EUROPEAN COMMISSION

Brussels, 30.11.2017

SWD(2017) 428 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Amended proposal for a Council Regulation

Amending Regulation (EU) No 904/2010 as regards measures to strengthen administrative cooperation in the field of value added tax

{COM(2017) 706 final}
{SWD(2017) 429 final}

TABLE OF CONTENTS

1. Introduction and policy context    

1.1. Common system of value added tax (VAT)    

1.2. Council Regulation (EU) 904/2010 on administrative cooperation and combating fraud in the field of value added tax    

2. What is the problem and why it is a problem    

2.1. The VAT gap – overall fraud assessment    

2.2. Problem tree    

2.3. Missing trader intra-community fraud (MTIC)    

2.3.1. Magnitude of the problem    

2.3.2. Problem drivers    

2.3.2.1. Weaknesses in the access and analysis of information mechanisms    

2.3.2.2. A current framework that makes it difficult to carry-out coordinated actions at EU level against fraudsters    

2.3.2.3. Insufficient multidisciplinary approach    

2.4. Fraud in relation to the dual VAT regime applicable to cars    

2.4.1. Magnitude of the problem    

2.4.2. Problem drivers    

2.4.2.1. Impaired access and mechanisms for information exchange and analysis    

2.5. Fraud in relation to the customs procedures No 42 and 63    

2.5.1. Magnitude of the problem    

2.5.2. Problem drivers    

2.5.2.1. VAT exemption is granted by customs authorities without proper checks of VAT numbers    

2.5.2.2.    Information collated by customs authorities on exempted importations is not shared with tax administrations and these transactions are not properly checked    

2.6. Evolution of the problem without action at EU level    

3. Why should the EU act?    

4. What should be achieved?    

4.1. General objectives    

4.2. Specific objectives    

4.3. Operational    

4.4. Consistency with other EU policies and with the Charter for fundamental rights    

5. What are the various options to achieve the objectives?    

5.1. Links between Problems/Drivers/Options/Sub-options    

5.2. Baseline scenario    

5.2.1. Legislative proposals with likely positive impacts on fight against VAT fraud    

5.2.2. Implementation of TNA on a voluntary basis    

5.3. Option 1: Implementation of Transaction Network Analysis    

5.3.1. Option 1.a: TNA without mandatory provision of data by all Member States    

5.3.2. Option 1.b: TNA with mandatory provision of data by all Member States    

5.4. Option2: Improving the operational framework with regard to coordinated actions of control between Member States    

5.4.1. Option 2.a: implementing joint audits    

5.4.2. Option 2.b: coordination of Member States' multilateral actions by Eurofisc    

5.5. Option 3: Developing exchange of data between Member States' tax administrations and law enforcement authorities at EU level    

5.5.1. Option 3.a: automated access to VIES data by Europol and OLAF    

5.5.2. Option 3b: Spontaneous transmission of VAT fraud cases involving at least two Member States by Member States' tax administrations to OLAF    

5.5.3. Option 3.c: access to Eurofisc data granted to Europol and OLAF    

5.5.4. Option 3.d: enrichment by and sharing with Europol of Eurofisc data    

5.6. Option 4: tackling fraud in relation to the dual VAT regime applicable to cars by improving access to car registration data    

5.6.1. Discarded option: improving automatic exchange of information between Member States    

5.6.2. Discarded option: access to the EUCARIS network without specific provisions in Regulation (EU) 904/2010    

5.6.3. Option 4.a: Access to car registration data by use of the EUCARIS network    

5.7. Option 5: fighting fraud in relation to the customs procedures No 42 and 63 by removing the shortcomings in the implementation of CP42/CP63 procedures    

5.7.1. Option 5.a: making compulsory granting an access to the register of VAT numbers to customs authorities    

5.7.2. Option 5.b: sharing CP42/CP63 data with tax authorities by way of VIES    

6. What are the impacts of the different policy options and who will be affected    

6.1. Methodology    

6.2. Analysis of the impacts of the various measures    

6.2.1. Baseline scenario: Status Quo with implementation of TNA on a voluntary basis    

6.2.2. Option 1: Implementation of a joint and automatic risk analysis system    

6.2.2.1. Common impacts    

6.2.2.2. Option 1.a. TNA without mandatory provision of data by all Member States    

6.2.2.3. Option 1.b. TNA with mandatory provision of data by all Member States    

6.2.3. Option 2: Improving the legal framework with regard to coordinated actions of control between Member States    

6.2.3.1. Option 2.a. Implementing joint audits    

6.2.3.2. Option 2.b. Coordination of multilateral actions by Eurofisc    

6.2.4. Option 3: developing exchange of data between Member States tax administrations and EU law enforcement authorities    

6.2.4.1. Common impacts    

6.2.4.2. Option 3.a. Automated access to VIES by Europol and OLAF    

6.2.4.3. Option 3.b. Spontaneous transmission of VAT fraud cases involving at least two Member States by Member States' tax administrations to OLAF    

6.2.4.4. Option 3.c. Access to Eurofisc data granted to Europol and Olaf    

6.2.4.5. Option 3.d. Enrichment of Eurofisc with Europol data    

6.2.5. Option 4: Improved access to car registration data    

6.2.5.1. Option 4.a. Access to Eucaris database    

6.2.6. Option 5: Removing existing shortcomings in the functioning of CP42/CP63 procedures    

6.2.6.1. Common impacts    

6.2.6.2. Option 5.a. Granting an access to the register of VAT numbers to customs authorities    

6.2.6.3. Option 5.b. Sharing CP42/CP63 information with tax authorities through VIES    

7. How do the Options compare?    

7.1. Summary assessment of the Impacts    

7.2. Identification of the preferred options    

7.3. Subsidiarity of the different options    

7.4. Proportionality of the different options    

7.5. Impact on SMEs    

8. How would actual impacts be monitored and evaluated?    

8.1. Indicators for monitoring and evaluation    

8.2 Monitoring structures    

9. ANNEXES    

9.1. Annex 1: Procedural information concerning the process to prepare the impact assessment report and the related initiative    

9.2. Annex 2: Consultations synopsis report    

9.3. Annex 3: Evaluation of Regulation (EU) 904/2010    

9.4. Annex 4: Consultation of Members States    

9.5. Annex 5: Consultation of stakeholders other than Member States    

9.6. Annex 6: Who is affected by the initiative and how?    

9.7. Annex 7: Methodology    

9.8. Annex 8: Additional information about the functioning of administrative cooperation in the field of VAT    

9.9. Annex 9: Transitional and definitive VAT system – Origin-based and destination-based VAT system    

9.10. Annex 10: Technical fiche on concrete examples of VAT fraud    

9.11. Annex 11: Findings and recommendations of the 2015 European Court of Auditors report: Tackling VAT fraud: more action needed    

List of figures and boxes

Figure 1: functioning of administrative cooperation in the field of VAT (source: ECA)

Figure 2: functioning of VIES (source: ECA)

Figure 3: VAT Gap as a percent of the VAT Total Tax Liability (VTTL) in EU-27 Member States, 2014 and 2013 (source CASE, 2016)

Figure 4: Problem tree: links between Drivers/Problems/Effect

Figure 5: Share of carousel fraud in VAT gap

Figure 6: Flowchart of a compliant import under CP42 (ECA, 2011, Special report no 2011/13)

Figure 7: Flowchart showing implementation flaws that may take place in a CP42 transaction chain of control

Figure 8: Links between Problems / Drivers / Solutions and Options

Box 1: MTIC fraud/carousel fraud

Box 2: Role of Europol and OLAF in fighting the most serious VAT fraud threats

Box 3: Presentation of the MTIC fraud on CO2 quotas

Box 4: New means of transport/second hand cars fraud schemes

Box 5: Presentation of a significant fraud case in relation to sale of cars

Box 6: Customs procedure 42

Box 7: Customs procedure 63

GLOSSARY

Abbreviations

Glossary of terms in their meaning within this document and for its specific purpose

1. Introduction and policy context

1.1. Common system of value added tax (VAT)

Value Added Tax (VAT) is a general tax on consumption applied to supplies of goods and services along the whole production and distribution process. It is a major and growing source of tax revenue in the European Union (EU). VAT raised slightly more than EUR 1 trillion in 2015, which corresponds to 7% of EU GDP or 17.6% of total national tax revenues 1 . One of the EU's own resources is also based on VAT (12.4% of the EU budget in 2015) 2 . As a broad-based consumption tax, it is considered to be one of the most growth-friendly forms of taxation.

One of the key strengths of VAT is that, by allowing taxpayers to exactly offset the tax incurred in previous stages of the production chain, it is much better suited than other types of indirect taxes to operate an single market free of tax distortions. This was the main reason for its early adoption by the EU. It is governed by the VAT Directive 3 which aims at ensuring that the principles underlying the functioning of this tax apply consistently in all Member States.

In recent years, however, the VAT system has been unable to keep pace with the challenges of the global economy and the opportunities offered by new technologies. Therefore, the Commission adopted on 7 April 2016 an Action Plan on VAT 4 (hereinafter "Action Plan") setting out ways to modernise the VAT system so as to make it simpler, more fraud-proof and business-friendly. In this context, the Commission announced its intention to adopt in 2017 four VAT-related proposals:

·a definitive VAT system for intra-EU cross-border trade based on the principle of taxation in the Member State of destination in order to create a robust single European VAT area (first step);

·a modernised VAT rates policy so as to allow Member States greater autonomy on setting the VAT rates;

·a comprehensive simplification VAT package for SMEs;

·a proposal to enhance VAT administrative cooperation and E urofisc.

As agreed by the European Parliament and the Council, the definitive VAT system will be based on the principle of taxation in the Member State of destination. In order to allow for a soft transition for tax administrations and businesses, this change will be implemented through a gradual approach 5 .The Commission envisages three successive legislative steps to implement it fully. Each step will require an unanimous agreement in Council. Therefore the final features of the definitive VAT system are unknown at this stage and it is a long term project.

As indicated in the Action Plan and as required by the Member States and the European Parliament, in the meantime, Member States need to adopt urgent conventional measures to contain the scale of cross-border fraud. This impact assessment relates to the possible measures that the Commission has examined with the stakeholders to enhance VAT administrative cooperation and Eurofisc in order to meet this requirement.

1.2. Council Regulation (EU) 904/2010 on administrative cooperation and combating fraud in the field of value added tax

Along with the abolition of the physical borders within the EU and the implementation of the ‘transitional arrangements’ for intra-EU trade 6 , a framework allowing Member States to cooperate to better manage and control VAT was put in place by the end of 1992 7 . With the introduction of Council Regulation (EU) No 904/2010 8 of 7 October 2010 (hereinafter ‘Regulation (EU) 904/2010’), the first objective was to gather, in a single piece of legislation, all provisions in relation to administrative cooperation in the field of VAT. On top of that, and in response to the most severe and always sophisticated VAT frauds, improving this administrative cooperation framework overall was also viewed as a necessity.

The current functioning of administrative cooperation in the field of VAT, as provided by Regulation (EU) 904/2010 can be summarised as showed below:

Figure 1: functioning of administrative cooperation in the field of VAT (source: ECA)

The first purpose of Regulation (EU) 904/2010 currently in force remains the exchange of information between Member States tax authorities with three different and complementary types of programmes: on request, spontaneous and automatic. To get information on businesses involved in intra-EU transactions, tax authorities can also rely on VIES 9 , an IT tool developed as of 1992.

Figure 2: functioning of VIES (source: ECA)

Certain information has also been made available to private stakeholders through ‘VIES on the web’ where VAT identification numbers delivered by Member States' tax administrations can be checked on-line.

Tax administrations are now well experienced in processing these different categories of exchanges and using VIES. Regulation (EU) 904/2010 also provides for other cooperation tools i.e. simultaneous controls and presence in administrative offices and during administrative enquiries. This Regulation has also introduced a new means for an enhanced cooperation, namely Eurofisc whose characteristics are presented in annex 8, where the functioning of administrative cooperation provided by Regulation (EU) 904/2010 through the different instruments made available to tax administrations is also summarised. Statistics in relation to the use of Regulation (EU) 904/2010 are presented in the Evaluation Report drafted to feed into the work on the possible measures to amend this Regulation. This Evaluation Report is presented in annex 3 to this Impact Assessment.

2. What is the problem and why it is a problem

2.1. The VAT gap – overall fraud assessment

The size of the VAT fraud is difficult to measure in itself. One of the most commonly accepted indicators used pointing to the scale of the problem is the ‘VAT gap’. It estimates the overall difference between the expected VAT revenue and the amount actually collected. It should be noted however that the VAT gap provides an estimate of revenue loss due not only to fraud and evasion, but also weaknesses in national tax collection systems, lack of compliance 10 of taxpayers leading to a shadow economy, impediments in the enforcement of the tax obligations by tax authorities, for instance a lack of tax audits carried out by tax administration, insolvencies as well as miscalculations and other irregularities. Despite its imperfections, the VAT gap offers a useful indicator to assess the size of VAT that is not collected by Member States.

According to the latest Commission's report on the VAT gap 11 which refers to data from 2014, the overall EU VAT gap is estimated at almost EUR 160 billion revenue losses each year The VAT gap varies considerably across Member States. The smallest gaps are observed in Sweden (1.24%), Luxembourg (3.80%), and Finland (6.92%). The largest gaps are registered in Romania (37.89%), Lithuania (36.84%) and Malta (35.32%). Overall, half of the EU-27 Member States represented in Figure 1 below 12 record a gap below 10.4% while the EU average registered in 2014 reached 14%. 

Figure 3: VAT Gap as a percent of the VAT Total Tax Liability (VTTL) in EU-27 Member States, 2014 and 2013 (source CASE, 2016)

When analysing the phenomena contributing to the VAT gap, it can be observed that some of them – notably the weaknesses in national tax collection systems, impediments in the enforcement, or lack of compliance – fall under the direct responsibility of the Member States that must design their own legislation and enforce it at national level. The Commission can however play a role by supporting Members States in their continuous efforts to improve the situation at national level or by providing platforms where dialogues between tax administrations and exchange of good practices can take place. Enhancing the administrative capacity of tax authorities is for example one of the objectives and priorities of Regulation (EU) 1286/2013 13 . The Action Plan adopted by the Commission on 7 April 2016 14 also includes actions with respect to tax administration or compliance.

There are however other areas – notably with respect to cross-border transactions and, linked to them, different types of fraud – where, in addition to measures taken at national level, it is necessary to put in place cooperation mechanisms at the EU level as fighting cross-border fraud requires coordinated actions within and between Member States. This has been recognised in particular in the VAT Action Plan which set out enhancing administrative cooperation between Member States by supporting the sharing and joint analysis of information as one of the key objectives in the short term.

There are three main types of cross-border fraud which are most widespread and most significant across the EU:

·carousel fraud (or missing trader intra-community fraud – MTIC fraud);

·fraud related to trading of second hand cars; and

·fraud related to the customs procedures 42 and 63.

As these types of fraud are distinct, exhibit different characteristics and are driven by - mostly - different factors, they are analysed individually as three main problems this initiative aims to tackle, followed subsequently by distinct sets of options addressing each of them.

Over the course of this review of the current administrative cooperation instruments , an evaluation of Regulation (EU) 904/2010 was undertaken. Its comprehensive results are presented in annex 3. This evaluation shows that beyond the question of VAT fraud and the availability of appropriate instruments to combat it, other problems also take place in the application of Regulation (EU) 904/2010. Amongst other things, the following shortcomings were highlighted by the Member States:

- high number of late replies from other Member States to requests for information;

- difficulties in using e-forms to exchange information;

- perceived administrative burden; and

- relative accuracy of VIES data.

All these shortcomings are well-known and documented. The Commission has been working with the Member States to improve the situation. However, amending Regulation (EU) 904/2010 would not bring any added-value in this respect as the instruments themselves are appropriate. It is their implementation in certain Member States that needs to be addressed. This explains why these issues are not taken into account in this impact assessment.

2.2. Problem tree

Figure 4: Problem tree: links between Drivers/Problems/Effects

2.3. Missing trader intra-community fraud (MTIC)

VAT fraud is a phenomenon which is targeting inherent shortcomings of the EU VAT system and, in general, involves a non-payment of output VAT or an overstatement of input VAT. There are a number of different kinds of VAT fraud. One of the most known types of fraud in B2B 15 transactions is the Missing Trader Intra-Community (MTIC) fraud or 'the carousel fraud'.

MTIC fraud occurs in many different ways and the schemes become more difficult every time and include both goods and services. What they have in common is that they exploit situations in which goods and/or services can be bought free of VAT. MTIC fraud systematically involves a defaulting trader, literally a trader that does not submit its VAT returns and does not pay the VAT due to the tax authorities. The major characteristics of an MTIC fraud are described in box 1 below.

Although MTIC fraud is first driven by the endemic weaknesses of the EU VAT system and weak tax administrations capacities in some Member States, these two specific drivers will not be addressed by the present initiative as they are out of the scope of Regulation (EU) 904/2010. It is nevertheless worthwhile to mention them for the sake of the analysis comprehensiveness.

2.3.1. Magnitude of the problem

Out of the EU VAT gap estimated at almost EUR 160 billion revenue losses each year, EUR 50 billion 19 would be due to carousel fraud alone. In a 2013 report Europol estimated the MTIC losses at EUR 100 billion. 20 Although it is difficult to precisely assess the weight of this phenomenon because it corresponds by essence to tax revenue not collected by Member States tax authorities, studies trying to tentatively assess its magnitude conclude that it is significant.

