ 1 billion euro annually

Up to - 3 billion euro annually

Up to - 5.7 billion euro annually

Consumers

0 51

++

++

Publishing industry

(SMEs)

0 52

(0)

++

(++)

++

(++)

Printing and retail sector

0

-

-

Environmental impact

0

++

++

Simplicity of VAT system

0

-

--

OVERALL ASSESSMENT
(objective and impacts)

0

+

+

0 : zero or negligible impact,

+ / - : limited positive / negative impact,

++ / -- : positive / negative impact,

+++/ --- : significant positive / negative impact


7.3.Preferred option

Option 1 and 2 are preferred over the baseline scenario, because they fulfil (at least partially) the specific objective of this targeted initiative. From a technical point of view there is no clear preferred option, but option 2 has the advantage that it is fully in line with what the Commission communicated in its Action Plan on VAT, namely that derogations (super-reduced and zero rates) only available to a few Member States should be granted to all Member States in a definitive VAT regime.

Setting VAT rates falls under the sole responsibility of Member States and thus, once such rules are adopted, Member States will not be obliged to change the rates they currently apply and in particular, Member States will not be obliged to align the VAT rates for printed publications and e-publications. They can continue to apply a standard rate to electronically supplied publications and continue to apply lower rates for printed publications.

All impacts depend on the assumption that Member States would align VAT rates as far as the VAT Directive would allow them. The impacts shown therefore illustrate a maximum impact; Member States would be able to modulate the impacts by acting on the rate level.

Granting Member States freedom in setting VAT rates by allowing the application of reduced, super-reduced and zero rates for publications does not entail any risk of creating distortion of competition or poses a problem for the functioning of the single market.

Eventually, Member States have to make a choice between raising VAT revenue or lower (up to zero) VAT for certain products. Member States do this already now, but under less flexibility. More flexibility also implies that Member States will have to deal with increased pressure from lobby for extending reduced VAT rates in a definitive VAT regime that is currently be dealt with at the EU level.

8.MONITORING

Monitoring is required with regard to the correct application of the VAT Directive.

If a Member State decides to apply other rates than the standard rate to
e-publications, it would need to introduce national provisions that specify the general provisions at the EU level, i.e. this Member State would need to define what is an "e-book", what is an "e-newspaper" and/or what is an "e-periodical". These definitions fall under Member States' responsibility, but Member States are not allowed to extend the scope of application of reduced, super-reduced or zero rates to other goods and services than books, newspapers and periodicals. Furthermore, Member States would be obliged to tax publications that predominantly consist of music and video content at the standard VAT rate. Consequently, national provisions will be monitored as part of the continuous monitoring of the correct application of the VAT Directive.

Beyond that, the monitoring at EU level takes place in a continuous review of the scope of the reduced rate in line with Article 100 of the VAT Directive. To this extent, there is also a continuous collection of information about the VAT rates in place in each Member State. Such information is currently published by the Commission in the document "VAT rates applied in the Member States of the European Union" 53 , which primarily serves as an information tool for researchers, businesses and final consumers. In future, with the implementation of a definitive VAT regime, such information should be collected on a mandatory and standardised basis. A web portal could provide accurate, timely and binding information on the reduced VAT rates in place in each Member States. A list and definitions of the products that are not covered by the standard rate in each Member State could be included in the web portal and regularly updated by the Member States.

As VAT rate setting falls under the Member States responsibility and does not influence the own resource payments and therefore EU budget, the Commission does not monitor VAT revenues in the Member States with regard to specific products. This is also not possible for Member States, because VAT rates payments by suppliers might be encoded with regard to the VAT rate they relate to, but they cannot be linked to a specific product, which would be the condition allowing for such monitoring.

9.ANNEXES 

9.1.Procedural information 

1. Agenda planning and Work Programme References

The Agenda Planning Reference is 2016/TAXUD/005.