According to a study carried-out by EY in 2015, the carousel fraud portion of the VAT gap ranges from 12% in Bulgaria to 39% in France (see figure 3 below on share of carousel fraud in VAT gap).

Figure 5: Share of carousel fraud in VAT gap (source "Own calculations" based on EY, 2015 study) 21

2.3.2. Problem drivers

2.3.2.1. Weaknesses in the access and analysis of information mechanisms

Each Member State collects and stores information on its own trader population. MTIC fraud is by its nature cross-border. Relying exclusively on national information does not allow tax administrations to efficiently detect and investigate such type of fraud. Therefore cross-border information exchange is crucial to monitoring the correct application of VAT on cross-border transactions and effective fight against fraud. Regulation (EU) 904/2010 sets out a specific framework for administrative cooperation and combating fraud in the field of VAT as described in section 1.2. and in Annex 8.

The VIES databases, requests for information and multilateral controls do not constitute the best instruments to combat MTIC/carousel fraud in the field of VAT: this fraud needs to be quickly detected and these communication channels do not allow Member States to get all necessary information to be able to prevent or put an end to VAT fraud. This is precisely why Eurofisc, designed as a multilateral early warning mechanism to improve their administrative capacity in combating organised VAT fraud and especially MTIC/carousel fraud, was set up in 2010 under Regulation (EU) 904/2010. Thanks to Eurofisc and by way of risk analysis tools, Member States can exchange multilaterally early warnings on businesses suspected to be involved in MTIC/carousel fraud. Eurofisc is a Member State-driven network, composed of national risks analysts working in different Working Fields per fraud risk area. The network is financed under the European Fiscalis 2020 Programme 22 . The Commission provides administrative and logistical support to Eurofisc and does not have access to operational data. Nevertheless, after 7 years of practice, it appears that Eurofisc has not so far met all expectations because of:

a- a cumbersome and lengthy process of information exchange, which can be described as follows:

·As a result of domestic risk analysis Member State 1 selects a potentially risky trader and puts it under monitoring. Member State 1 uploads on CIRCABC 23 a spreadsheet which includes intra-EU deliveries / invoices issued / Vies-on-the-WEB 24 data of the potentially risky trader;

·Member State 2 – a Member State where the potentially risky trader is registered – checks on him. Based on the results of the checks, Member State 2 provides a qualification 25 on the potentially risky trader and uploads a spreadsheet on CIRCABC to this end;

·Member State 1 checks CIRCABC if the potentially risky trader was qualified. If so, then this qualification can be used for domestic risk analysis and follow-up.

b- instruments that are not adequate as they do not permit to exchange vital information on risky traders or transactions immediately. The current set-up of Eurofisc is challenged by the following:

·limited scope of exchanged data, largely attributed to the limitations arising from the manual work that needs to be carried out for data sharing;

·slow processes to request and retrieve targeted data;

·poor targeting of traders to be put under monitoring;

·lack of prioritization of suspicious cases based on risk assessment;

·no common risk analysis standard or common EU framework for risk analysis;

·issues of data quality and accuracy;

·risks of data losses.

All these problems were emphasised by the European Court of Auditors (hereinafter "ECA") 26 in its 2015 Report, which concluded that Eurofisc is a mechanism that is time consuming, labour intensive, with slow and not always well targeted exchanges made with non-user friendly instruments. All this significantly decrease the value of information exchanges while swiftness is key to fight MTIC fraud.

These shortcomings were also highlighted by the Member States in the course of Regulation (EU) 904/2010 evaluation. Member States recognise that Eurofisc has had positive effects on the fight against VAT fraud and the collection of VAT on intra-EU transactions and report a high level of satisfaction. Nevertheless, they also mention the same drawbacks in its functioning as the ECA did and indicate that, due to them, Eurofisc has certainly not reached its full potential yet.

2.3.2.2. A current framework that makes it difficult to carry-out coordinated actions at EU level against fraudsters

Regulation (EU) 904/2010 provides Member States with the possibility to agree to conduct simultaneous controls or so-called ‘multilateral controls’ (MLCs) of the tax liability of one or more related traders, if they consider such controls to be more effective than controls carried out by only one Member State. In addition, the Member States can agree that foreign tax auditors be present during administrative enquiries.

These instruments are considered useful by Member States and can be used as flexible means to exchange targeted information in a swifter and more coordinated fashion than exchanging information on request. Nevertheless, these tools are first of all designed to exchange information but cannot be considered as instruments allowing tax administrations to coordinate their audits:

·although presence of foreign officials in administrative enquiries (be it in the framework of a MLC or not) is possible, this does not enable them to actively participate in enquiries carriedout in other Member States;

·MLCs are carried out by separate audit teams in each country. Results are shared thereafter by way of spontaneous exchanges of information. This makes it difficult to have a single strategy and to correctly coordinate actions of all national authorities involved in such simultaneous controls and operating according to their national guidelines.

The ECA 2015 Report as well as the Commission Report to the Council and the European Parliament on the application of Regulation (EU) 904/2010 27 both reported that all Member States considered MLCs as a useful tool for combating fraud, including VAT fraud. Some evidence from the case studies conducted under the final evaluation of the Fiscalis 2013 programme 28 and the MLC reports also proved its effectiveness in actually helping to identify fraud. According to data collected from closed MLCs for which a report was provided (85%), these led to the identification of additional tax revenue with a value of approximately EUR 3.26 billion.

Nevertheless, the number of multilateral controls initiated annually is stable, at around 40. While searching for possible causes, some were offered by the Member States and included aspects such as: overall (very) slow character of the instrument; long duration of the MLCs (most lasting more than one year, making it difficult to be used to tackle the most serious fraud in an efficient fashion); difficulties with inserting MLCs initiatives in established annual audit planning programmes; extra workload for local officers combined with lack of experiences; or difficulty in convincing management that it was worthwhile investing in audits that may only show a benefit for the other Member States. All this was highlighted again by Member States over the course of Regulation (EU) 904/2010 evaluation (see annexes 3 and 4).

The Commission Report to the Council and the European Parliament 29 also mentioned that better coordination at both national and EU levels would be valuable, in particular by reinforcing ties between national anti-fraud units and Eurofisc. National anti-fraud units are at the forefront in fighting VAT fraud and are equipped with - often unique - legal measures to do so (e.g. right to search or seize). Despite a strong expertise and the fact that Eurofisc liaison officials are the first to be aware of new fraud schemes and would consequently be in the best position to dismantle these fraudulent networks by joining their efforts, Eurofisc does not play any role in coordinating multilateral controls at EU level.

2.3.2.3. Insufficient multidisciplinary approach

According to the ECA and Europol, the most damaging VAT frauds are committed by organised crime groups (OCGs) through MTIC schemes. They benefit from their international criminal structures and connections to establish efficient MTIC schemes to extort money from national budgets. The ECA and Europol estimate that EUR 40-60 billion of the annual VAT revenue losses are caused by organised crime groups and that 2 % of those groups are behind 80 % of the MTIC fraud.

The proceeds of MTIC fraud are usually reinvested in new criminal activities 30 or laundered. This is corroborated by the real examples of MTIC fraud presented in annex 10. The figures provided by Europol during the public hearing before the PANA Committee on 14 November 2016 confirmed a link between companies appearing in the Panama Papers and ongoing investigations for MTIC fraud 31 . In the same hearing Eurojust mentioned that many of its proceedings on money laundering were connected with VAT fraud.

Tax authorities are not equipped to identify these criminal organisations and administrative measures against such serious VAT fraud cannot be effective. A multidisciplinary approach involving law enforcement and judicial bodies is more effective to:

·identify OCGs behind VAT fraud;

·investigate; and

·prosecute.

Such were also the conclusions of the ECA 2015 Report. Most Member States are already taking this approach to a certain extent at national level. Tax authorities involve law enforcement bodies to stop criminals behind VAT frauds. In some Member States, tax authorities even received law enforcement powers which simplified the process.

However, according to the 2017 SOCTA report made by Europol 32 70% of OCGs are active in more than 3 countries and 45% of them are involved in more than one type of criminal activity. Therefore, the multidisciplinary approach should become international as well. Most of Member States are already taking actions in this field. 27 Member States participated in the EMPACT MTIC priority 33 that was created after the SOCTA 2013 report 34 . The temporary activity was driven by Member States with a support from Europol. It allowed launching several operations during the period 2013 – 2017 against criminal organisations active in MTIC fraud.

Despite national initiatives towards a multidisciplinary approach, there are still fields that could be improved. In particular, and as confirmed by the Member States in their replies to the targeted consultation made in the course of the Evaluation Report (see annexes 3 and 4), there is no exchange of VAT information and of intelligence on organised crime structures involved in serious VAT fraud and coordination of actions between, on the one hand, tax administrations and Eurofisc and, on the other hand, other law enforcement authorities at EU level such a Europol and OLAF – see box 2 below. A systematic cross-checking of information on missing traders between Eurofisc and Europol databases on OCGs is also absent.

The VAT fraud on CO2 quotas, as detailed below in box 3, that caused EUR 5 billion of revenue losses, can be cited as an example illustrating MTIC fraud in the light of a lack of multidisciplinary approach. Although Eurofisc could now help to avoid such frauds to happen, this case demonstrates a lack of cooperation and connections between all authorities in charge of fighting VAT fraud. This shortcoming was also stressed in the ECA report where it was recommended to improve the cooperation and remedy overlapping competences of administrative, judicial and law enforcement bodies.

2.4. Fraud in relation to the dual VAT regime applicable to cars

The sale of cars is also an economic sector recognised as heavily affected by VAT fraud. This situation is mainly the result of two set of rules provided for in the VAT Directive.

In relation to new means of transport 36 , including cars, and unlike the general rule for the cross-border delivery of goods between businesses and consumers 37 , new means of transport acquired by non-VAT liable persons, are subject to VAT in the country where goods are delivered 38 .

The rationale of this rule is that the car industry is not equally widespread across Member States, while consumers are located throughout the entire EU territory. The taxation of purchases of cars in the country of the supplier as it is the case in B2C transactions would attract VAT revenue in countries where cars are produced or with low standard rates and, reciprocally, result in significant VAT losses in the country of their use i.e. where consumers live. The high value of these goods and VAT revenues associated made it even more difficult for Member States to accept taxation in the Member State of the supplier.

In relation to second-hand means of transport as defined in Article 327 (1) of the VAT Directive, two different VAT regimes exist:

·on the one hand, transactions on second-hand means of transport are not subject to VAT in the country where goods are delivered. Instead, VAT is calculated on the margin realised by the supplier and must be paid in the country where this supplier is located;

·on the other hand, sales of second-hand cars for which the right to deduct the input VAT has been exercised by the owner are subject to VAT on the full transaction amount. In that case, VAT must be paid in the country where goods are delivered, as it is the case for new means of transport 39 .

It results from this that trading in cars is often used to commit VAT fraud, the easiest way being to sell recent or new means of transport taxable on the whole amount as if they were second-hand goods, taxable on the margin only. The modus operandi on car fraud is illustrated in Box 4 below. A typical example of this kind of fraud is also detailed in annex 10.

2.4.1. Magnitude of the problem

There is no study specifically targeting the fraud of new and second-hand means of transport so the hard data to assess the weight of this fraud in the overall VAT gap is scarce. However, the amount of data referring to cars supplies exchanged under Eurofisc Working Field 2 40 is significant: the monitored transactions that is, the total amount of transactions that were suspected by Member States to be fraudulent and needing further checks by tax authorities, amounted to EUR 5 billion in 2013, EUR 3 billion in 2014 and EUR 1.5 billion in 2015. Although the trend shows a decrease in the total amount of transactions under monitoring due to better targeting methodologies, this amount remains high and demonstrates that the selling of cars is a sector subject to predominant concerns at EU level.

Fraudulent trade into cars causing significant VAT losses and distortion of competition have also been reported by compliant car resellers. The high-value of recent cars gives a major advantage to fraudulent actors that sell these vehicles without collecting VAT on the total amount of transactions. Compliant businesses have no other appropriate means to combat these distortions than decrease their mark-ups, when possible, or to sell goods at prices not reflecting the real costs. This significantly hampers the proper functioning of the single market and does not place all economic actors on an equal footing.

For example, according to the Fédération Nationale des Artisans de l’Automobile (FNAA) in France which bases its estimations on customs data, the damage caused by the fraudulent trade to cars retailers represents a loss of EUR 5 billion annually, while EUR 1 billion is lost in VAT by the French Treasury, for a total of 160 000 cars sold on the French market.

This situation directly results from the VAT regime in relation to means of transport which was described above, and some loopholes in the exchange of information under Regulation (EU) 904/2010.

2.4.2. Problem drivers

2.4.2.1. Impaired access and mechanisms for information exchange and analysis

Due to the free movement of persons and goods within the EU, international cooperation is crucial to efficiently monitor the VAT regimes that apply to transactions on cars. A prerequisite to exchange information on suspicious trade of cars is to know the Member State that must receive the information, leading to a two-step exchange: in first instance, an exchange of information to determine the Member State where the car, the sale of which is under monitoring, was registered and, thereafter, a second exchange of information allowing the relevant Member State to check the correct application of VAT.

One shortcoming reported by Member States is that there are, to date, no suitable administrative cooperation instruments to collect identification or registration records on vehicles in an efficient fashion. The main inefficiencies that occur are:

·in relation to exchange of information on request, requests are handled manually on a case-by-case basis, without the possibility to receive a swift answer because of not well targeted procedures;

·in relation to automatic exchanges, these exchanges provided for in Article 13 of Regulation (EU) 904/2010 in conjunction with Article 2 of Commission Implementing Regulation (EU) No 79/2012 42 are not mandatory 43 and when they occur, the information provided is difficult to process quickly and efficiently;

·in relation to Eurofisc, it is not possible in all instances to quickly determine the Member State to whom crucial information must be directed.

Outside the legal framework of Regulation (EU) 904/2010, dedicated public or business organisations do keep information in relation to car registration in each Member State. Information maintained in these registers is already exchanged between Member States for purposes specifically defined in several pieces of legislation 44 by way of a European platform named EUCARIS.

EUCARIS started in 1994 as a co-operation among national registration authorities from five European countries to fight international vehicle crime and driving licence tourism by means of exchanging vehicle and driving licence information between its members. All EU Member States are now EUCARIS members, as well as Iceland, Norway, Switzerland, Jersey, Isle of Man and Gibraltar.

The organisation of EUCARIS is based on the EUCARIS Treaty and its underlying Rules of Procedure. The General Assembly appoints the EUCARIS Secretariat and the responsible party for the execution of the operational system management, which are chosen from among the Member States. At the moment the Netherlands Vehicle Authority (RDW) is responsible for both the EUCARIS Secretariat and EUCARIS Operations .

EUCARIS is an information exchange system that provides an infrastructure and software to countries to share, among others, their car and driving licence registration information helping fight car theft and registration fraud. EUCARIS is developed by and for governmental authorities and is able to support all kinds of transport related information exchange based on treaties, directives, bi- and multilateral agreements.

This platform allows Member States to:

·have an automated access on real time to information kept in the national repository of the requested authority;

·send out bulk requests to identify where a specific means of transport has been registered.

To this end, a client application which has been developed by the EUCARIS Secretariat, is distributed in each Member State and can be used for specific purposes such as enforcement of road offences.

2.5. Fraud in relation to the customs procedures No 42 and 63

The customs procedures No 42 and 63 (CP42 and CP63) are two VAT regimes provided for by Article 143(1) of the VAT Directive that allow for a VAT-free importation of goods in a Member State if it is followed by an exempted B2B supply or transfer to another Member State. This simplification has been implemented to allow for transit of Community goods without imposing on traders an unnecessary VAT burden.

In principle, an import of goods should be subject to VAT and this input VAT reported and offset in the VAT return of the importer. In normal scenario the importer will sell the goods in the same country enabling a compensation of import VAT (input VAT) with the VAT on the sales (output VAT). However, in case the importer does not have VAT taxable transactions in the country of importation compensation of input VAT with output VAT is not possible. The importer will need to request a reimbursement of the VAT to the tax authorities and is then supporting the burden of financing VAT while in the end no VAT is due in the Member State of importation. To compensate this situation, a VAT exemption on importation of goods that are transiting to other Member States has been introduced. It improves the cash flow situation of businesses and reduces their administrative burdens.

The functioning of these procedures is shown in the flowchart below.

Figure 6: Flowchart of a compliant import under CP42 (ECA, 2011, Special report no 2011/13) 45

To differentiate such VAT exempt importation from others, customs authorities have established two specific customs procedures namely CP42 and CP 63 46 :

These two procedures are widely used. There were 8.5 million importations under VAT exemption in 2015, representing a total value of EUR 74 billion. In terms of value, 60% of these importations were concentrated in Belgium (EUR 33.1 billion for 570 000 transactions) and Germany (EUR 17.3 billion for 707 000 transactions), followed far behind by France, Austria, the UK and Sweden. 17 Member States did not exceed 12 000 of such customs procedures in 2015. In some Member States, like the Netherlands, they are not used at all as the importers make use of the possibility to defer the payment of the import VAT until the submission of their VAT return 48 . With respect to the countries of origin, the top of the list is dominated by China, Switzerland, the USA and Japan. If one looks at the Member States of destination to where the goods imported under CP 42 are subsequently transported, depending on the criteria (value or number of items) one can find France, Germany, the UK, the Netherlands and Italy.

If the benefit for legitimate business to use CP42 and CP63 cannot be denied, the risk of VAT fraud linked to these importations cannot be ignored as well. Based on the information available, it would appear that the main VAT fraud linked to CP42 and CP63 is avoiding paying VAT and selling the goods in the shadow economy. Fraudsters selling goods on the black market in the EU would be looking for any ways to bring goods into the single market at the lowest possible costs. This includes reducing the customs duties by under valuating and avoiding VAT at importation with the use of CP42 and CP63. The last OLAF investigation (see below section 2.5.1.) and ECA reports confirm that these two phenomena are closely connected.

2.5.1. Magnitude of the problem

No precise assessment of the level of fraud committed by misuse of the CP42 and CP63 is available. In its report published in 2011 on the control of CP42 49 in seven Member States 50 , the ECA mentioned that for year 2009 and by extrapolation, the level of VAT losses in relation to the CP42 only would approximately reach EUR 2.2 billion. For this specific year, this represented 29% of the VAT theoretically applicable on the taxable amount of all the imports made under the CP42. This estimation was made before the EU legislation in relation to this procedure was amended, providing for stricter application conditions, although these new rules do not seem to always be consistently applied (see below, section 2.5.2.1). In its report released in 2015 51 , the ECA did not further estimate the level of this fraud. Nevertheless, several recommendations were made, demonstrating that defects in the application of this procedure still remain.

Recently, in March 2017, OLAF imposed a fine of EUR 2 billion on the UK for uncollected customs duties on under-evaluated imports of footwear and textiles through the UK during the period 2013-2016. Imports were done mainly with CP42 and under VAT exemption. According to OLAF, goods were sold by criminal organisations on the black market in several Member States. If this were the case, VAT losses for these transactions would amount to several billion of Euro, the estimation of VAT losses being at that early stage, only an extrapolation based on the findings of customs operations 52 . Another concrete example of CP 42 fraud involving traders in China, the UK and France is also presented in annex 10.

It is also possible that exempted imports, instead of being dispatched to the supposed country of destination, are in fact locally sold free of VAT. It is also possible for these goods to be involved in a carousel fraud with a supposed recipient in the country of destination that is, in fact, a missing trader.

This fraud, like carousel and fraud on cars, heavily distorts competition within the single market as goods can be introduced VAT free and sold at underestimated prices. VAT fraud can then be combined with customs duties fraud as the goods are also often strongly under-evaluated. Some EU producers and retailers in certain economic sectors 53 are more heavily affected and could not face competition with non-EU operators in a position to sell under-priced goods in the single market.

2.5.2. Problem drivers

Already in 2011, the ECA report on the control of CP42/CP63 54 in seven Member States 55 pointed out that the control process of the CP42/CP63 was deficient and recommended improving the communication of key data within and between Member States, encouraging the automatic verification of VAT identification numbers, creating a common EU risk profile for imports under CP42/CP63 and encouraging action on a proposed amendment to the VAT Directive. Answering this last recommendation, the general conditions for the exemption were added to the VAT Directive in 2009 under Article 143(2) (see below). Goods that are imported onto the EU territory under CP42/CP63 should at least be subject to the following checks:

·VAT numbers of the importer and customers;

·evidence that goods are intended to be transported to another Member State.

In practice, as showed in flowchart 5 below, several implementation flows in the CP42/CP63 chain of customs controls and in the exchange of information were identified at national level:

·these preliminary checks are not in all instances carried out by customs authorities upon import, as detailed under 2.5.2.1;

·tax authorities in the Member State of import are not always notified that goods have entered the EU territory VAT free, preventing them to further check whether information reported upon import, on the one hand, and information reported by the importer to tax authorities upon submission of its recapitulative statement, on the other hand are consistent and if the goods have actually been dispatched to another Member State (see 2.5.2.2);

·information on CP42/CP63 stays in the country of importation. Only the information on the subsequent intra-EU supply is shared with the second Member State (if the importer reports it correctly in the recapitulative statements); However, the link with the importation is lost. This prevents tax authorities in the Member State of destination to properly cross-check this information with that provided by the recipient of the goods (see 2.5.2.2).

2.5.2.1. VAT exemption is granted by customs authorities without proper checks of VAT numbers

As mentioned above, CP42 and CP63 can be granted upon provision of the valid VAT numbers of the importer and customers. Registers of VAT numbers are usually maintained and kept up-to-date by tax authorities. Access to all VAT identification numbers delivered by Member States tax authorities and registration data concerning enterprises to which these VAT identification numbers have been allocated is granted to all Member States through VIES 56 (see annex 8). It appears that, in practice, national databases as well as access to VIES are not made available to customs authorities in certain Member States. As a result, customs authorities are not, in all instances, in a position to check on the spot if VAT numbers declared by the importer are valid.

It can therefore well be the case that the recipient of goods does not have any valid VAT number. In such instance, there is high risk that goods will be sold on the black market in the country of importation or any other Member States without VAT being paid in both cases.

This problem was stressed by the Commission in its 2014 report 57 . It results from this assessment that a significant number of Member States failed to check systematically the validity of VAT identification numbers (BE, BG, FR 58 , HU, IE, LU, NL, PT, and UK) at the moment of importation. This was reported again by the ECA in its more recent report dated 2015.

2.5.2.2.    Information collated by customs authorities on exempted importations is not shared with tax administrations and these transactions are not properly checked

Once CP42/CP63 is approved by the customs authorities, tax administrations should take over the responsibility for following up the physical flow of goods: onward supply to and acquisition in another Member State. To this end, cross-checks between customs data on imports under CP 42/CP63, VAT recapitulative statements and return submitted by the importer, and the VAT return of the recipient are key for ensuring that onward supplies of goods do not remain untaxed in the territory of the Member State of consumption of the goods or in another Member State.

However, both tax administrations of the Member States of importation and of destination do not systematically receive from customs authorities, information on imports made under CP42 and CP63 despite the risks inherent to these procedures. In addition, traders are not obliged to flag separately or identify the intra-EU supplies following these imports in their VAT recapitulative statements as it is the case for triangular transactions.

As a result, tax authorities can only rely on the compliance of the importer to be informed. In any event, if reported, the information is transmitted at the earliest one month after the importation and the tax administration cannot see that the reported intra-EU supply was preceded by a risky exempt importation. Therefore, tax administrations are not aware that such risky transactions have taken place and cannot check if goods have been actually dispatched to another Member State and if VAT has been paid by the recipient.

Recommendations in relation to these procedures were made again by the ECA in its more recent report released in 2015 where it is stressed that there are no effective cross-checks between customs and tax data in most of the Member States visited during the course of the audit. It is then very easy for fraudulent stakeholders to use the weaknesses of the system to introduce VAT free goods onto the market. In addition, lengthy procedures to carry-out all necessary checks in all Member States concerned once the fraud took place allow the fraudster to be active for a long time before it can be identified and deregistered by tax authorities. And even in that case, the fact that customs authorities will not, in all instances, check VAT numbers upon importation will not make sure that fraudsters will be eliminated.

Figure 7: Flowchart showing implementation flaws that may take place in a CP42 transaction chain of control

2.6. Evolution of the problem without action at EU level

As mentioned in this report, the level of VAT fraud remains significant within the EU 59 . The recent legislative proposal submitted by the Commission to the Council 60 , which echoes the request from certain Member States to be allowed to apply a generalised reverse charge mechanism (GRCM) as an urgent measure to combat MTIC and carrousel fraud, reveals how the shortcomings of the current transitional VAT system can severely affect certain Member States. It also reveals the limits of conventional measures to combat such fraud. This is corroborated by recent reforms implemented in several Member States (e.g. split payment 61 introduced in Italy in respect of certain transactions, compulsory electronic invoicing transiting via the tax administration in Portugal) with a view to improving VAT collection.

In the longer term, it is expected that the definitive VAT system based on the taxation of intra-EU transactions will efficiently curb MTIC and carousel fraud across the EU by addressing the root of the problem. Such a reform has the potential according to a recent study 62  to increase tax revenues by about EUR 40 billion per year. This proposal is being tabled in 2017 but will take several years to be agreed upon and fully implemented.

Without immediate action at EU level, the weakness of the current transitional VAT system will continue to be exploited by fraudsters to commit MTIC and carousel fraud. Fraud levels might be stable but this will be to the detriment of compliant businesses that will pay the price either through stable compliance costs (compliance costs associated with cross-border VAT obligations are already 11% higher than for domestic trade) or even extra compliance costs (with additional reporting obligations or audits or the management of a dual VAT system with the introduction of the generalised reverse charge in certain Member States).

The complexity of the current transitional VAT system without physical borders will continue to negatively impact the functioning of the single market by failing to capture new business models, new markets and technologies and by translating into losses of competitiveness for honest EU businesses and efficiency losses for Member State tax administrations.

At the same time, with a VAT system organised at EU level, administrative cooperation between Member States remains necessary to ensure that VAT legislation can correctly be implemented and enforced in each Member State. It is also important to make sure that unreliable stakeholders do not take advantage of cross-border situations and do not gain from the fact that weaknesses in streams of information between Member States create risks in domestic situations. Only administrative cooperation between Member States organised at EU level can establish the necessary conditions to make the VAT system robust and efficient.

3. Why should the EU act?

As demonstrated under the problem analysis, the root causes behind the three types of fraud partly lie within the VAT framework. It explains why the Commission has tabled proposals to change the system itself. Instruments already available to Member States to resolve or at least overcome weaknesses inherent to the VAT system – amongst others through administrative cooperation within and between tax administrations – are not powerful enough to prevent the EU-wide fraud.

However as mentioned earlier on, to prepare the implementation of the definitive regime and, pending its implementation, to contain cross-border fraud, in the VAT Action Plan of April 2016 the Commission proposed actions on three fronts, one of them being achieving better administrative cooperation. Already the ECA in its 2015 report on the VAT fraud recognised that tools for administrative cooperation available to tax administrations were not being sufficiently exploited, or ambitious enough.

Views from the Member States, expressed over the course of the evaluation report presented in annex 3 have confirmed the need to improve the current administrative cooperation instruments by removing their shortcomings, but also to reinforce them by making for instance a better access and use of already existing information.

It all led the Commission to commit to endeavour shifting from the existing cooperation models based on Member States exchanging information, to new models of sharing and jointly analysing information and acting together, allowing Member States benefiting from a risk management capacity at EU level, as well as enabling them to rapidly and more effectively identify and dismantle fraudulent networks. In addition, new and always more sophisticated fraud schemes require innovative solutions. In the field of VAT, with fraud being cross-border in the most severe instances, it tends to belong to no-one while affecting everybody. Having this in mind and recalling the earlier described swiftness with which fraud takes place, any solution can be better implemented, effective and efficient when done jointly at EU level.

The Commission identified altogether 20 measures to tackle the VAT gap 63 . The present initiative will fit under a considerable number of them, including:

·the possibility of extending the use of automated access to data and exploring with Member States the possibility to develop an automated mechanism that would allow a cross-matching between the data reported by each party of every single transaction. That would allow detecting fraud in early stages and ultimately prevent a missing trader fraud, be it domestic or intra-EU;

·reinforcing the role and impact of Eurofisc on tackling intra-EU VAT fraud, including making better use of the information available within the network;

·evaluating Regulation (EU) 904/2010 and proposing legal and operational solutions to address the weaknesses of the current legislation, including by introducing joint audits;

·supporting deeper cooperation between different authorities.

The management, collection and control of VAT are first and foremost a national competence of the Member States. However, VAT fraud is often linked to cross-border transactions within the single market or involves traders established in other Member States than the one where the tax is due. Moreover, VAT fraud has a negative impact on the functioning of the single market and causes serious losses to Member States’ revenues and consequently to the EU budget.

Therefore pursuant to Article 113 of the TFEU 64 the EU has implemented cooperation tools organising in particular an exchange of information between tax administrations and supporting common audit activities and the Eurofisc network.

Improving the efficiency of these tools, in particular strengthening Eurofisc, would offer value over and above what could be achieved at Member State level.

4. What should be achieved?

The objective of the initiative is to reduce the level of cross-border VAT fraud that takes place within the EU. To this end, the initiative will provide tax administrations with more efficient and effective instruments to fight the three main sources of VAT fraud taking place in cross-border situations i.e. MTIC fraud, fraud on sale of second-hand cars and fraud in relation to Customs procedures No 42/63. A better use and sharing of information between authorities and the reinforcement of Eurofisc will enable Member States' tax administrations to better target fraudsters, to put an end to fraudulent schemes at an earlier stage and ultimately to ensure a level playing field for businesses.

4.1. General objectives

The general objectives of the initiative are:

·to contribute to fiscal consolidation within the EU by ensuring that taxes due are collected to feed national and EU budgets ;

·to contribute to fighting organised crime; and

·to contribute to a closer cooperation between Member States in the EU VAT area, moving away from exchange of information only to a model based on data sharing and joint actions.

4.2. Specific objectives

The specific objectives of the initiative are:

·to better exploit the existing administrative cooperation instruments in the field of fighting VAT-related fraud;

·to contribute to fighting organised crime through rapid and more effective identification and dismantling of fraudulent networks related to VAT; and

·to improve the multidisciplinary approach to fighting and preventing VAT-related fraud through swifter and more coordinated reaction capacity.

4.3. Operational

·Faster information exchange and joint processing of data related to VAT by Eurofisc officials;

·Improve identification and targeting of potential fraudsters;

·Provide for new/improved channels for access to and sharing of VAT–related information between tax administrations and other authorities or institutions;

·Improve effectiveness of checks and the sharing of VAT-related information in the context of imports; and

·To facilitate joint audits.

Linking the objectives to the problems

4.4. Consistency with other EU policies and with the Charter for fundamental rights

The objective of fighting tax fraud and evasion to help secure national and EU revenues and prevent distortion of competition has been amongst top Commission priorities for the last few years. The political guidelines 65 of the present Commission called for stepping up the efforts to combat tax evasion and tax fraud, including through improved administrative cooperation between tax authorities. These priorities were directly reflected in the already mentioned 2016 VAT Action Plan. More recently, the creation of a simple, modern and fraud-proof VAT system was one of the fiscal priorities set out by the Commission for 2017 (Annual Growth Survey 2017 66 ).

This initiative goes hand in hand with other initiatives in the VAT area and plays an important role in securing the success of the most ambitious proposals, which is the introduction of the definitive VAT regime. These several initiatives were already discussed before in this impact assessment. Additionally, MTIC fraud is also one of the nine EMPACT priorities 67 , the European Union’s priority crime areas, under the 2014-2017 EU Policy Cycle of Europol. The same approach has been followed for the next policy cycle.

The impacts on fundamental rights of the different options that are considered in this impact assessment are further detailed under Chapter 6. In short, it is expected that this proposal will trigger new exchange and joint processing of VAT information which could include personal data. Nevertheless, these new developments will ultimately be subject to Article 8 of the Charter for Fundamental Rights and, once introduced in Regulation (EU) 904/2010, Paragraph 5 of Article 55 of that Regulation that mirrors Directive 95/46/EC 68 on the protection of individuals with regard to the processing of personal data and on the free movement of such data 69 .

5. What are the various options to achieve the objectives?

5.1. Links between Problems/Drivers/Options/Sub-options

As detailed under section 2 of this impact assessment, there are three main areas where cross-border VAT fraud schemes are developed. These three distinct problems, and their underlying factors, need to be addressed in different ways. Improving the instruments of Regulation (EU) 904/2010 may be the solution to resolve some of them. For others, the solution may be found in pieces of legislation that are out of the scope of the proposal this impact assessment relates to. They nevertheless needed to be mentioned to give a precise overview of the problems and provide better understanding of the issue at stake.

When designing the different options presented in this impact assessment, diverse views were gathered. Most of them result from discussions between the Commission and tax administrations in fora where VAT fraud issues and administrative cooperation matters are regularly discussed, such as the Eurofisc network meetings, the Anti-Tax Fraud Strategy Group, the Standing Committee on Administrative cooperation or Fiscalis 2020 activities. The ECA 2015 report on VAT fraud 70 also suggested different ways to improve administrative cooperation, in particular by involving law enforcement authorities to better combat the most severe types of fraud. Finally, over the course of the evaluation of Regulation (EU) 904/2010, opinions of the Member States and of other stakeholders on these options were also sought 71 . All this is further detailed below. 

The links between the problems and their drivers previously described, on the one hand, the solutions and options that could be developed to address them, on the other hand, are detailed in figure 6 below. The baseline scenario has not been reported in this figure although it is further described under section 5.2 and constitutes the framework against which all options will be assessed.

Figure 6: links between Problems / Drivers / Options/ and Sub-options

5.2. Baseline scenario

The baseline reflects the current legal framework inclunding all ongoing initiatives which have a sufficient degree of certainty to be in a position to weight and assess them against all options that will be developed in this impact assessment.

There are on the current agenda of the Commission several legislative proposals whose objectives are to improve the functioning of the EU VAT system, to make it stronger and remove some of the weaknesses described in this impact assessment.

However, the precise characteristics of some proposals are yet unknown, some others envisage a progressive implementation and all of them would need to be agreed upon unanimously by all Member States after including possible far-reaching amendments. Hence, the baseline scenario that will serve as the benchmark against which the other options will be assessed will not include these legislative initiatives. They are nevertheless presented to give a complete overview of the current situation (section 5.2.1. of this impact assessment). In parallel, the Member States and the Commission have been working on the implementation of the Transaction Network Analysis on a voluntary basis. Since this instrument will be implemented whichever the results of the Council discussions on amending Regulation (EU) 904/2010, it has been considered as part of the baseline scenario.