2. Inter-Service Steering Group

An Inter-Service Steering Group was set up for the follow-up of the Action Plan on VAT. The Steering Group met on four occasions, on 8 June, 5 July, 20 July and 1 September 2016 to discuss VAT for e-publications. An additional meeting was held on 24 October 2016 to discuss a revised version of the Impact Assessment following integration of comments from the Regulatory Scrutiny Board. The following directorates and services were consulted: DG Agriculture and Rural Development, DG Budget, DG Communications Networks, Content and Technology, DG Competition, DG Economic and Financial Affairs, DG Education and Culture, DG Environment, DG Internal Market, Industry, Entrepreneurship and SMEs, DG Justice and Consumers, the Secretariat-General and the Legal Service. The feedback received from these directorates and services was taken into account in the report and the first draft by DG TAXUD was substantially amended during the process.

3. Consultation of the Regulatory Scrutiny Board

The Impact Assessment Report was submitted to the Regulatory Scrutiny Board for a first time on 14 September 2016 and resubmitted on 26 October 2016.

On 11 November, the Regulatory Scrutiny Board gave a positive opinion with reservations. The report was further adjusted in order to integrate the Board's recommendations.

Board recommendation

What has been done?

Where?

1.

The report should clarify how the present initiative is articulated with the Commission's approach to the definitive VAT regime.

It is explained that the Action Plan puts forward two options for the rules on VAT rates in a definitive VAT regime, which both would make available to all Member States the derogations (super-reduced and zero VAT rates) currently available to only a few Member States.

Section 5.3

2.

The report should further elaborate the baseline scenario to include those elements of the definitive VAT regime that have already been agreed and those that remain uncertain at this stage.

It should revise the general objective of the initiative.

It is explained that the foreseen elements of the definitive VAT regime have no impact on the baseline scenario.

The general objective of the initiative has been revised and complemented by a second general objective.

Section 5.1

Sections
4.1 and 4.2

3.

The report should clarify the scope of the options and assumptions behind the impact analysis and

compare the options in a more transparent and balanced way.

It is emphasised that the impacts are maximum impacts based on a technical, but not realistic, assumption that all Member States would lower VAT rates for e-publications as far as they could.

The technical analysis does no longer identify a preferred option.

Sections 6.2 and 6.3

Sections 7.2 and 7.3

Further recommendations for improvements by the Board:

What has been done?

Where?

The planned VAT reform is not fully reflected in the report

The options outlined in the Action Plan were explained.

Section 5.3

Naming the options should be consistent across sections 5 and 6

Names of the options in section 6 have been replaced with the names of options in section 5.

Sections 6.2 and 6.3

Descriptions of option 1 should be clearer

It has been emphasised that that option 1 retains the existing leeway for super-reduced and zero-rates for some publications in some Member States.

Section 5.2

The impact analysis section should clearly state the assumption that all Member States will exercise to the maximum extent all leeway that the options provide, i.e. they will lower VAT rates to the fullest possible extent.

Such a statement has been introduced for both options.

Sections 6.2 and 6.3

There appears to be little evidence to corroborate that a reduced VAT on e-publications will lead to lower prices for consumers. The report shows contradictory outcomes between the Commission's study and practice and past experience. The report should therefore explain why it bases the impact assessment, and in particular the options, on this assumption.

The reference to past experience has been deleted and the discussion enirely refers to economic theory that suggests that high price elasticities are an evidence for lower prices as a result of at least a partial pass on of the VAT cut to final consumers.

Section 6.2

The report should not aggregate across all respondents (e.g. pages 23, 33 and 45). Documented differences across groups of stakeholders contradict the blanket statements on pages 45 and 46 that responses do not differ significantly across groups of respondents.

On page 23 differences across different groups of stakeholders have been added.

On page 33 is has been made clear that only businesses are referred to.

The blanket statement on page 45 and 46 has been corrected.

Sections 6.2 and 9.2

 

9.2.Stakeholder consultation 

The consultation strategy, besides verifying the results of recent public consultations in 2008, 2011 and 2013 on VAT rates and in particular e-books, had two main purposes. The first was to get the views of stakeholders on the Commission’s commitment to allow Member States the application of reduced VAT rates for e-publications, and the second was to provide them with an opportunity to present their views on the most appropriate definition(s) for e-publications.