5.2.1. Legislative proposals with likely positive impacts on fight against VAT fraud

The Action Plan adopted in April 2016 by the Commission sets out the objectives and measures envisaged by the Commission to modernise the EU VAT system. The creation of a robust single European VAT area is one of the key actions announced in the Action Plan. It will require the setting up of the definitive VAT system for B2B cross-border trade, moving away from the origin-based system towards a destination-based system with the taxation of intra-EU trade in the country of destination (please see Annex 9 for better understanding of the transitional system). The 'destination principle' has been agreed by the European Parliament 72 and the Council 73 .

After having examined in detail the possible options for implementing the destination principle, the Commission announced its intention (see Action plan), and the Council took note of it, to present a proposal for the definitive VAT system that consists in taxing all intra-EU transactions (while still tracking the flow of the goods), just like it would be the case for domestic transactions. This option was assessed as the best solution to tackle MTIC fraud as well as decrease the complexity of the system and thus the compliance costs for all stakeholders.

Although the preliminary results on the above-mentioned taxation option are very promising, preparing and adopting such a major change is likely to take some time. Indeed, the Commission envisages three successive legislative steps to implement it fully. Each step will require an unanimous agreement in Council. Therefore the final features of the definitive system are unknown at this stage and it is a long-term project.

In addition, the definitive regime will not completely address frauds in relation to second-hand cars and customs procedures 42 and 63. Actually, transactions taking place under these regimes will still be subject to specific VAT regimes that could be exploited by fraudsters to evade VAT. In this context, administrative cooperation measures to combat these frauds will still be needed whether the definitive regime is implemented or not.

New types of frauds could also take place once the definitive VAT regime is implemented. For instance, fraudsters could exploit differences in VAT rates between Member States to create new forms of missing trader frauds. If this were the case, Member States could use the cooperation instruments which are considered here to combat these new fraud schemes.

Considering the urgency to take robust measures to improve the current situation when it relates to VAT fraud, the Commission has tabled four proposals that are currently being discussed at the Council level:

·a proposal for a temporary Generalised Reverse Charge Mechanism 74 . The current VAT system is based on fractioned payment ensuring that the VAT is collected at each stage of the production and distribution chain. Under this system, the supplier is the person liable for payment of VAT to the tax authorities. Under a reverse charge mechanism, the person liable for payment of the VAT to the tax authority is the acquirer. The proposal under consideration aims at, under certain conditions, extending the reverse charge mechanism to all B2B supplies of goods and services above a certain threshold that take place on a domestic market to prevent missing trader VAT fraud to take place;

·a proposal for a Directive modernising VAT for cross-border B2C 75 e-commerce 76 .
E-commerce has become a key part of the economy and an important driver of economic growth. The proposal aims at solving several concerns in relation to e-commerce, in particular losses in VAT revenues for Member States. It is expected that a change in the small consignments exception rule for imports and the distance sales thresholds, along with an extension of the Mini One Stop Shop, may be solutions to tackle these increasing VAT losses;

·the future entry into force of the Directive on the fight against fraud to the Union's financial interests by means of criminal law which will cover serious EU-wide VAT fraud 77 ; and

·the new legislation establishing the European Public Prosecutor's Office. On 8 June 2017, 20 Member States reached a political agreement on its establishment under enhanced cooperation. It will be an independent and decentralised prosecution office of the European Union with competence for investigating, prosecuting and bringing to justice crimes against the EU budget, such as fraud, corruption, or cross-border VAT fraud above EUR 10 million damage.

All these initiatives along with the new instruments that could be implemented by amending Regulation (EU) 904/2010 would constitute ways to immediately improve the situation with regard to VAT fraud. It would also bring tools that could be further used and reinforced to combat new fraud schemes. Nevertheless, even though all these initiatives are successfully implemented, the defintive regime will remain necessary to put an end to the current endemic weaknesses of the transitional arrangements and reduce compliance costs for businesses engaged in intra-EU trade within the single market.

5.2.2. Implementation of TNA on a voluntary basis

Member States, on a voluntary basis,on the basis of the positive outcomes of a pilot project led by Benelux countries, have decided to develop a tool for information exchange and joint processing of VAT data for Eurofisc officials called Transaction Network Analysis (hereinafter ‘TNA’) 78 . TNA is being implemented so it is considered as part of the baseline scenario.

TNA is highly targeted to how Eurofisc officials already exchange information. It will not substantially change the "what", but improve the "how" by addressing the following points:

·automating the collection of targeted information, e.g. VIES;

·improving the detection by introducing advanced data analytics;

·being able to visualize suspicious networks without manual intervention. In the future Member States will be monitoring the whole chains with the same information being available to all participating Member States;

·improving Member States ability to share data and qualify traders identified by Eurofisc by making them less dependent upon manual uploads of spreadsheets, resulting in a more structured and fluid way of information exchange;

·improving the quality of information shared by enhancing IT facilities for collecting and sharing the information. That would allow Member States to better allocate resources on which cases to investigate

As with Eurofisc, participation in TNA is voluntary for Member States as this is merely a tool to facilitate the exchange of information. Most of the Member States so far supported the development of TNA and 25 countries have joined the project as full members, two countries being observers. This means that TNA will be able to access their data to identify potential fraudsters and their Eurofisc officials will have direct access to TNA to investigate identified traders and provide feedback. However it is envisaged that Member States not joining TNA will also benefit from the development as they will receive an extract of the results of TNA analysis through usual Eurofisc channels.

It is expected that this new instrument hosted by the Commission will provide a new framework allowing Member States to better and in a very efficient fashion target fraudsters so that tax authorities can quickly react to put an end to VAT fraud schemes. This voluntary TNA will be implemented in 2018.

In the answers to the public consultation, 47 respondents reported that a joint processing of data would be helpful to fight VAT fraud and 36 considered that Eurofisc would be the right structure to coordinate this new instrument.

In order for TNA to reach its full potential and bring all its positive effects, all Member States should join the project and grant TNA access to their VIES data. Nevertheless, one Member State requires a clearer legal framework in order to participate.

5.3. Option 1: Implementation of Transaction Network Analysis

Two alternative options can be drawn up. The introduction of an explicit legal base for TNA is envisaged under each of them. The only difference between them relates to the access by TNA to the VIES data which is necessary to identify potential fraudsters.

5.3.1. Option 1.a: TNA without mandatory provision of data by all Member States

In contrast with the baseline scenario this option would introduce an explicit legal base for TNA in Regulation (EU) 904/2010. However, this option would not require that the Member States which would not participate in TNA would have to grant it access to their VIES data.

5.3.2. Option 1.b: TNA with mandatory provision of data by all Member States

Under the baseline scenario and option 1.a, non-participating Member States would not be required to grant TNA access to their VIES databases. As a consequence, TNA would not be able to track network ramifications and relationships in all Member States. This would hamper the performance of the tool.

Under option 1.b, as under option 1.a, an explicit legal base would be introduced in Regulation (EU) 904/2010. However, all Member States, regardless of their participation in TNA, would be required to grant TNA access to their VIES data on intra-EU transactions in order for the application to identify potential fraudsters established in all Member States.

5.4. Option2: Improving the operational framework with regard to coordinated actions of control between Member States

Two sub-options or variants that can be implemented separately or combined with each other have been identified: the implementation of joint audits and the coordination of Member States' multilateral actions by Eurofisc. When asked about the usefulness of joint audits, Member States reported that there would be an added value in implementing this new instrument: 15 strongly agree or agree that it would be a useful means to audit cross-border activities, 14 that it would be cost effective for tax administrations and 15 for businesses, and 15 that it would provide legal certainty to businesses. At the same time, their replies also showed that at the moment, the administrative cooperation instruments that score the best are coordinated actions of control under the form of simultaneous controls or presence during administrative enquiries.

Stakeholders other than Member States are balanced when it comes to the added-value of joint audits with 19 agreeing that an active participation of foreign tax officials would be valuable, 17 disagreeing and 19 without any position. At the same time, 40 of them recognise the added value of having a single report audit to provide taxpayers with more legal certainty. In addition, 43 respondents reported a benefit from having cross-border administrative enquiries of fraudulent network being coordinated at Eurofisc level.

5.4.1. Option 2.a: implementing joint audits

Today, to coordinate their actions, Member States can engage in:

·simultaneous controls, where the tax authorities of several Member States carry out audits in their respective country and thereafter exchange the information collected by way of spontaneous exchange of information; or

·administrative enquiries where foreign tax auditors can be present in other tax administration's offices and taxpayers' premises during national audit activities. However, foreign auditors are not allowed to directly carry out these activities.

The joint audit is an instrument whereby a taxable person is subject to a coordinated audit with the following features:

·two or more countries join together to form a single audit team to examine issue(s) / transaction(s) of one or more related taxable persons (both legal entities and individuals) with cross-border activities, including cross-border transactions involving related affiliated companies established in the participating countries;

·in which participating countries have a common or complementary interest;

·the taxpayer shares information with the single audit team and all participating countries at once; and

·the team includes competent officials 79 from each country. No individual (bilateral) requests for information between the auditors are needed because the information is shared within the mandate for the joint audit between all members of the single audit team.

Introducing the concept of joint audit would be a step forward in achieving closer cooperation between tax administrations in controlling cross-border supplies, by removing legal obstacles taking place in simultaneous audits and presence of foreign auditors during administrative enquiries. In a joint audit, with two or more countries joining together to form a single audit team, the joint audit team carries out a single audit i.e. the examination of books and records. To this end, tax auditors are allowed to actively participate in the audit (with powers of inspection) held in the jurisdiction of another participating Member State.

5.4.2. Option 2.b: coordination of Member States' multilateral actions by Eurofisc

Antifraud units are at the forefront in fighting serious VAT fraud and within these units, the information gathered and analysed by Eurofisc officials is used by auditors carrying out cross-border audits (currently under 'simultaneous controls' and 'presence in the office/participation in the administrative enquiries'). Eurofisc officials have to transfer the relevant information to national auditors, within their tax administration. It is then up to these auditors to consider if the information is relevant enough to propose the initiation of a simultaneous control following the procedures of the MLC-platform. The audit units in the tax administrations coordinate the cross-border administrative enquiries/controls.

Another domain in which Eurofisc could take a more active role is the coordination of cross-border administrative enquiries (simultaneous controls or joint audits) of possible fraudsters identified by Eurofisc officials. Taking into account that Eurofisc officials are often the first to be warned about new fraudulent networks and they have a strong expertise in serious VAT fraud, they would be the best placed to coordinate the corresponding necessary investigations and administrative enquiries in the Member States involved. This would also allow better utilising the result of the TNA and the information from the audits could be immediately processed with TNA.

Such coordination would be targeted to fraudulent networks identified within Eurofisc and requiring quick reaction. That means that only a fraction of MLCs would be covered. In practice, coordination would be carried out in the working field having the relevant expertise by one or several Eurofisc officials of the Member States involved in the audits.

5.5. Option 3: Developing exchange of data between Member States' tax administrations and law enforcement authorities at EU level

Member States use more and more multidisciplinary approaches against serious VAT fraudsters by involving national law enforcement bodies or participating in EMPACT 80 . This national effort could be enhanced at EU level by involving Europol and OLAF in this multidisciplinary approach. To date, no formal exchanges of information are organised between Member States' tax authorities and these two organisations. As reported by the ECA in its 2015 report, this "reduces Europol and OLAF's ability to tackle VAT fraud through the identification and disruption of organised crime groups behind the carousels and even their ability to assess the real impact of intra-EU VAT fraud".

The targeted consultation whose results are presented in annex 4 shows that Member States exchange information with law enforcement authorities in a rather limited number of instances and that the vast majority of them have no experience at all in exchanging information with Europol and OLAF (19 out of the 27 replying Member States seem to have no exchange of information with these two enforcement authorities at EU level). The vast majority of them also demonstrate a limited appetite to develop such exchanges in particular in the light of the pending questions regarding the powers of law enforcement authorities to investigate VAT fraud .

Stakeholders other than Member States see more interest in these sharing of information with 34 respondents being of the view that cooperation between tax authorities and law enforcement bodies should be changed.

Four sub-options or variants which can be combined with each other have been identified.

5.5.1. Option 3.a: automated access to VIES data by Europol and OLAF

As recommended by the ECA 81 , the first option would be to grant an automated access to VIES data to Europol and OLAF. The information that would be made available covers:

·recapitulative statements from businesses making intra-EU supplies. These statements list the aggregate value of supplies of goods and services made to VAT registered customers elsewhere in the EU;

·VAT numbers and details of enterprises registered for VAT purposes within the EU.

This option should be analysed in conjunction with Recommendation 14 of the ECA Report 24/2015 where it is mentioned that the European Parliament and the Council should:

(a) include VAT within the scope of the proposed directive on the fight against fraud to the Union’s financial interests by means of criminal law (PIF directive) and the Regulation on the establishment of the European Public Prosecutor’s Office; and

(b) grant OLAF clear competences and tools to investigate intra-EU VAT fraud.

Currently, VAT-based revenue is part of the EU’s own resources. However, the definition of the notion of ‘irregularity’ provided for by Council Regulation (EC, Euratom) No 2988/95 on the protection of the European Communities financial interests only applies to infringements affecting own resources collected directly on behalf of the Communities. Therefore, there are still pending questions regarding the possibility for OLAF to conduct on-the-spot checks in the field of VAT.

5.5.2. Option 3b: Spontaneous transmission of VAT fraud cases involving at least two Member States by Member States' tax administrations to OLAF

Under the second option tax authorities would spontaneously provide OLAF with information on serious VAT fraud cases involving at least two Member States.

As reported under section 5.5.1., OLAF does not have the tools to conduct investigations in the field of VAT successfully on its own. However, OLAF has, in the past, when information was shared, supported multilateral investigations of serious VAT fraud by offering coordination and facilitation. The opening of such coordination cases became rare when the sharing of information was framed legally in Regulation (EU) 904/2010 and access to Eurofisc was limited to Member States administrations.

5.5.3. Option 3.c: access to Eurofisc data granted to Europol and OLAF

Under this option an access to data held by tax authorities within Eurofisc would be granted to Europol and OLAF. Having regard to the remit of these two bodies, links with Eurofisc can be made since they are all involved in the fight against the most severe tax fraud threats. Hence, access to targeted Eurofisc data could be considered when it relates to organised criminality, including:

·information on carousel fraud, including Eurofisc data (traders under monitoring, qualifications, feedbacks etc.) and TNA results;

·information on CP 42 & CP 63 fraud schemes; and

·information on VAT fraud trends.

Considering their scopes of intervention, it seems that Europol is more likely to be interested by information on organised crime, in particular, carousel fraud, while OLAF has probably an interest in receiving information at a broader spectrum. Nevertheless, the intention, here, is not to segregate the information that could be accessed by the two bodies based on their activities, but to allow them to receive the same level of information.

In practice, this information is today exchanged by way of CIRCABC. An access to the restricted CIRCABC fields where this information is made available to Eurofisc officials could be granted to new Eurofisc officials designated at Europol's and OLAFs levels.

5.5.4. Option 3.d: enrichment by and sharing with Europol of Eurofisc data

Considering its scope of activities, Europol holds information on criminal organisations that could be useful to fight serious VAT fraud in particular to Eurofisc liaison officials.

Another option would be to allow Eurofisc officials to provide Europol with information so Europol could enrich Eurofisc data pieces with its own information and facilitate the analysis carried out by Eurofisc liaison officials and the administrative enquiries they could coordinate. Europol could also share the information from Eurofisc with Member States law enforcement authorities. According to Europol rules, it would be up to Eurofisc officials to decide whether the information can be shared within Europol and with whom.

5.6. Option 4: tackling fraud in relation to the dual VAT regime applicable to cars by improving access to car registration data

Enhancing the exchange of information on car registrations between Member States would constitute a useful means to help national tax administrations. This was in particular reported by the Member States over the course of the targeted consultation with 20 Member States being of the view that car registration data should be added to the list of categories for automatic exchange of information and 19 that data on car owners should be exchanged over VIES. With regard to other stakeholders, out of the 49 respondents, 29 consider that tax authorities should exchange data on car registration. To this end, three alternative options/variants have been identified.

5.6.1. Discarded option: improving automatic exchange of information between Member States

As detailed above under section 2.4.2.1., automatic exchange of information on new means of transport already exists between Member States. Improving these exchanges could also be considered as a means to answer tax administrations' needs. In this case, different shortcomings that exist today in the automatic exchange of information would be addressed:

·automatic exchange of information on new means of transport would become mandatory for all Member States, which is not the case so far as seven Member States do not exchange such information 82 ;

·more frequent exchange of information would be introduced, since information exchanged within three months after the quarter when it has become available, as it is currently the case, does not permit to have information on real time.

However, this option is unlikely to suit needs expressed by Member States in relation to access to car registration data. The objective, here, is to be able in a very efficient fashion to query all Member States about where a specific vehicle is registered to further direct fraud signals or information. Improving automatic exchange of information would, on the one hand, improve the availability of information in relation to cars. On the other hand, it would not allow Member States to be able to know where a car is registered, except if all Member States forwarded all their car registration data and if at national level each Member State put in place necessary development to implement this function. This is clearly not proportionate to Member States' needs and this option must then be discarded.