Therefore, an open public consultation for the initiative was held for 8 weeks between 25 July and 19 September 2016 54 using the EU Survey tool; in the rare cases, where respondents wished to provide a filled in questionnaire by email rather than in the EU Survey tool, the answers were introduced in the system on behalf of the respondents. The consultation period had to be shortened in order to be able to present a proposal before the end of 2016 in accordance with the Action Plan on VAT. Due to this short time frame and to technical difficulties, the questionnaire was only provided in English, but upon request respondents were informed that answers to the few open questions included in the consultation could be given in any of the 24 official languages in the EU. The Commission received 858 submissions via EU Survey and four other contributions by email. All public submissions will be made available on the DG TAXUD website in October 2016.

Although Member States already reacted in the Council in May 2016 to the Action Plan on VAT and invited the Commission to present a proposal on e-publications, the Belgian Federal Public Service of Finance, the Slovenian Ministry of Culture, the HM Treasury of the United Kingdom, the Spanish tax authorities and the Swedish government reacted again to the open public consultation and expressed support for such a Commission initiative. Other Member States have also reiterated their support, mainly in bilateral meeting with members of the European Commission.

A large majority of respondents were readers, but with 251 responses also business and organisations participated largely in the consultation together with 107 authors. Nearly all respondents indicated under 'other' were libraries, for which no specific category had been foreseen.

Profile of respondents Open Public Consultation

Author    12%

Company/enterprise    16%

Organisation, association, consultancy, law firm    14%

Member State or other Public authority    1%

Researcher, academic institution, think tank    4%

Reader    50%

Other    4%

The main field of economic activity of the participating businesses and organisation were publishing of books, newspapers, periodicals, music or video recordings (56%).

There is evidence of a campaign due to a large number of replies from Poland (28%), mainly in the 'readers' category. However, these responses have no material impact on the results and therefore all contributions have been included in the analysis. In terms of the participation from other countries, a significant 19% indicated that the respondent was based in the UK and 15% in France.

The consultation also included an open question at the end of each section. These questions gave respondents an opportunity to elaborate on their replies. Mainly businesses and organisations elaborated on their responses and pointed out the need to enable Member States to apply the same reduced VAT rates to printed publications and e-publications.

The title of the open consultation was "Public Consultation on reduced VAT rates for electronically supplied publications". The consultation did not aim at discussing the current treatment of printed publications or publications on other means of physical support. It was therefore appropriate to carry out an online consultation.

Overview of consultation results

It is important to note that the responses differ slightly between the different groups of respondents. If all answers are aggregated, the key findings are:

94% of the respondents agreed that Member States should be allowed to apply a reduced VAT rate to e-books.

88% of the respondents agreed that Member States should be allowed to apply a reduced VAT rate to e-newspapers and e-periodicals, too.

A plurality of respondents (45%) found that super-reduced and zero rates for printed publications should not be abolished. Out of these 45%, an overwhelming majority (90%) was in favour of allowing all Member States to apply super-reduced or zero VAT rates to printed publications and e-publications.

The results of this open public consultation are in no way representative, because participation was made on a voluntary basis and for specific groups of respondents the number of responses was simply too low. In addition, asking consumers and in particular readers about a potential tax cut by Member States, triggers positive answered towards a tax cut. However, the answers relating to existing super-reduced and zero rates show that respondents are well aware of a situation where tax cuts might go too far.

Open public consultations also give a good picture about the interest of stakeholders in a specific issue and the overall high amount of responses in a very limited time frame shows great interest by readers and the publishing industry in this subject. It is important to carve out the nuances of differences in responses to the closed questions as far as possible, in order to value the results of this open public consultation. Otherwise, it can only be concluded from this public consultations that readers have already been influenced by several campaigns advocating for reduced VAT rates for e-books, because obviously most respondents were motivated to reply to this consultation because of books and when being asked about newspapers they were slightly less in favour of allowing reduced VAT rates.

Therefore, stakeholders were asked about the effects they expected from a reduction of VAT rates in two distinct sections (one on books and one on newspapers and periodicals). According to the stakeholders, the predominant effects would be the following:

For e-books, prices go down, either by the amount the VAT is reduced (45%) or less than the amount the VAT is reduced (39%).

For e-newspapers and e-periodicals prices go down, either by less than the amount the VAT is reduced (42%) or by the amount the VAT is reduced (32%).

If the VAT cuts are not fully passed on to final consumers, profit margins for publishers would increase; the sector would, in the case of books, invest in new content (47%) and, in the case of newspapers, not only invest in new content (42%), but also be less dependent on advertising revenue (41%).