5.6.2. Discarded option: access to the EUCARIS network without specific provisions in Regulation (EU) 904/2010

As detailed under section 2.4.2.1., dedicated public or business organisations do hold information in relation to car registration in each Member State and exchange such information through EUCARIS. An option would be to allow tax authorities to have a direct access to the EUCARIS Web Client Application.

Nevertheless, in this situation, legal issues would be raised as, under the current legal framework, tax administrations are excluded from the possibility to directly use the EUCARIS network or to have access to its data, even indirectly 83 .

A new EUCARIS Treaty providing for exchanges of information for purposes other than preventing, investigating and prosecuting offences in relation to cars with other administrative bodies was signed on 8 June 2017. Once this treaty is in force, exchange of information for tax purposes will be possible. However, a 2-3 years of ratification process can be expected and only 8 EU Member States are signatories to this Treaty. This means that under such scenario these specific tax administrations may only have access to the EUCARIS data of Member States participating in that multilateral agreement (all data of other EUCARIS countries would be excluded from the access). This would constitute a significant loophole and would definitely not answer Member States' needs.

5.6.3. Option 4.a: Access to car registration data by use of the EUCARIS network

To solve the above-mentioned legal issue, the automated access to relevant data stored in the national repositories of vehicle registration authorities could be provided by a specific provision added to Regulation (EU) 904/2010.

In practice, the EUCARIS network would be accessible by way of a client application to be developed to suit Member States' needs and either provided at EU level or distributed in each tax administration. Under this alternative, functionalities would be the same as under section 5.6.2.

As variant, as it is already the case as regards certain existing VIES data, it could be envisaged that only Eurofisc officials could be granted automated access to the information.

5.7. Option 5: fighting fraud in relation to the customs procedures No 42 and 63 by removing the shortcomings in the implementation of CP42/CP63 procedures

Improving exchanges of information between tax and customs authorities is seen by the Member States as a suitable means to be more efficient in tackling VAT fraud in relation to CP42/CP63 procedures. In their answers to the targeted consultation – see annex 4 – 19 Member States reported that it would be useful to exchange data on imports made under CP42/CP63 procedures over VIES.
24 out of 27 also mentioned that customs authorities should have an access to VIES data, which was reported as being already the case in 23 Member States. These views were also for a large extent reiterated by the Member States during the Anti-tax Fraud Strategy meeting that took place on 10 July 2017.

Stakeholders other than the Member States also demonstrated an interest in these sharing of information since 41 out of the 58 respondents to the public consultation consider that tax administrations should have automated access to information on exempt importations from customs authorities. The following options can be considered when it comes to fighting fraud in relation to CP42/CP63 procedures.

5.7.1. Option 5.a: making compulsory granting an access to the register of VAT numbers to customs authorities

Checking the validity of the VAT numbers of the importer and the final customer by customs authorities before approving the customs declaration under CP42/63 is necessary to check the application of the exemption and to prevent VAT fraud. Accessing VIES data is key to carry out these checks.

Under this option, access to VIES data, in particular VAT registration information would be granted to customs authorities in all Member States. Granting access to customs authorities would not be decided at national level in each Member State.

5.7.2. Option 5.b: sharing CP42/CP63 data with tax authorities

As described before, one weakness of CP42/CP63 procedures is the slowness and complexity of the verification of the entire process despite the risk of fraud occurring quickly. The tax authorities in the Member States of import and of destination have to wait for the recapitulative statement of the importer, which is often not submitted, to carry out these checks. The tax authorities in the Member States of the final customer are not informed that there is an additional risk in relation to specific intra-EU supplies as such transactions are not flagged in the recapitulative statement.

To remedy this implementation flaw, the relevant information on CP42/CP63 (i.e. VAT numbers, value of the imported goods, type of commodities) which is submitted electronically with the customs declaration would be shared either immediately or on a regular basis by the Member State of importation with the tax authorities of the Member States of destination (determined on the basis of the VAT number of the customer reported on the customs declaration). This sharing of information could be done via a new database accessible through CCN/CSI.

Such a measure would allow the Member States of importation and of destination to detect suspicious transactions much faster and stop the fraudsters much sooner. Tax authorities of both countries would be able to cross-check this information with that reported by the importer in its recapitulative statement and VAT return and with that declared by the recipient of goods in its VAT return. In addition, if the Member State of destination detects that the VAT number of the customer declared to the customs authorities, albeit valid, has been hijacked (i.e. the identified business is not the real customer – that can hardly be detected by the customs authorities), it could immediately inform the Member State of importation so it can control the importer.

In addition to this measure, as it is already the case as regards existing VIES data, all Eurofisc officials could be granted automated access to the whole database to detect fraudulent chains operating in several Member States in particular by using TNA (Transaction Network Analysis).

6. What are the impacts of the different policy options and who will be affected

6.1. Methodology

The revision of Regulation (EU) 904/2010 takes place in the context of the Action Plan 84 presented by the EU Commission in April 2016 and further endorsed by the EU Council in May 2016 85 and the EU Parliament in November 2016 86 . No specific tools or methodology were developed for the assessment of the options. No independent study could be launched either. An evaluation of Regulation (EU) 904/2010 has been undertaken on the basis of, amongst other things, a consultation of Member States, an open public consultation of private stakeholders in the first half of 2017, statistics provided by Member States and other evidences available to the Commission. This evaluation is presented in annex 3.

In addition, several reports on the implementation of the current VAT administrative framework were presented and have been used when designing the different options and comparing them : see Commission report on the functioning of Regulation (EU) 904/2010 made in accordance with Article 59 of the Regulation 87 , the Commission report underArticle 12 of Regulation (EC) No 1556/1999 88 , the March 2015 ECA special report on the effectiveness of the EU in tackling intra-EU VAT Fraud 89 and an upcoming report in accordance with Article 12 of Regulation (EC) No 1556/1999.

All other evaluations, studies and consultations that have been used to assess the different options are further described in Annex 7.

6.2. Analysis of the impacts of the various measures

The different options that have been presented under section 5 of this impact assessment are expected to swiftly and efficiently improve the situation when it relates to VAT fraud. In most instances, these new instruments will improve the use and dissemination of already existing information on the one hand, and relationships between administrations and authorities involved in the fight against VAT fraud on the other hand. Additional burdens that may be imposed either on national administrations or other stakeholders are rather limited or non-existent as regards the latter, while positive effects on the level of VAT fraud and the functioning of the single market can be expected. All this is further detailed below.

Social impacts are difficult to assess. It could be argued that, indirectly, eliminating fraudulent businesses from the market will level out the playing field for legitimate businesses, improve their competitiveness and promote their growth, which in turn may shift employment upwards. It is nevertheless impossible to assess with any level of credibility the extent of such an impact.

Similarly, one could expect elimination of fraud to improve the health, safety and security of products. Goods in the fraudulent circulation are more likely to be of unknown provenance and thus escaping standards, norms and procedures they should normally meet. Therefore, reducing their presence in the market by dismantling some of the fraudulent schemes could indirectly improve the real or perceived uncertainty as to products' quality and reduce the risks involved.

Consumer prices may be somewhat affected by these measures. The fight against fraud will eliminate businesses that abuse the VAT system and can, consequently, propose under-priced goods to consumers. Although prices may be negatively impacted, this is the result of a correct application of the rules currently framing the intra-EU VAT system and this will be to the profit of the correct functioning of the single market and contribute to tax fairness that is in the interest of all citizens.

Indirectly, a well-functioning taxation system has a stronger distributive role; taxes evaded by a business or an individual benefit only to them directly whereas overall fair taxation system, with high compliance and low level of irregularities, converts the revenues into (quality) public goods of benefit to all citizens. Conversely, a tax system where VAT can be avoided incentivises businesses and individuals to follow suit into the 'profiteering' approach, which ultimately erodes the market equality and the citizens' trust in the system as a whole.

Finally, as fraudulent VAT schemes are often linked to organised crime with the proceeds of MTIC fraud usually reinvested in other criminal activities, the adoption of a common and multidisciplinary approach to tackle intra-EU VAT fraud and a better targeting of MTIC fraud will help eradicating the roots of these criminal networks.

In terms of economic impacts, only indirect consequences could be mentioned here but neither their extent nor the causality relationship with the initiative can be credibly assessed. At the micro-level, as already mentioned above, eliminating fraudulent businesses from the market would improve the economic and competitive standing of legitimate businesses, prompting them to grow, invest and employ. Overall, impacts on legitimate businesses are considered overly positive with no significant additional costs of adjustment, compliance or any other administrative burden. The conduct of the legitimate businesses is neither targeted nor expected to change, if excluding reduced mitigating measures to counter fight VAT fraud in any given business, which may potentially reduce some operational costs and smoothen the operations.

Where pertinent, alongside these general impacts, some more specific effects on businesses coming from the individual policy options, including on administrative burdens, are described further on in greater detail, although no quantitative data are available to specifically support the analysis.

At the macro-level, similar growth-inducing effects are normally expected from the well-functioning taxation system, which generates resources being partially and indirectly returned to the economic sectors through infrastructure and services, legal certainty of the enforceability of the tax rules and ultimately mutual trust. There exists unfortunately little research material on the causal links between the level of fraud and its impact on the economic growth in general. Malfunctioning tax system however, is by all considered as eroding the basic market principles and further dis-incentivising any compliance.

Positive impacts on the functioning of the single market are expected from the different options that are considered in this impact assessment. Cross-border VAT fraud cannot be tackled by a single administrative cooperation measure and all options will help national tax administrations to eradicate VAT fraud schemes roots, prevent the fraud to happen and rebuild the conditions of a level playing field for compliant enterprises across the EU. Negative effects of these schemes that have been described under Chapter 2 of this impact assessment (in particular on enterprises and on prices) will be reduced. These compliant enterprises will benefit from a more efficient and less distorted single market.

Ex post audit and investigation, while important, is unlikely to forestall considerable loss of revenue, because the essence of the fraud is that money is made quickly, in the time gap before the missing trader is required to remit the VAT it has supposedly charged on its sales. Once the money has disappeared into the complex web of transactions, tracing and recovering unjustified VAT refunds becomes time-consuming and costly. Preventing fraud to happen by jointly identifying potential fraudsters very early is therefore key.

As the objective of this initiative is to reduce VAT fraud through improved cooperation between Member States, it is in that area that the impacts of the initiative are expected to be the most significant and are analysed below in detail for each policy option. Since fighting VAT through better cooperation across the EU necessitates some coordination, organisational and, albeit to a lesser extent, financial inputs, these are also part of the more extensive analysis below. Therefore, the detailed analysis of the impacts of the policy options concentrated on the following specific aspects:

·impact on compliance costs to business, in particular SMEs;

·impact on administrative costs to the tax administrations;

·impact on VAT fraud;

·impact on fundamental rights.

No significant operational and implementation risks have been identified in relation to the options described in Chapter 5 and whose impacts are detailed under this chapter. Two may potentially take place. It is first possible that tax administrations do not develop on time or do not update when needed the IT components necessary to implement certain measures. This is typically the case for options 4 and 5 where national authorities will have to develop IT programmes or functions to provide and exchange information in relation to car registration data and customs procedures 42 and 63.

A second risk that has been identified is the possibility that some instruments made available to national tax authorities are finally not used as much as expected or that their potentialities are not fully exploited. Since Regulation (EU) 904/2010 proposes a variety of tools without - for the majority of them - imposing to the Member States to use them, this risk may happen. This can be the case because of a lack of available human resources or because Member States are reluctant to use some instruments or see them as useless (for example participation in joint audits)

Nevertheless, combating VAT fraud is a constant concern for all Member States, meaning that they will certainly use all these new possibilities if their impacts on national revenues are positive and quickly felt by the tax authorities. For instance some Member States were initially reluctant to the implementation of the Transaction Network Analysis and have finally accepted to join the project. In sum, and even though the risk that some instruments are not used is not negligible, it remains rather limited.

6.2.1. Baseline scenario: Status Quo with implementation of TNA on a voluntary basis

6.2.2. Option 1: Implementation of a joint and automatic risk analysis system

6.2.3. Option 2: Improving the legal framework with regard to coordinated actions of control between Member States

6.2.4. Option 3: developing exchange of data between Member States tax administrations and EU law enforcement authorities

6.2.5. Option 4: Improved access to car registration data

6.2.6. Option 5: Removing existing shortcomings in the functioning of CP42/CP63 procedures

7. How do the Options compare?

7.1. Summary assessment of the Impacts

The tables below analyses and evaluates the various impacts of the different options.

The main elements of each option have been weighed (0: no significant, + positive impact, - negative impact) according to their respective expected impacts. These elements are the following:

·impact on compliance costs of business, in particular SMEs;

·impact on administrative costs of the tax administrations;

·impact on VAT fraud;

Table 1: Summary assessment of the options

Table 2: Comparison table

7.2. Identification of the preferred options

This impact assessment has presented three different problems that amendments to Regulation (EU) 904/2010 could address. All these problems are driven by different factors which would require separate measures to remedy them. Therefore, it is not possible to compare the five options previously described against each other and to identify a best option since they do not address the same problems and drivers. All options are worth considering and could have a positive impact on VAT fraud. However, some seem to be more effective for each driver. After a careful analysis, the Commission concluded that at the analytical level, none of the policy options would establish a clear advantage over the others. Therefore, benefits and risks have been outlined for each of the options in order to prepare the ground for a decision at the political level. The same holds true with respect to sub-options which do not always address the same shortcomings.Nevertheless, several considerations should be taken into account by policy makers at the time they decide which of these options should be retained or not:

·each option addresses one of the five specific drivers identified in this impact assessment. This means that at least one sub-option should be considered for each option and that at least five different measures should be retained;

·options 1.a. and 1b. are exclusive and one of these two sub-options should be retained. Options 1.b scores better than option 1.a and will have a more deterrent effect on MTIC fraud. All other sub-options address different shortcomings that have been identified and all these options can be combined with each other;

·all sub-options would be cost-effective (efficient). None of them would entail additional costs for businesses and only two would require specific IT developments from tax administrations to create new access to already existing information not available to tax authorities. These developments seem to be proportionate to the benefits that could be expected from this information sharing;

·two options would have significant positive impacts on the level of VAT fraud. Option 1.b., (implementation of TNA with mandatory provision of data) and 5.b. (sharing of CP42/CP63 information between tax and customs administration) score higher than any other options. They would best meet the Member States' needs;

·positive effects of three other options could also be ascertained, with lower benefits than option 1.b and 5.b. though. These options are option 1.a. (TNA without mandatory provision of data), option 4.a. (access to Eucaris information), and option 5.a. (access to VAT registration numbers by customs authorities);

·in the table comparing the options, some of them score low. This is the case for all options 2 and 3. At the same time, these options would have positive impacts on VAT fraud. They would not deteriorate the current situation and would not entail significant costs for their implementation. They would bring new administrative cooperation instruments to Member States and would contribute to develop new working methods or new relationships between authorities involved in fighting VAT fraud. Therefore, increasing positive benefits from these measures are expected in the long term.

7.3. Subsidiarity of the different options

The management, collection and control of VAT are first and foremost a national competence of the Member States. However VAT fraud is often linked to cross-border transactions within the single market or involves traders established in other Member States than the one where the tax is due. Moreover VAT fraud has a negative impact on the functioning of the single market and causes serious losses to Member States’ revenues and consequently to the EU budget.

Therefore pursuant to Article 113 of the TFEU 96 the EU has implemented cooperation tools organising in particular an exchange of information between tax administrations and supporting common audit activities and the Eurofisc network.

Improving the efficiency of these tools in particular strengthening Eurofisc and establishing new ways of cooperation and new exchanges of tax information between tax authorities and with other law enforcement bodies would offer value over and above what could be achieved at Member State level.

7.4. Proportionality of the different options

All options described in this impact assessment are considered to be consistent with the principle of proportionality as while having positive effects on the VAT fraud level, they would entail no additional costs for business and for administrations except for options 4.a. and 5.b. where IT developments would be necessary but the associated costs would remain limited.

As with the subsidiarity test, it is not possible for Member States to address in an efficient and effective manner the problems and drivers identified in this impact assessment without amending Regulation (EU) 904/2010.

7.5. Impact on SMEs

All options presented in this impact assessment have no impacts on businesses in general and SMEs in particular since they would not trigger any additional administrative burdens and compliance costs. As detailed above under Chapter 6, eliminating fraudulent businesses from the market will level out the playing field for the legitimate businesses, improve their competitiveness and promote their growth. This will be the case for all enterprises, including SMEs.

8. How would actual impacts be monitored and evaluated?

8.1. Indicators for monitoring and evaluation

The table below gives an overview of the objectives, the indicators to measure whether they will be achieved, the tool for measuring them and the operational objectives.

8.2 Monitoring structures

Measuring the effectiveness, efficiency relevance, coherence and EU added-value of these new instruments will trigger the provision of new statistics and evaluation indicators. The precise list of these statistics will be adopted in accordance with Article 49 of Regulation (EU) 904/2010 with the assistance of the Standing Committee on Administrative Cooperation.