In case prices for e-publications fall, consumers would buy roughly the same amount of printed books, but buy more e-books (46%) and some consumers would switch from printed to e-newspapers/e-periodicals (40%).

No impact of reduced prices for e-books is assumed by the majority of stakeholders (55%) for the printing industry, the pulp and paper industry, and bookstores. With regard to newspapers stakeholders were more divided: 37% assume a reduced demand for print and reduced number of newspaper stores and equally 37% assume no impact.

These answers show that respondents even though the large majority of them advocates for a reduction in VAT rates for e-publications are well aware that a reduction in VAT will not automatically passed on to final consumers.

When asked about the definition of the terms book, newspaper and periodical at EU level, the majority (59%) of respondents indicated that there is no need to define the terms in a uniform way at EU level. Even when respondents replied that there was a need to define the terms in a uniform way at EU level, they often argued differently in their comments. Some respondents tried to elaborate on this complex issue and concluded that there was even no need in some Member States to define further the terms book, newspaper or periodical in their national law. One respondent summarised the situation, recalling the diverging definitions in several Member States, and concluded that the International Standard Book Number (ISBN) could qualify as a criterion for books; however, this respondent also noted that the ISBN encompasses a range of educational materials, some of which may not be regarded as books, such as videos and films.

In two final sections, businesses were asked about the inclusion of audio and video material in e-publications:

25% of the businesses already now sell e-books with audio and video content alongside text, 52% do not offer it but think that the market may soon demand it and only 23% do not envisage, in the foreseeable future, selling e-books with audio and video content alongside text.

39% already offer now paid services including audio and video content alongside text, 41% do not offer it but think that the market may soon demand it and only 20% do not envisage, in the foreseeable future, offering paid services including audio and video content alongside text.

In the same section businesses were subsequently asked if the introduction of reduced VAT rates for e-books would increase the administrative burden for their business. A 12% share of the businesses see such an increase with regard to e-books and 6% with regard to e-newspapers and e-periodicals. However, the large majority (around 68%) of the businesses do not expect an increase in administrative burden. In this regard, those businesses that commented further nearly unanimously argued that they would face a lesser administrative burden with regard to bundle sales, and some mentioned the challenge of complying with diverging VAT rates, particularly for smaller businesses.

Observations relating to specific groups of respondents

As previously mentioned, answers to the consultations do not differ significantly between the different groups of respondents. Most importantly, the overall sample of the consultation is not representative and this is in particular true for partial samples. Hence, differences between groups have to be interpreted carefully.

A small difference can be observed with regard to the support for a reduced rate for e-books, which was the highest among readers (97%), researchers (98%) and libraries (100%), whereas a lower level support was found in the printing industry (73%) and, surprisingly, the publishing industry (still however overwhelming at 83%). This less-than-unanimous support among publishers might perhaps result from entries of businesses whose main focus is on the traditional printed medium. However, with regard to e-newspapers and e-periodicals the support was between 88% and 90% among all groups.

As previously mentioned, readers are mainly located in Poland (53%). On the other hand, the large majority of authors are located in the United Kingdom (81%), in line with the greater development of the e-book sector there. Researchers participating in the survey come mainly from Italy, Spain and the Netherlands.

The respondents from the publishing industry indicate mostly France (21%), Germany (17%) and the United Kingdom (17%) as their location. The biggest difference in their answer to the other respondents relates to a possible EU level definition of books, newspapers or periodicals. Compared to an average of 59% of respondents, 80% of the publishing industry respondents indicate that no specific EU level definition was required. A second difference is that 60% (with regard to books) and 62% (with regard to newspapers/periodicals) of these respondents declare that their sector would invest in new content, in case profit margins for publishers would increase. This is considerably higher than what the average of all respondent (47% with regard to books and 42% with regard to newspapers/periodicals) believes.

From the 22 answers from the printing industry 27% came from Germany and 23% from the United Kingdom. The most remarkable observation about this group relates to the fact that the majority of these respondents do not see any impact on the printing industry in case of reduced prices for e-publications (68% with regard to e-newspapers/e-periodicals and 50% with regard to e-books) and only 18% expect a reduced demand for print. The difference among sectors might be due to already high penetration of digital publications in the newspaper market.