9. ANNEXES

9.1. Annex 1: Procedural information concerning the process to prepare the impact assessment report and the related initiative

9.2. Annex 2: Consultations synopsis report

9.3. Annex 3: Evaluation of Regulation (EU) 904/2010

9.4. Annex 4: Consultation of Members States

9.5. Annex 5: Consultation of stakeholders other than Member States

9.6. Annex 6: Who is affected by the initiative and how?

9.7. Annex 7: Methodology

9.8. Annex 8: Statistics and additional information about the functioning of administrative cooperation in the field of VAT

9.9. Annex 9: Transitional and definitive VAT system – Origin-based and destination-based VAT system

9.10. Annex 10: Technical fiche on concrete examples of VAT fraud

9.11. Annex 11: Findings and recommendations of the 2015 European Court of Auditors report: tackling VAT fraud: more action needed 97

9.1. Annex 1: Procedural information concerning the process to prepare the impact assessment report and the related initiative

9.1.1. Agenda Planning and Work Programme References

The proposal to amend Regulation (EU) 904/2010 is linked to the VAT Action Plan 98 .

TAXUD is the lead DG for this Initiative. The Agenda Planning reference is PLAN/2017/596. The inception impact assessment was published on 28 February 2017 99 .

9.1.2. Inter-Service Steering group (ISSG)

The first meeting of the Inter-Service Steering Group has taken place on 25 January 2017 where the draft questionnaire for the Member States and open-public consultations were discussed. The following directorates and services were present: SG, SJ, BUDG, and OLAF. The feedback received from these directorates and services has been taken into account in the final public consultation document.

The second meeting of the Inter-Service Steering Group has taken place on 7 April 2017 where the first impact assessment draft was discussed. The following directorates and services were present: SG, SJ, BUDG, CNECT, and OLAF. The feedback received from these directorates and services has been taken into account in the report.

The third meeting of the Inter-Service Steering Group will take place on 30 June 2017. The following directorates and services were present: SG, SJ, BUDG, CNECT, and OLAF.The feedback received from directorates and services that were present or had commented before the meeting has been taken into account in the report.

9.1.3 Consultation of the Regulatory Scrutiny Board (RSB)

The impact assessment Report was submitted to the Regulatory Scrutiny Board on 25 July 2017. On 15 September 2017, the Regulatory Scrutiny Board gave a positive opinion with a recommendation to further improve the report with respect to the following key aspects:

(1) The context could better explain the interaction with other recent and announced VAT initiatives, in particular the definitive VAT system;

(2) The report needs clearer motivations for the design of the policy options, including a description of stakeholders' support;

(3) The uncertainties and limits regarding available data and robustness of the evidence base are not sufficiently apparent. It is not clear why the report is not able to select a set of preferred options.

The report was adjusted to take into account the RSB's recommendations.

9.1.4. Anti-tax fraud strategy (ATFS) meeting

An ATFS meeting involving all Member States took place in the Commission premise on 10 July 2017. The outcome of the open-public and Member States consultations were set out and discussed. All options contained in this impact assessment were presented to Member States on that occasion. Delegates were invited to comment and provide their inputs. These views were taken into account to the largest extent possible.

9.2. Annex 2: Consultations synopsis report

 Synopsis report

of the consultations conducted to evaluate the functioning of and assess possible improvements to Regulation (EU) 904/2010 of 7 October 2010 on administrative cooperation and combating fraud in the field of Value Added Tax

1. Background to the initiative

The purpose of the evaluation of the use of Regulation (EU) 904/2010 on administrative cooperation and combating fraud in the field of Value Added Tax was:

·to assess whether and to what extent Regulation (EU) 904/2010 contributed to a closer cooperation between Member States;

·to assess to what extent Regulation (EU) 904/2010 has facilitated cooperation between Member States by making it smoother and faster and less burdensome;

·to assess to what extent Regulation (EU) 904/2010 has contributed to preventing budget losses stemming from tax avoidance and evasion;

·to assess whether and to what extent the provisions of Regulation (EU) 904/2010 continue to correspond to the needs of the Member States; and

·to assess to what extent the provisions of Regulation (EU) 904/2010 are in line with other policies and priorities of the EU and whether Member States have achieved similar results without acting at EU level.

2. Summary of the consultation activities carried out

a.On-going discussions with Member States at EU-level

There is a variety of fora organised at EU level where administrative cooperation and tax fraud matters are discussed amongst Member States and with the Commission. The following can inter alia be mentioned:

- the Standing Committee on Administrative Cooperation as provided for by Article 58 of Regulation (EU) 904/2010 and its Expert Group;

- the Anti-Tax Fraud Strategy group;

- the Eurofisc Group and its targeted working fields;

- project groups and seminars organised under the Fiscalis 2020 programme.

All these fora constitute platforms where issues in relation to the implementation of Regulation (EU) 904/2010 or the need to strengthen it are addressed on an on-going basis. Moreover, solutions envisaged to tackle them can be discussed at an early stage to further determine whether a need or an appetite to develop new instruments emerge. All these fora have, in particular, been used over the course of this policy cycle to collect inputs from the Member States when designing the targeted questionnaire and the different options on the table to improve the administrative cooperation framework in place.

b.Targeted consultation

A targeted consultation of the Member States on the functioning of the administrative cooperation and fight against fraud in the field of VAT was carried out. This consultation which was done via a questionnaire sent out to the Member States' tax authorities responsible for administrative cooperation and fight against fraud in the field of VAT ran from 7 March until 14 April 2017. On 30 June 2017, date on which the Evaluation Report was finalised, all Member States but one had replied to this consultation 100 .

This consultation sought to answer the following questions:

·to what extent current arrangements for administrative cooperation correspond to the needs of Member States; and

·what can be done in addition in that field in order to improve Member States ability to collect VAT and fight fraud.

For this purpose the Commission looked at the:

·effectiveness of the current arrangements for administrative cooperation to assess to what extent the objectives of the intervention have been achieved;

·efficiency of the current arrangements for administrative cooperation to assess to what extent the costs borne are proportionate to the benefits;

·relevance of the current arrangements for administrative cooperation to assess to what extent the objectives of the intervention are still corresponding to the needs of the Member States;

·coherence of the current arrangements for administrative cooperation to assess how its various internal components operate together to achieve the objectives of the intervention; and

·the EU added value to assess to what extent having common rules and tools at EU level makes the difference compared to what Member States could achieve at a national level.

To assess possible changes to the current arrangements for administrative cooperation the Commission sought to gather Member States' feedback in particular on:

·the possibility to introduce additional instruments and enhance existing ones for cross-border administrative cooperation;

·the possibility to expand data sources available to tax administrations;

·possible ways to ensure that law enforcement authorities have access to relevant VAT data in order for them to perform their tasks and in particular carry out criminal investigations when necessary.

The questionnaire contained statement questions allowing respondents to determine the extent to which they agreed with the statements. Most questions were followed by open text boxes where they could further develop their opinion.

27 Member States replied to this questionnaire. A summary report of their replies has been made available to the Member States. Germany did not manage to reply at the time when the Evaluation report had to be finalised, that is on 30 June 2017 101 . Hence, Germany's answers to the questionnaire are not taken into account in this Report.

Member States were invited to upload any document that could help the Commission to evaluate Regulation (EU) 904/2010 and the current arrangements for administrative cooperation. Three Member States (UK, IT, LV) provided such documents.

The overview of Member States’ answers to the questionnaire has been presented to them during an Anti-Tax Fraud Strategy meeting which took place on 10 July 2017.

Results of the the targeted consultation of Member States' tax authorities

The current arrangements for administrative cooperation and fight against fraud provided for in Regulation (EU) 904/2010 meet the Member States' needs either to a very high or to a high extent. Most Member States either strongly agree or agree that the new provisions introduced into the framework of administrative cooperation have improved their ability to monitor cross-border transactions and to collect VAT. It has also contributed to improve the quality, reliability and timeliness of information exchange as well as to increased legal certainty for traders.

Most Member States are of the opinion that the costs for participating in administrative cooperation are proportionate to the benefits achieved although opinions vary across the range of tools offered under the current legal framework. However, in practice, the vast majority of Member States have not been able to provide data precisely detailing the costs incurred by their participation in administrative cooperation and the balance with the benefits achieved. It is therefore not possible to provide any relevant quantitative results in relation to this matter. Even more, from the rather limited and not comparable information received, it is not possible to make comparisons and to extrapolate a robust quantification.

The automated access to information (VIES) is highly valued by the Member States. Ability to check information over VIES is considered to be very relevant and effective to control intra-EU transactions. This instrument also offers the best costs/benefits ratio. The extended rights to consult VIES for Eurofisc officials are also highly valued by Member States. However many of them consider that restricting these rights only to Eurofisc officials along with additional requirements is an ineffective and out of date arrangement that could hinder an efficient fight against fraud.

Eurofisc is considered by the majority of Member States as an effective mechanism to fight VAT fraud. However not all Eurofisc Working Fields are equally effective. While Working Field 1 and 4 rank high, further actions are needed in relation to Working Field 2 and 3 to improve targeting of signals. Issues with data quality, timeliness of feedback and targeting of early warning signals remain. Most Member States have high expectation towards TNA and are looking forward to having this tool at their disposal.

Collaboration with law enforcement authorities remains limited with only few Member States exchanging VAT relevant data. Very few Member States would like to see granting law enforcement agencies access to such information.

Cooperation between customs and tax administrations is of growing importance for Member States. Some of them mentioned that VAT fraud often involves transactions across the external border of the EU making information exchange with customs authorities crucial. Most Member States seem to exchange VAT relevant information with customs administrations and would like to have access to data on importations using Customs Procedure 42/63.

Most Member States would also like to have an automated access to cars registration information to fight VAT fraud.

The Member States rather agree than disagree that joint audits could potentially be a useful tool in auditing companies involved in intra-EU trade. However a number of issues were raised in relation to the rights of the auditors participating in such audits and the extent to which the decision of the audit would be binding.

In sum, the results of this consultation show the need, for the Member States, to improve the current instruments in place to fight VAT fraud. In particular a need to better use, share and analyse already existing information was expressed (be it already available to tax administrations or not). All this requires legislative measures that can be best introduced by amending Regulation (EU) 904/2010.

At the same time, issues in relation to the implementation of the Regulation current provisions have also been reported. This is the case with respect to the timeliness of answers, accuracy of VIES data, or the content of e-forms used to exchange information. As mentioned above, there are several fora at EU level where these questions of implementation are discussed. Many of them are already being addressed, for instance by introducing an e-forms central application from 2018.

As these issues are matters of implementation that are best examined and resolved with activities involving all Member States or by allocating more resources at national level rather than by amending the Regulation itself. This is why, over the course of the current process, it has not been considered that all these issues would deserve amendments to Regulation (EU) 904/2010 since they rather refer to implementation of the current arrangements.

c. Open-public consultation

An open-public consultation running from 2 March 2017 until 7 June 2017 and available in all EU languages was also carried out.

The purpose of this consultation was:

·to gather views from stakeholders other than tax administrations about their experience of the current rules governing administrative cooperation and fight against cross-border fraud in the field of VAT;

·to bring new insights for the on-going evaluation of Regulation (EU) 904/2010;

·to provide information about possible improvements including the on-line service for checking VAT numbers for intra-EU transactions : VIES on-the-web; and

·to collect data on possible reduction or increase of regulatory costs/benefits (administrative burden and/or compliance costs) for businesses (in particular SME’s) although in practice, such information was actually not provided in the answers received.

To this end, the Commission collated respondents’ views on the following topics:

·general principles governing administrative cooperation and fight against fraud in the field of VAT;

·control of VAT compliance in cross-border transactions – joint audits;

·fight against VAT fraud – role of Eurofisc; and

·functioning of VIES on-the-web.

All stakeholders – citizens, companies, organisations, institutions, public authorities, academic researchers – were invited to provide their views on this matter.

Though the public consultation was announced in several fora, publicly announced on Commission websites including ‘VIES on the web’ page and being made available in all EU official languages bar the Gaelic, in total, 58 individual replies from 12 Member States were received through the on-line survey tool. Respondents were mainly professionals responding on behalf of their organisation (74% - 47 replies).

These professionals generally work for a private enterprise (75%). The others respondents were citizens, replying in their personal capacity (11 replies).

It is important to note that since only 58 replies from 12 Member States were received over the course of the open-public consultation, these responses are not statistically representative of the target population. Answer ratios must therefore be interpreted with care without any possibility to draw any general conclusions from these replies.

Results of the open-public consultation

A large majority of respondents (around 50) support the role of the EU to assist and to ensure administrative cooperation amongst Member States. They widely approve the idea that VAT fraud is an international issue that must be dealt with jointly by all competent authorities at national and EU level. In their opinion, administrative cooperation at EU level has an added-value. In that perspective, 23 respondents considered that the costs for cooperation can be justified by additional VAT revenues against 9 that disagreed.

Additionally, respondents frequently emphasised that the exchange of information should be highly-secured, framed and used only for the purpose of fighting VAT fraud. Apart from this objective, the exchange of information would be irrelevant and unwanted. On possible improvements and extensions of the current system, respondents insisted that the EU should concentrate on making existing systems efficient and performant before thinking of creating new entities. They also frequently noted that all improvements should be designed with great carefulness taking into consideration their impact on businesses.

Respondents widely agree that VAT fraud is an international issue that must be dealt jointly at EU level. To this end, they see administrative cooperation as valuable. Nevertheless, 27 of them considered that the current instruments provided by Regulation (EU) 904/2010 are not very effective to fight cross-border VAT fraud and not very suitable for new business models such as e-commerce or collaborative economy. 29 considered that the current instruments are not effective to fight serious VAT fraud organized by criminal organizations. New automated exchanges of information as regards cars registration and exempted importations should be envisaged.

Respondents mainly support the role of Eurofisc to coordinate and carry out joint risk analysis (66%) and would allow it to coordinate cross-border administrative enquiries (78%). Regarding the implementation of an automated joint risk analysis, 47 respondents considered that it would be very useful to fight VAT fraud in an efficient manner. However, few respondents believed that making the current system more efficient and transparent would allow better control and enforcement possibilities without creating another body – which would generate unnecessary burdens for legitimate businesses.

Respondents also insisted on the need to develop contacts and cooperation between Eurofisc, national tax agencies and European administrative entities such as OLAF and Europol. Cooperation between Eurofisc and national financial administrations would also improve legal certainty of intra-EU transactions. It was also suggested by some respondents to develop contact points between tax authorities and businesses, as they usually possess very useful information and have a common interest to collaborate in order to reduce fraud risks.

From the 58 replies received, 28 respondents strongly agree or agree that more intense exchange of data is justified by the objectives pursued, while only 14 disagree or strongly disagree. 45 respondents consider that these exchanges should be strictly proportionate to the specific objectives of fighting VAT fraud.

Respondents expressed an overall positive opinion on joint audit. They widely agree that joint audits would constitute a step forward in administrative cooperation regarding fight against VAT fraud and would prevent duplication of tax audit which is burdensome for both tax administration and businesses. In specific, 40 of them consider that having a single report at the end of the audit would provide taxpayers with more legal certainty.

However, some of them observed that joint audits could be useful if a common unique procedure is clearly defined and if the rely on a legal basis. Joint audits outcomes should be directly shared with other Member States as well. It was emphasized that as the expertise of tax officials matters more than their citizenship, joint audits could be valuable if they are led by highly-qualified auditors knowing all Member States’ tax regulations. Respondents also observed that language barriers are an issue to be considered.Two respondents insisted that joint audits could be very useful in the field of transfer pricing, by preventing double taxation and legal disputes.

Although checking information by way of VIES on the Web is very useful as it helps reducing delays in processing VAT information, a majority of respondents considered that improvements should be made in this area. A large proportion of them asks for more harmonization of information collected and provided by Member States. They suggested that at least the name and address of the legal entity concerned should be added. The period of validity of the VAT number could also be provided in the system. Regarding the services proposed by VIES on the web, respondents asked for a batch processing for VAT checks with download possibility and for a report functionality in case of invalid VAT check, in order to report potential fraudsters.

Improvements in relation to the accuracy of data shown by the system could also be made. These concerns are serious since they can impede the correct functioning of active businesses involved in intra-EU trade. However, they do not require any amendments to Regulation (EU) 904/2010 since it is a matter of implementation of the current rules by the Member States. The Commission works on an on-going basis with them to help improve the situation.

d. Other information collected

Other sources of information were also used in the course of Regulation (EU) 904/2010 evaluation:

·annual statistics provided by Member States in accordance with Article 49 of Regulation (EU) 904/2010;

·discussions held within the Standing Committee on Administrative Cooperation and its Expert Group meetings. 102 In particular, the evaluation report uses the Working document SCAC-EG 47 – statistic for 2016 (committee 2016);

·comments expressed by Member States' delegates in several Fiscalis Project groups and workshops;

·statistics of the use of Regulation (EC) 1798/2003;

·previous reports :

oReport from the Commission to the Council and the European Parliament. Seventh report under Article 12 of Regulation (EEC, Euratom) n° 1553/89 on VAT collection and control procedures; 103

oReport from the Commission to the Council and the European Parliament on the application of Council Regulation (EU) no 904/2010 concerning administrative cooperation and combating fraud in the field of value added tax; 104

oEuropean Court of Auditors special report no 24/2015: Tackling intra-EU VAT fraud: More action needed 105 ;

4. Conclusion

Member States’ opinion is overall positive on the legal and practical framework implemented with Regulation (EU) 904/2010. Most of them consider that it has contributed to improve the administrative cooperation between Member States. Amongst the instrument provided by the Regulation, the exchanges of information on request, the automated exchanges (VIES), Eurofisc and the multilateral controls are viewed as the most effective, efficient and relevant instruments.