Finally, with regard to super-reduced and zero rates, a difference can be observed between readers, authors and researchers on the one hand and the printing industry and publishing industry on the other hand. The first group is undecided about abolishing super-reduced and zero rates, whereas both the printing and publishing industry not surprisingly argue strongly against abolishing super-reduced and zero rates (74%) and demand to allow Member States applying such rates to all publications (on average 66%).

Previous public consultation

In 2008, as a result of the open consultation "Review of existing legislation on VAT reduced rates" 55 , there were suggestions that the reduced rate applied to supplies of books, newspapers and periodicals might need to be updated to also cover the supplies of these products in electronic form, as it would result in an equal reduced rate treatment of classic and similar digital goods.

In 2011, in the open consultation on the "Green Paper on the Future of VAT - Towards a simpler, more robust and efficient VAT system" 56 , the supply of electronically supplied publications and other cultural goods or services were highlighted by interested stakeholders and some Member States as candidates for reduced rates, as they were perceived as suffering from VAT rate discrimination. Publishers’ representatives largely supported the extension of the reduced and even zero rates currently applied to printed publications to electronically supplied publications.

The results obtained from the first two open consultations on VAT reduced rates served as a basis for a targeted public consultation on the "Review of the existing legislation on VAT reduced rates" 57 in 2013, asking specific questions on how to implement an option for Member States to apply reduced VAT rates to electronically supplied services, in particular e-books:

Do you agree that those electronic services that would qualify for the reduced rate will have to be precisely defined in a uniform way at an EU level or do you consider that a broad definition in the VAT Directive would be sufficient?

Considering the need for a uniform and future proofed approach at EU level, what should be the definition of an e-book in EU-law?

The large majority of replies to the first questions concerned online and off-line publications (books, newspapers and periodicals). Quite a number of respondents advocated the application of the same VAT rate to similar goods and services and stress the need to precisely and uniformly define at an EU level those electronic services that would qualify for the reduced rate. However, about the same number of respondents argued that a precise definition reduces the ability of Member States to adapt the terms to existing national legislation and that it would add complexity to the VAT system. In addition, a significant number of respondents pointed out that a definition would fail, if it was not broad and robust enough to cater for future technological developments.

9.3.Who is affected by the initiative and how?

The Commission intend to grant options to Member States by the legislative initiative. Consequently, the use of the options by Member States is determining the impacts. If Member States made use of the options, it would affect consumer, the publishing industry, the printing industry, the environment and the simplicity of the VAT system. Lower VAT rates for publications would lead to lower prices for final consumers and increase profits of the publishing industry.

9.4.Analytical models used in preparing the impact assessment 

The study used for this impact assessment:

Ramboll, The Evaluation Partnership and Europe Economics, 2013

“Economic study on publications on all physical means of support and electronic publications in the context of VAT”

aims at assessing the substitutability between physical and electronically-supported publications; evaluating the impact of current VAT reduced rates; and evaluating the impact of extending either the reduced rate or the standard rate regime to all kinds of publications.

Data were obtained by case studies and surveys.

The assessment of the impact of changes in VAT rates looks at a number of potential effects separating between supply and demand impacts (in print and digital formats):

Demand-side effects are the reactions consumers might have if faced with a change in prices. The key aspect to consider when assessing demand-side effects is the extent to which consumers are sensitive to price changes (changes in price that might result from an increase or decrease in the tax rate), and the extent to which consumers consider print and digital materials to be substitutable (including the extent to which an increase or decrease in the price of material in one support may result in migration to products delivered in the other support).

(1)Supply-side effects concern the supply chain of publishing material and how this might be affected by a change in the tax rate. Publishers and players within the supply chain might differ in their ability to absorb a tax rise or to react to a tax decrease. The extent to which suppliers can pass on to consumers any increases (or decreases) in their costs (including changes as a result of tax increases or decreases) is known as the ‘pass-on effect’.

The impacts on the supply side, in conjunction with the demand-side effects, play a key role in determining the extent to which a change in tax is actually reflected in the final retail price.

As the study did not analyse the budgetary impacts for Member States, own calculations on the basis of simplified assumptions, which are described under option 1, were carried out. The impact is calculated on the basis of household final consumption expenditure in Europe on books, newspapers and stationery. The calculation relies on several simplified assumptions, which are the following:

- The first assumption relates to the additional VAT loss arising from non-deductible input VAT by exempt sectors and public bodies. This is estimated at 20% of the VAT paid by households.

- Expenditure on physical and digital books, newspapers and stationery amounted to EUR 91 billion in 2014. By deducting 10% to exclude miscellaneous printed matter and stationery and drawing materials, we estimate expenditure on books, newspapers and periodicals at approximately 82 billion. These figures are gross of VAT. The bulk of household expenditure originates from Germany (28%), France (16%), the United Kingdom (15%), Italy (10%), Spain (5%), Poland (4%), the Netherlands (4%), Belgium (2%), Austria (2%), Finland (2%), Portugal (2%) and Sweden (2%). The share of the other Member States is much more limited and ranges from 1.4% in Denmark to 0.1% in Malta.

- Expenditure for books, newspapers and periodicals has constantly decreased during the last decade, but remained stable in 2014 for the first time. On the basis of this stabilisation, we assume for the sake of modelling that total expenditure will remain roughly constant in the future. This simplified assumption also take into account a potential increase in the number of e-publications due to lower VAT rates. Whereas, the number should increase, the expenditure should remain constant as new e-publications are expected to be less expensive than the printed publications that disappear from the market.

- With regard to e-publications, their market share of currently 5% is expected to grow to 20% in the medium term and an average VAT rate differential between the standard and the reduced VAT rate of 13.5 percentage points is taken into account.

Full details on the methodology used for the study are given in the following documents:

Final Report

http://ec.europa.eu/taxation_customs/resources/documents/common/publications/studies/economic_study_vat_on_publications_finalreport.pdf

Annex A: Member States reports

http://ec.europa.eu/taxation_customs/resources/documents/common/publications/studies/annex_a_ms_reports.pdf

Annex B: Case studies

http://ec.europa.eu/taxation_customs/resources/documents/common/publications/studies/annex_b_case_studies.pdf

Annex C: Surveys

http://ec.europa.eu/taxation_customs/resources/documents/common/publications/studies/annex_c_surveys.pdf

Abstract

The study is motivated by the Commission’s assessment of the current VAT rates structure and by the need to understand: the extent to which publications delivered on physical means of support and in electronic format should be considered as substitutable products; the impact their current VAT rate differentiation has on the market; and the potential impact of any changes in VAT in order to reduce such rate differentials.

The similarity (or dissimilarity) of products is investigated by assessing whether the products delivered through different means (print or digital) are regarded by consumers as substitutable. The underlying approach to substitutability is based on the concept of cross-price elasticity and tries to establish whether there are critical price differences that make consumers switch significantly from one product to another. An economic model populated with parameters estimated from survey data is used to simulate how potential tax changes would impact on different market outcomes (turnover and tax revenues).

In relation to the potential substitutability of print and digital publications the study concludes that: (a) print and digital publications are not part of the same relevant market; (b) consumers value print and digital products differently and this is linked to the different attributes and features of the different supports; (c) consumers are not likely to significantly substitute one product for another as a result of an impact in prices (although some migration from print to digital can be observed when the price of digital decreases, but the magnitude of such migration has been estimated as small).

In relation to the potential impacts of VAT changes the study concludes that: (a) a large decline in the sales for publications in print will result as a consequence of the equalising rates up; and a large increase in the tax revenue (in the order of between 100 and 300% for large geographical locations); (b) a moderate decline in sales for publications in print will result as a consequence of the equalising rates down; and a small loss (in absolute terms) in tax revenue (although simulations also show that some large Member State, like the UK, may have a reduction of 100% due to the complete drop in tax).