However, developing new instruments or new ways of cooperating seems to be needed. In particular, Eurofisc seems to be at a turning point as Member Sates particularly support the possibility to implement a joint processing of VAT data. There is also room to further develop automated exchange of information or access to new sets of data. The exchange of information with other EU law enforcement authorities remains however a sensitive area.

The other stakeholders support the role of the EU to assist and to ensure administrative cooperation amongst Member States. However, some of them consider that the current instruments are not adapted to new business models and to the fight against cross-border or organized crime fraud. New automated exchanges of information and a greater collaboration between the competent authorities should be envisaged. They support a greater role for Eurofisc. VIES on the Web is considered as very useful by most stakeholders even if some improvements could also be made. They expressed an overall positive opinion on joint audit.

9.3. Annex 3: Evaluation of Regulation (EU) 904/2010

TABLE OF CONTENT

List of Figures    

1. EXECUTIVE SUMMARY    

2. INTRODUCTION    

3. BACKGROUND TO THE INITIATIVE    

3.1. Situation prior to Regulation (EU) 904/2010    

3.2. Regulation (EU) 904/2010 and its objectives    

4. EVALUATION QUESTIONS AND METHOD    

4.1. Evaluation questions    

4.2. Evaluation materials    

4.2.1. Questionnaire to the tax authorities    

4.2.2. Open-public consultation    

4.2.3. Yearly statistics    

4.2.4. Other sources    

4.3. Evaluation process and matrix    

5. IMPLEMENTATION OF THE REGULATION    

6. ANALYSIS OF THE ANSWERS TO THE EVALUATION QUESTIONS    

6.1. To what extent has Regulation (EU) 904/2010 contributed to a closer cooperation between Member States? (effectiveness)    

6.1.1. Development and stakeholder assessment across the various forms of exchange of information between Member States    

6.1.1.1. General views of Member States on changes introduced to Regulation (EU) 904/2010    

6.1.1.2. Evolution of the total number of exchanges of information on request and qualitative assessment of the perceived effectiveness of the tools for exchange of information    

6.1.1.3. Speediness of answers to request for information and late answers    

6.1.1.4. Evolution of the total number of spontaneous exchange of information and qualitative assessment of the perceived effectiveness of this tool    

6.1.1.5. Evolution of automatic exchange of information and qualitative assessment of the perceived effectiveness of this tool    

6.1.1.6. Evolution of the total number of feedback requested and provided and qualitative assessment of the perceived effectiveness of this tool    

6.1.1.7. Overall assessment of the effectiveness of the different exchange of information programmes    

6.1.2. Developments and stakeholder assessment of other forms of administrative cooperation    

6.1.2.1. Evolution of the total number of requests for administrative notification and qualitative assessment of the perceived effectiveness of this tool    

6.1.2.2. Evolution of the total number of administrative enquiries and presence of officials in the administrative premises of other administrations and qualitative assessment of the perceived effectiveness of this tool    

6.1.2.3. Evolution of the total number of the MLCs and effectiveness of this instrument    

6.1.2.4. Overall assessment of the effectiveness of administrative cooperation programmes other than exchange of information    

6.1.3. Development and stakeholder assessment of VIES    

6.1.3.1. Evolution of the number of queries made on VIES by Member States    

6.1.3.2. Qualitative assessment of the use, usefulness and quality (e.g. up-to-datedness, accuracy, reliability, etc.) of data in VIES of the quality of data in VIES    

6.1.3.3. Overall assessment of the effectiveness of VIES    

Summary of findings: to what extent has Regulation (EU) contributed to a closer cooperation between Member States? (effectiveness)    

6.2. To what extent has Regulation (EU) 904/2010 facilitated cooperation between Member States by making it smoother, faster and less burdensome? (efficiency)    

6.2.1. Positive development and stakeholder assessment on the use of e-forms    

6.2.1.1. Evolution and qualitative assessment of the use of e-forms for EOI on request    

6.2.2. Costs and benefits of administrative cooperation between Member States    

6.2.2.1. Costs and benefits associated with the implementation of Regulation (EU) 904/2010    

6.2.2.2. Perceived administrative burden associated with participation in administrative cooperation viewed by Member States    

6.2.2.3. Stakeholder other the Member States' assessment of the costs/benefits for Member States    

Summary of findings: to what extent has Regulation (EU) 904/2010 facilitated cooperation between Member States by making it smoother and faster and less burdensome? (efficiency)    

6.3. To what extent has Regulation (EU) 904/2010 contributed to preventing budget losses stemming from tax avoidance and evasion and from errors in the application of VAT rules? (effectiveness)    

6.3.1. Extent to which Regulation (EU) 904/2010 has helped Member States to assess and recover more VAT    

6.3.1.1. Amounts assessed and recovered on the basis of Regulation (EU) 904/2010    

6.3.1.2. Member States assessment on the extent to which delays and irregularities incurred in the process of information exchange are detriment to the correct assessment, control and recovery of VAT    

6.3.1.3. Overall assessment of the extent to which Regulation (EU) 904/2010 has helped Member States to assess and recover more VAT    

6.3.2. Extent to which Member States are better equipped to fight VAT fraud thanks to Regulation (EU) 2010/94    

6.3.2.1. Assessment of the use, utility and future of Eurofisc    

6.3.2.2. Views of stakeholders other than Member States on the effect of administrative cooperation on VAT fraud    

6.3.2.3. Overall assessment of the extent to which Member States are better equipped to fight VAT fraud thanks to Regulation (EU) 904/2010    

6.3.3. Extent to which Regulation (EU) 904/2010 helped stakeholders other than the Member States to correctly apply VAT rules    

6.3.3.1. Evolution of the use of VIES on the web by stakeholders other than the Member States    

6.3.3.2. Assessment of stakeholders other than the Member States with regard to their experience with VIES on the web (VoW)    

6.3.3.3. Overall assessment of the extent to which Regulation (EU) 904/2010 helped stakeholders other than the Member States to correctly apply VAT rules    

Summary of findings: to what extent has Regulation (EU) 904/2010 contributed to preventing budget losses stemming from tax avoidance and evasion and from errors in the application of VAT rules? (effectiveness)    

6.4. To what extent the provisions of Regulation (EU) 904/2010 continue to meet the needs of the Member States? (relevance)    

6.4.1. The administrative cooperation instruments meet the needs of Member States    

6.4.1.2. Assessment of the use and problems encountered by Member States in using Regulation (EU) 904/2010 for their own needs    

6.4.1.3. Qualitative assessment of the problems encountered by tax administrations when exchanging information with other law enforcement authorities    

6.4.1.4. Qualitative assessment of the relevance of VIES for Member States needs and problems encountered when using VIES    

6.4.1.5. Qualitative assessment of the relevance of automatic and spontaneous exchange of information in confront with other mechanism    

6.4.1.6. Qualitative assessment of the extent to which Eurofisc continues to meet the needs of Member States    

6.4.1.7. Overall assessment as to whether administrative cooperation instruments still fits the needs of Member States    

6.4.2. Views of stakeholders other than the Member States on VIES on the web    

6.4.2.1. Qualitative assessment of the extent to which VIES on the web (VoW) continues to fit the needs of stakeholders other than the Member States    

6.4.3. Identified areas in need for further changes to Regulation 904/2010    

6.4.3.1. Qualitative assessment of the need for change in the area of Eurofisc & Risk analysis    

6.4.3.2. Qualitative assessment of the need for change in the area of multilateral activities (MLCs, joint audits, administrative enquiries)    

6.4.3.3. Qualitative assessment of the need for change in e-forms    

6.4.3.4. Qualitative assessment of the need for change in the area of automatic exchange of information, including car registration data    

6.4.3.5. Qualitative assessment of the need for change in the area of customs data    

6.4.3.6. Qualitative assessment of the need for change in the area of VIES    

6.4.3.7. Qualitative assessment of the need for change in the area of strengthening cooperation with other law enforcement bodies    

6.4.3.8. Overall assessment of identified areas in need for further changes to Regulation (EU) 904/2010    

Summary of findings: to what extent the provisions of Regulation (EU) 904/2010 continue to correspond to the needs of the Member States? (relevance)    

6.5. To what extent are the provisions of Regulation (EU) 904/2010 in line with other policies and priorities of the EU? Could Member States have achieved similar results without acting at EU level? (coherence and EU added value)    

6.5.1. Administrative cooperation in VAT is a cornerstone for the proper functioning of the single market and the fight against VAT fraud    

6.5.1.1. Qualitative assessment of the level of convergence of the objectives of Regulation (EU) 904/2010 with other Commission's strategies    

6.5.1.2. Qualitative assessment of the inter-dependence of provisions governing administrative cooperation in the VAT with other Commission's strategic initiatives and programmes (e.g. the developments in the VAT area towards the definitive VAT system, Fiscalis activities)    

6.5.2. Administrative cooperation in VAT contributes to the creation of a single EU VAT area    

6.5.2.1. Qualitative assessment of the Regulation (EU) 904/2010's contribution to the creation of a common system for cooperation between the Member States    

6.5.3. Joint EU approach in administrative cooperation in VAT has advantages over other forms of national and international forms of tax cooperation    

6.5.3.1. Cases of cross-border fraud (e.g. MTIC), MLC indicator, VAT gap)    

6.5.3.2. Exchange of best practices    

Summary of findings: To what extent are the provisions of Regulation (EU) 904/2010 in line with other policies and priorities of the EU? Could Member States have achieved similar results without acting at EU level? (coherence and EU added-value)    

7. CONCLUSIONS    

List of Figures

Figure 1: Administrative cooperation instruments provided for by Regulation (EU) 904/2010

Figure 2: Impact of changes introduced with Regulation (EU) 904/2010

Figure 3: Number of requests for information sent out by Member States

Figure 4: Effectiveness of the tools for administrative cooperation

Figure 5: Exchange of information on request. Number of requests sent out by Member States Evolution between 2011 and 2016

Figure 6: Percentage of late replies

Figure 7: Number of spontaneous exchanges of information

Figure 8: Non-established taxable persons - Volume of data exchanged per year

Figure 9: New means of transport - Volume of data exchanged per year

Figure 10: Impact of Member States abstaining from exchanging information without prior request

Figure 11: Number of feedbacks requested per year

Figure 12: Number of feedbacks provided per year    

Figure 13: Requests for administrative notification - Number of requests received per year

Figure 14: Requests for notification - Number of requests sent out per year

Figure 15: Administrative enquiries/Presence of officials - Number of requests made per year

Figure 16: Simultaneous controls - Number of MLCs initiated per year

Figure 17: Simultaneous controls - Total number of participating Member States

Figure 18: VIES: total number of requests made by Member States per year

Figure 19: Usefulness of data exchange over VIES and impact of inaccurate data in VIES

Figure 20: Effectiveness of automated access to VIES

Figure 21: Costs in relation to benefits of participating in administrative cooperation

Figure 22: Proportionality of costs and benefits in relation to participation in administrative cooperation

Figure 23: Burden in relation to participating in administrative cooperation

Figure 24: Administrative costs borne by Member States to cooperate at EU level are justified by additional VAT revenues

Figure 25: Impact of late replies to request for administrative cooperation

Figure 26: Effectiveness of Eurofisc in relation to collection of VAT on intra-EU transactions

Figure 27: Eurofisc liaison officials' rights to consult VIES

Figure 28: The current instruments are effective to prevent cross-border VAT fraud

Figure 29: The current instruments are effective to fight serious VAT fraud organised by criminal organisations.

Figure 30: The current instruments are sufficient to fight VAT fraud occurring in new business models such as in the collaborative economy or ecommerce.

Figure 31: VIES on the Web - Number of requests made per year

Figure 32: VIES on-the-web is a useful tool for businesses carrying out intra-EU transactions

Figure 33: How do you rate your general experience with the consultation of the VAT identification number of your customers through VIES on-the-web?

Figure 34: Your VAT identification data was correctly reflected in the VIES on-the-web system?

Figure 35: The VAT identification data of your customers were correctly reflected in the VIES on-the-web system?

Figure 36: Complaints received in relation to VoW data

Figure 37: To what extent are the tools for administrative cooperation provided by Regulation (EU) 904/2010 relevant for the control of intra-EU transactions and the collection of VAT? Please provide your opinion on each tool?

Figure 38: Problems in information exchange with law enforcement authorities

Figure 39: Data available through VIES and Member States' needs

Figure 40: Do you encounter problems with getting automated access to the information exchanged through VIES as provided for in Article 21?

Figure 41: Replacing automated and spontaneous exchange of information with other mechanisms

Figure 42: The availability of the ‘VIES on-the-web’ server is good

Figure 43: Do you consider that a system of e-mail notification to inform users about technical issues with VIES on-the-web would be useful?

Figure 44: Do you consider that the current way of checking VAT numbers should be replaced with a system that inform you about the changes to your clients VAT data such as the validity of their VAT number?

Figure 45: Member States' opinions on TNA

Figure 46: The fight against cross-border VAT fraud would benefit from an EU wide automated joint risk analysis system.

Figure 47: Member States experience with joint audits

Figure 48: Member States' opinion on joint audits

Figure 49: It would be more efficient to also authorise an active participation of tax officials in audits carried out by other Member States.

Figure 50: Being audited by a single team of auditors from different Member States in which you carry out economic transactions and controlling your operations taxable therein would be less burdensome.

Figure 51: Having a single report at the end of the audit signed by all the tax administrations involved would give taxpayers more legal certainty

Figure 52: Need to update standard forms

Figure 53: Modifying categories for automatic exchange of information

Figure 54: Tax administrations should have automated access to information related to cars registration from other Member States

Figure 55: Exchange of information in relation to CP42

Figure 56: Tax administrations should have automated access to information on exempt importations from customs authorities

Figure 57: In your opinion should the rules for automated access to VIES data (Art 21(2)) be changed?

Figure 58: Adding categories of information to be exchanged over VIES

Figure 59: The scope of information directly accessible in VIES should be extended when relevant.

Figure 60: Allowing law enforcement authorities access to data exchanged over VIES

Figure 61: Possible collaboration between Europol, OLAF and Eurofisc

Figure 62: The EU is the best placed to provide Member States with common rules to allow them to work more closely together in order to monitor intra-EU trade for VAT purpose and ensure the correct application of VAT.

1. EXECUTIVE SUMMARY

This evaluation of the use of the EU framework for administrative cooperation and combating fraud in the field of VAT accompanies the impact assessment carried out in conjunction with a legislative proposal to amend Council Regulation (EU) No 904/2010 of 7 October 2010 on administrative cooperation and combating fraud in the field of value-added tax (hereinafter ‘Regulation (EU) 904/2010’). It examines to which extent the overall objective of this Regulation to contribute to a closer cooperation between Member States, to avoid budget losses, to fight VAT fraud and to preserve principles of fair taxation have been met.

The evaluation questions relate to the effectiveness, the efficiency, the relevance, the coherence and the EU added value of this EU framework (see section 4.1.).

The evaluation is mainly based on the comments and responses provided by the Member States' tax authorities, the responses provided by stakeholders other than Member States in response to an open public consultation and the yearly statistics on the use of this EU framework. Other sources of information such as issues addressed in the Standing Committee on Administrative Cooperation and its Expert Group meetings, in Fiscalis 2020 activities and in other external reports (see section 4.2.) have been used when relevant.

Member States’ opinion is overall positive on the legal and practical framework implemented with Regulation (EU) 904/2010. The vast majority of them consider that it has contributed to improve administrative cooperation between Member States. Amongst the instruments provided by the Regulation, Member States reported that exchanges of information on request, automated exchanges (VIES 106 ), Eurofisc 107 and multilateral controls are the most effective instruments.

Drawbacks in the manner in which administrative cooperation takes place are however reported. Late replies to requests for information is the most significant. Although in decrease, the number of instances where it happens remains significant - 33% - and Member States should work to remedy it. The accuracy of data exchange through VIES is a domain where concerns are expressed by both the Member States and businesses. Businesses need accurate and up-to-date information to properly apply the VAT rules to their intra-EU transactions and Member States should improve this situation.

Regulation (EU) 904/2010 appears to be an efficient instrument for organising administration cooperation. Most Member States consider that costs incurred by participating in administrative cooperation in the field of VAT are to a high extent proportionate to the benefits achieved. Automated access to information through VIES, Eurofisc, simultaneous controls, requests for information and administrative enquiries score particularly high. These instruments are considered as the most efficient by the Member States but at the same time rather burdensome. This is certainly the result of the large number of instances where they are used, requiring administrative resources to handle them.

Member States consider that costs incurred by administrative cooperation are proportionate to the benefits although, due to a lack of comparison elements, it has not been possible to precisely assess the extent to which Regulation (EU) 904/2010 has contributed to prevent budget losses. In the field of fight against VAT fraud, Eurofisc appears to be the most powerful instrument. A vast majority of Member States reports a high level of satisfaction and agrees that it has had a positive impact on the collection of VAT on intra-EU transactions. However, there are still shortcomings regarding its functioning leading to the conclusion that Eurofisc has certainly not reached its potential yet.