(1) See http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=URISERV%3Al31057
(2) According to Article 14 of the VAT Directive, “supply of goods” shall mean “the transfer of the right to dispose of tangible property as owner”.
(3) According to Article 24 of the VAT Directive, “supply of services” shall mean “any transaction which does not constitute a supply of good”.
(4) Member States can maintain up to two different reduced rates and may choose which of these to apply to a specific category.
(5) Exception to the general rule that standard rate is applicable.
(6) Electronically supplied services are covered by the definition of “supply of services”. Annex II to the VAT Directive gives an indicative list of electronically supplied services.
(7) COM(2016) 148 final; http://ec.europa.eu/taxation_customs/taxation/vat/action_plan/index_en.htm
(8) As part of the Regulatory Fitness and Performance Program (REFIT).
(9) Conclusions adopted by the Council at its 3468th meeting held on 25 May 2016.
(10) See chapter 9.4
(11)  See also recommendations of a report produced in the framework of the 2015-2018 Work Plan for Culture by the Working Group of EU Member States' experts on promoting reading in the digital environment:
http://bookshop.europa.eu/en/promoting-reading-in-the-digital-environment-pbNC0116151/
(12) See http://curia.europa.eu/jcms/upload/docs/application/pdf/2015-03/cp150030en.pdf
(13) This is documented in many sources. See inter alia, for a variety of countries: Communauté française de Belgique, Service de promotion des Lettres, Vademecum du livre numérique, Brussels, March 2011; Jean-François Gayrard, Comparaison entre les coûts de la chaine du livre papier et celle du livre numérique, Nov. 9, 2014, available at http://medium.com/pour-lamour-du-lire/comparaison-entre-les-couts-de-la-chaine-du-livre-papier-et-celle-du-livre-numerique-321c262d0d1d#.r078wiafr ; G. Colleu, Coûts du livre papier, coûts du livre numérique: une comparaison, March 2013, available at http://alliance-lab.org/archives/1072?lang=fr#.WAn8fHpRq_I ; New York Times, 28 February 2010, Math of Publishing Meets the E-Book. Although these studies are not new, the cost structure is a structural variable that usually evolves slowly over time, and the costs for digitalisation tend to fall over time.
(14) M. Rich, "Math of publishing meets the E-book', New York Times, 28 February 2010
(15) Marginal costs, particularly in competitive markets, tend to affect the price more readily than changes in overheads.
(16) "Economic study on publications on all physical means of support and electronic publications in the context of VAT2013" available at "http://ec.europa.eu/taxation_customs/resources/documents/common/publications/studies/economic_study_vat_on_publications_finalreport.pdf
(17) This is defined as the expected percent change in sales following a one-percent increase in price. For example, given an elasticity of -3, sales would decline by three percent for every 1 percent increase in price.
(18) Some publishers are sticking to a high-price strategy for e-books, with discounts against print editions of only 20-25% on average, but in other cases front-line fiction e-books are available for just half the price of print. This is because the publications market is highly segmented, with different supply and demand elasticities. On the latest price trends in the e-book market, see Global eBook 2015, A report on market trends and developments, Rüdiger Wischenbart et al.
(19) For periodicals and newspapers, sales volumes, or more precisely circulation, take on a special significance because the rates paid by advertisers, which make up a significant share of revenue, depend on circulation. Hence a drop in volumes has an effect that goes beyond the value of the forgone sales.
(20) A detailed description of the calculation method and assumptions is included in chapter 6.2 (option 1).
(21) Conclusions adopted by the Council at its 3468th meeting held on 25 May 2016.
(22) In this context, it is worth recalling that, according to well-established VAT case law, Member States are always free to restrict the scope of application of reduced rates, although they cannot widen it beyond the categories listed in Annex III.
(23) The more ambitious one abolishes the list of goods eligible for the reduced rate, replacing it with criteria, while the narrower one opens up the list to update it.
(24) Global eBook 2015, A report on market trends and developments, Rüdiger Wischenbart et al.
(25) Most of the data and conclusions in this section are obtained from the study "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)". http://ec.europa.eu/taxation_customs/resources/documents/common/publications/studies/economic_study_vat_on_publications_finalreport.pdf
(26) WAN-IFRA, World Press Trends 2015: Facts and Figures, available at http://www.wptdatabase.org/world-press-trends-2015-facts-and-figures . Print circulation has also fallen in the US, where the number of newspapers published has shrunk from roughly 2700 to about 2000 since 2008 (New York Times, " Yes, the News Can Survive the Newspaper", Sept. 4, 2016.)
(27) All conclusions and data in this section are obtained from the "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)".
(28) See also discussion under option 1.
(29) The impact on VAT revenues is expressed as percentage of VAT revenues accruing from each category (books, newspapers and periodicals) and based on own calculations taking into account the different level of VAT rates applied in the Member States.
(30) An overall pass-on of 50% has been estimated and justified by the "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)".
(31) See Baromètre, cit.
(32) http://ec.europa.eu/eurostat/statistics-explained/index.php/Publishing_activities_statistics_-_NACE_Rev._2
(33) More detailed Eurostat data on the turnover for each Member State for 2010 can be obtained from the study "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)" or directly from the Eurostat webpage: http://ec.europa.eu/eurostat/statistics-explained/index.php/Publishing_activities_statistics_-_NACE_Rev._2 .
(34) In addition, with regard to newspapers, the Eurostat data also include advertising newspapers, which cannot benefit from a reduced VAT rate.
(35) VAT is a consumption tax and according to Article 78 of the VAT Directive the taxable amount shall include all taxes, duties, levies and charges, excluding the VAT itself, and all other incidental expenses charged by the supplier to the consumer.
(36) By way of a simple illustrative example the following book (“Harry Potter and the Prisoner of Azkaban”) shows the following prices differences in May 2016 on www.amazon.fr Softcover: EUR 6.55, Kindle format: EUR 5.79 (12% lower).
(37) This is borne out by replies in the open public consultation. Respondents answered that, in the case of newspapers, the expected increased margins would not only lead publishers to invest in new content (42%), but also to be less dependent on advertising revenue (41%).
(38) In general, less printing on paper would also have an impact on the paper and pulp industry and eventually, at the end of the chain, the forest industry. However, less printing of newspapers and books would in the first instance reduce the demand for recycled paper.
(39) Assessing the Health of Independent Bookshops, by Stephen Heyman, New York Times, Feb. 25, 2015
(40) Heyman, op.cit.
(41) See The Death of the bookstore was greatly exaggerated, by Lev Grossman, TIME Magazine, July 11/July 18 2016, and also American Booksellers Association, Retail Sales at Bookstores Up in June, by Sidney Jarrard, Aug 15, 2016.
(42) New York Times, August 31, 2009. Are E-Readers Greener than Books? by Joe Hutsko,
(43) http://www.greenpressinitiative.org/documents/ebooks.pdf
(44) http://www.theatlantic.com/business/archive/2014/01/the-decline-of-the-american-book-lover/283222/
(45) Pew Research Centre, op. cit, p 5.
(46) Businesses have already now to determine the status of their customer (a taxable or a non-taxable person) and the place of supply (in which Member States of the EU or outside the EU) along with the standard VAT rate to be applied.
(47) This overall percentage has been estimated for the European Commission by a study by Deloitte on modernising VAT for cross-border e-commerce, which has not yet been published.
(48) To demonstrate the complexity of the issue, it should be borne in mind that printed publications will continue to be taxed as a good and e-publications will continue to be taxed as a service in the VAT Directive, independently from a change in the rules on VAT rates. That means that the place of taxation might not always the same for similar products and even when they are part of a bundle.
(49) The impact on VAT revenues is expressed as percentage of VAT revenues accruing from each category (books, newspapers and periodicals) and the estimates are based on the "Economic study on publications on all physical means of support and electronic publications in the context of VAT (2013)".
(50) By way of a simple illustrative example the following book (“Harry Potter and the Prisoner of Azkaban”) shows the following prices differences in May 2016 on www.amazon.fr Softcover: EUR 6.55, Kindle format: EUR 5.79 (12% lower).
(51) The impact would be negative in Italy and France.
(52) The impact would be negative in Italy and France.
(53) https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/vat_rates_en.pdf
(54) https://ec.europa.eu/taxation_customs/consultations-get-involved/tax-consultations/public-consultation-on-reduced-vat_en
(55) See summary report of the consultation: https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/summary_report_consultation_vat_rates_en.pdf
(56) See consultation page: https://ec.europa.eu/taxation_customs/consultations-get-involved/tax-consultations/green-paper-future-vat-towards-a-simpler-more-robust-efficient-vat-system_en
(57) See consultation page: https://ec.europa.eu/taxation_customs/consultations-get-involved/tax-consultations/review-existing-legislation-vat-reduced-rates_en
Top