Member States consider that all administrative cooperation instruments in the field of VAT are relevant. Some instruments score very high such as VIES and requests for information. Developing new instruments or new ways of cooperating seems to be needed. In particular, Eurofisc is at a turning point as Member Sates particularly support the possibility to implement a joint processing of VAT data. There is also room to further develop automated exchange of information or access to new sets of data. In this context, Member States are particularly interested in an access to customs data or car registration information. The best way to access it still needs to be discussed, although Member States are more in favour of automated than automatic exchanges. On the opposite, Member States do not demonstrate a strong appetite to develop exchange of information with other EU law enforcement authorities such as Europol or OLAF.

Improving administrative cooperation in the field of VAT is fully coherent with other EU policies currently under development. This is particularly the case with regard to mutual assistance in the field of direct taxation where a number of improvements took place over the recent years. It is also coherent with other EU initiatives underway such as the draft Directive to protect the EU financial interests or the European Public Prosecutor's office. All this demonstrates that there are several common initiatives taken at EU level with similar objectives: improving cooperation between law enforcement authorities and finding new way of fighting the most severe threats to tax revenues.

Administrative cooperation is essential for the proper functioning of the EU VAT area which itself is a cornerstone of the single market. Common VAT rules have been accepted by the Member States because they were supported by reciprocal, binding and evolving administrative cooperation rules. As such, administrative cooperation is an essential component of the free circulation of goods and services which is one of the four freedoms enshrined in the EU. Other forms of cooperation between Member States also demonstrate the need to use the EU level as the platform of reference to discuss and develop administrative cooperation in the field of VAT.

In conclusion, it results from this evaluation that Regulation (EU) 904/2010 constitutes the cornerstone of administrative cooperation and fighting fraud in the field of VAT. This instrument, coherent with other already or to be implemented EU policies, suits the needs of the Member States. The high level of cooperation developed between them demonstrates a good knowledge and appropriation of this Regulation and further-on its effectiveness and efficiency. Drawbacks in its implementation still exist and Member States should work at national and EU levels to improve its functioning. At the same time, this evaluation shows that new administrative cooperation instruments are needed by the Member States. Reinforcing Eurofisc or providing new tools to fight the most severe fraud threats appear to be amongst their highest expectations for the near future.

2. INTRODUCTION

On 7 April 2016, the Commission adopted an Action Plan on VAT 108 (hereinafter ‘Action Plan’) setting out ways to modernise the VAT system so as to make it simpler, more fraud-proof and business-friendly. In this context, the Commission announced its intention to adopt in 2017 four VAT-related proposals:

·a definitive VAT system for intra-EU cross-border trade based on the principle of taxation in the Member State of destination in order to create a robust single European VAT area (first step);

·a modernised VAT rates policy so as to allow Member States greater autonomy on setting the VAT rates;

·a comprehensive simplification VAT package for SMEs;

·a proposal to enhance VAT administrative cooperation and Eurofisc 109 .  

This current framework for administrative cooperation and fight against VAT fraud is laid down in Regulation (EU) 904/2010.

The purpose of this evaluation is to assess whether and to what extent the implementation of Regulation (EU) 904/2010 by the Member States have made administrative cooperation and the combat against VAT fraud more efficient and effective.

It will also assess whether this framework continues to be relevant and coherent for Member States' needs and what is the added value of organising this cooperation at EU level.

Finally, it will assess to what extent this administrative cooperation needs to/can be improved, and make recommendations to take into account the current needs of the Member States and the functioning of the single market in a quickly changing economic and political environment.

3. BACKGROUND TO THE INITIATIVE

3.1. Situation prior to Regulation (EU) 904/2010

Along with the abolition of the physical borders within the EU and the implementation of the ‘transitional arrangements’ for intra-EU trade 110 , a framework allowing Member States to cooperate to better manage and control VAT was put in place by the end of 1992 111 . With the introduction of Regulation (EU) 904/2010, the first objective was to gather, in a single piece of legislation, all provisions in relation to administrative cooperation in the field of VAT. On top of that, and in response to the most severe and always sophisticated VAT frauds, improving this administrative cooperation framework overall was also viewed as a necessity.

3.2. Regulation (EU) 904/2010 and its objectives

The first purpose of Regulation (EU) 904/2010 currently in force remains the exchange of information between Member States tax authorities with three different and complementary types of programmes: on request, spontaneous and automatic. To get information on businesses involved in intra-EU transactions, tax authorities can also rely on VIES, an IT tool developed as of 1992 112 . Certain information has also been made available to private stakeholders through ‘VIES on the web’ 113 where intra-EU VAT numbers delivered by Member States' tax administrations can be checked on-line.

Regulation (EU) 904/2010 also provides for other cooperation tools i.e. simultaneous controls and presence in administrative offices and during administrative enquiries. Regulation (EU) 904/2010 has also introduced a new means for an enhanced cooperation, namely Eurofisc.

The functioning of administrative cooperation in the field of VAT, as provided for by Regulation (EU) 904/2010 can be summarised as showed below in figure 1:

Figure 1: Administrative cooperation instruments provided for by Regulation (EU) 904/2010

Different objectives were pursued with the introduction of Regulation (EU) 904/2010 of which two can be viewed as its main objectives:

·to contribute to a closer cooperation between Member States; and

·to avoid budget losses, to fight VAT fraud and to preserve the principle of fair taxation.

4. EVALUATION QUESTIONS AND METHOD

4.1. Evaluation questions

The evaluation is based on the following questions:

·To what extent has Regulation (EU) 904/2010 contributed to a closer cooperation between Member States? (effectiveness)

·To what extent has Regulation (EU) 904/2010 facilitated cooperation between Member States by making it smoother faster and less burdensome? (efficiency)

·To what extent has Regulation (EU) 904/2010 contributed to preventing budget losses stemming from tax avoidance and evasion? (effectiveness)

·To what extent the provisions of Regulation (EU) 904/2010 continue to correspond to the needs of the Member States and other stakeholders? (relevance)

·To what extent are the provisions of Regulation (EU) 904/2010 in line with other policies and priorities of the EU? Could Member States have achieved similar results without acting at EU level? (coherence, EU added value)

4.2. Evaluation materials

4.2.1. Questionnaire to the tax authorities

a.    In order to gather the information necessary for the preparation of this report, the Commission has asked for the opinion of the 28 Member States' tax authorities dealing with administrative cooperation in the field of VAT. At the time this evaluation report was finalised (30 June 2017), all Member States but one had replied to the targeted evaluation questionnaire, which was sent to them on 7 March 2017 with expected replies by 14 April 2017 at the latest. This questionnaire was prepared by the Commission and discussed with the inter-service steering group (ISSG) in a meeting that took place on 25 January 2017. One set of replies was received per responding Member States. Overall quality and completeness of replies were satisfactory.

b.    The overview of Member States' answers to the questionnaire has been presented to them during an ‘Anti-Tax Fraud Strategy’ meeting taking place on 10 July 2017 114 .

4.2.2. Open-public consultation

a.    An open public consultation took place from 2 March to 31 May 2017. It was prepared by the Commission and discussed with the ISSG in a meeting that took place on 25 January 2017.

Though the public consultation was announced in several fora, publicly announced on Commission websites including ‘VIES on the web’ page and being made available in all EU official languages bar the Gaelic, there was only a limited number of responses to this public consultation (58).

However, as Regulation (EU) 904/2010 deals with cooperation between tax authorities, the stakeholders other than Member States might have thought that there was only a limited interest in replying to the open-public consultation, although some parts of it directly related to the business (in particular the use of 'VIES on the web' and joint audits).

It is important to note that since only 58 replies from 12 Member States were received over the course of the open-public consultation, these responses are not statistically representative of the target population. Answer ratios must therefore be interpreted with care without any possibility to draw any general conclusions from these replies.

b.    A summary report of the responses to this public consultation has been presented to Member States during an Anti-Tax Fraud Strategy meeting taking place on 10 July 2017 and published on the Commission's website. 115

4.2.3. Yearly statistics

a.    The replies to the above consultations completed the information already available to the Commission from the yearly statistics that Member States have to provide to the Commission in accordance with Article 49 of Regulation (EU) 904/2010. The statistics cover all types of administrative cooperation instruments provided for by the Regulation:

·exchange of information on request;

·spontaneous exchange of information;

·automatic exchange of information;

·simultaneous controls;

·presence during administrative enquiries;

·presence in administrative offices; and

·use of VIES.

The Standing Committee on Administrative Cooperation Committee has adopted detailed guidelines with regard to the calculation of the statistical data. Nevertheless, the statistical data provided by the Member States do not always match. Mismatches can still be noted in the numbers of the requests sent and received. Such differences in the reported figures can be explained in certain Member States by a poor communication between some tax administration' departments and the competent authority which provides he Commission with the information. This is typically the case when liaison offices have been implemented without suitable tracking instruments.

Under these circumstances, the accuracy of these statistics cannot be fully guaranteed although they have been checked by the Member States. Nevertheless, they provide an estimate of the level of administrative cooperation which has taken place between Member States over the past 5 years 116 .

b.    The usefulness of the statistics could be improved if more information was collected and if their reliability could be checked. However, the statistical information that Member States have to provide annually cannot be decided without their agreement 117 and they are reluctant to provide more statistics, as this could entail additional administrative burden (e.g. the information on additional VAT assessed and collected by Member States with the help of administrative cooperation instruments). In addition, the annual reports which are prepared by the Eurofisc Working field coordinators as provided for in Article 37 of Regulation (EU) 904/2010 and compiled in the annual report prepared by the Eurofisc chair are not made public. Therefore, they cannot be used despite the fact that they relate to a significant activity provided for by Regulation (EU) 904/2010.

4.2.4. Other sources

a.    Discussions on certain issues which were held within the Standing Committee on Administrative Cooperation and its Expert Group meetings. 118  

b.    The evaluation also relied on comments expressed in several Fiscalis project groups and workshops.

c.    The evaluation took account of previous reports, including the following:

·Report from the Commission to the Council and the European Parliament. Seventh report under Article 12 of Regulation (EEC, Euratom) n° 1553/89 on VAT collection and control procedures; 119  

·Report from the Commission to the Council and the European Parliament on the application of Council Regulation (EU) No 904/2010 concerning administrative cooperation and combating fraud in the field of value added tax; 120

·The European Court of Auditors 2015 special report no 24/2015: Tackling intra-EU VAT fraud: More action needed. 121

4.3. Evaluation process and matrix

a.    The evaluation process was monitored by a Commission inter-service Steering group supporting the evaluation and revision of Regulation (EU) 904/2010. 122  

b.    The evaluation covered all Member States for the entire period from the date from which the provisions of the Regulation had to be applied in the Member States (1 January 2012). 123  

c.    The following evaluation matrix has been used:

5. IMPLEMENTATION OF THE REGULATION

Council Regulations do not have to be transposed by Member States and are directly applicable. Hence, Regulation (EU) 904/2010 applies from the date laid down in Article 62 that is 1 January 2012 except for Croatia that joined the European Union on 1 July 2013.

6. ANALYSIS OF THE ANSWERS TO THE EVALUATION QUESTIONS

6.1. To what extent has Regulation (EU) 904/2010 contributed to a closer cooperation between Member States? (effectiveness)

6.1.1. Development and stakeholder assessment across the various forms of exchange of information between Member States

6.1.1.1. General views of Member States on changes introduced to Regulation (EU) 904/2010

Member States’ opinion is overall positive regarding changes introduced in 2010 in Regulation (EU) 904/2010 ( Figure 3 2). A large majority of Member States consider that they contributed to improve the administrative cooperation framework at EU level, to increase legal certainty for the taxpayers, to improve the monitoring and collection of VAT for cross-border transactions and to improve the quality, reliability, timeliness and governance of the information exchanged.

One Member State (CZ) considers that the quality of information exchanged has improved mainly due to the introduction of Eurofisc. However Italy notes that the quality of information remains an issue though signals received led to numerous follow up actions. Three Member States (SE, FI and LU) insist on the growing need for the exchange of information to be faster.

Figure 2: Impact of changes introduced with Regulation (EU) 904/2010

6.1.1.2. Evolution of the total number of exchanges of information on request and qualitative assessment of the perceived effectiveness of the tools for exchange of information

As a whole (Figure 3), and following the entry into force of Regulation (EU) 904/2010, an increasing number of outgoing requests for information has been reported by the Member States with a total number of 46.000 requests in 2012 and 49.000 in 2013. Since then this number has been stable close to 49.000 requests per year with small fluctuations over the last years.

In sum, Regulation (EU) 904/2010 may have had a positive effect on the total number of outgoing requests for information at the beginning. The number reached a plateau after about two years.

No targets have been set as regards the number of requests for information sent annually. Such targets would be difficult to set as the intensity of cooperation is an outcome of many externalities such as the volume of cross-border economic activities, changing trends in fraud schemes, changes in the administrative capacity of the Member States, the implementation of domestic anti-fraud measures or of best practices identified as a result of cross-border cooperation or the maturity in the use of the available instruments. The volume of exchange is thus prone to change in an unpredictable manner at any point in time. It also depends on the trends by country (see below Figure 5) with some countries making a better use of this instrument while others use it less.

When it comes to the effectiveness of exchange of information on request provided by Regulation (EU) 904/2010 as it is actually perceived by the Member States, and indicated in Figure 4, it appears that all Member States either find this instrument effective or very effective to ensure the correct collection and application of VAT. The European Court of Auditors (hereinafter 'ECA') in its 2015 report 124 on tackling intra-EU VAT fraud came to the same conclusion by indicating that Member States consider information exchanges as the most effective administrative cooperation tool.

However Bulgaria, Latvia, Croatia, Lithuania, Portugal, and the United Kingdom stressed that all Member States should fully commit to participating in information exchanges and provide complete, accurate and properly targeted information – otherwise effective action to fight VAT fraud could not be carried out.

As regards the number of requests for information sent out by Member States between 2011 (last year where Regulation (EC) 1798/2003 125 was in force), and year 2016 (last year for which statistics for Regulation (EU) 904/2010 are available), as illustrated in Figure 5, it appears that the number of requests has increased for 21 Member States and decreased for 8 of them (Austria, Denmark, Greece, France, Ireland, Romania, Sweden). For some Member States, this increase is very significant with 12 Member States having an increase above 50%: Lithuania (+512%), Czech Republic (+ 295%), Latvia (+276%), Poland (+254%), Bulgaria (+105%), Estonia (+114%), Slovenia (+90%), Cyprus (+75%), Finland (+70%), , Portugal (+65%), Malta (+59%), and Slovak Republic (54%).

It is nevertheless difficult to draw a conclusion as to whether this trend is due to the entry into force of Regulation (EU) 904/2010. In particular, most of the Member States with the most significant increases belong to the group of most recent Member States while older Member States have a stable or an even decreasing number of such requests. One could argue that these increases are the result of a better appropriation of the administrative cooperation instruments by these Member States while the older ones have already approached a certain level of maturity in this respect. On the other hand, the fact that some businesses moved their activities to Central and Eastern Europe and the increase of intra-EU trade with these countries over the years following their accession could also explain the increase of their requests.

6.1.1.3. Speediness of answers to request for information and late answers

The number of late replies decreased between 2012 and 2016 from 19.500 requests not answered within the mandatory three months deadline in 2012 to 16.800 requests in 2016. At the same time, the total number of outgoing request increased. Overall, the proportion of requests not answered within three months, decreased from 43 % in 2012 to 33 % in 2016 (see figure 6).

Member States usually have a monitoring system in place to follow-up requests for information in order to meet the three months deadline. Despite this, the high level of late replies demonstrates that there are still shortcomings in the way some Member States deal with the incoming requests. This was already highlighted in the Commission Report to the Council and the European Parliament on the application of Regulation (EU) 904/2010 126 .

The high level of late replies was also stressed as unsatisfactory by the ECA in its 2015 report. It was also emphasised that six Member States replied late more than 50 % of the time, which seems to indicate that some Member States experience difficulties in effectively organising and managing the administrative cooperation within their tax administration. Moreover, an analysis of the number of requests received per Member State showed that delays were not always explained by the workload caused by the number of requests. With the exception of the United Kingdom, none of the Member States’ tax authorities audited by the ECA at that time had set operational targets for reducing the percentage of late replies. 

6.1.1.4. Evolution of the total number of spontaneous exchange of information and qualitative assessment of the perceived effectiveness of this tool

The spontaneous exchange of information 127 is an instrument that already existed under Regulation (EC) 1798/2003 and in which Member States had already a high level of experience. Therefore, the number of exchanges remained stable with the entry into force of Regulation (EU) 904/2010 (about 7.000 exchanges per year). The only exception was year 2013 where only 6.000 of such exchanges took place (Figure 7).

Regarding the effectiveness of this instrument, as showed in figure 4, 6 Member States find it very effective and 17 effective to monitor the correct application of VAT on cross-border transactions and ensure their ability to collect VAT. In its 2015 report, the ECA also indicated that Member States clearly found this tool useful and provided many examples (additional VAT assessments, information about missing traders, corrections in VIES) in which the exchange of information without prior request had been valuable to them.

6.1.1.5. Evolution of automatic exchange of information and qualitative assessment of the perceived effectiveness of this tool

The experience of Member States in automatic exchange of information is long-standing since such exchanges regarding non-established taxable (NETP) persons and new means of transport already took place under Regulation (EC) 1798/2003. The replies from the Member States to the questionnaire (see figures 4) demonstrate that they are less convinced that this instrument is useful for the correct application of VAT on cross-border transactions. Only one Member State sees it as a very effective instrument, 15 as effective and 11 as either somewhat effective or not very effective. Such perception is explained partially by the statistics (see below), which may indicate that the limited usefulness is due rather to the lack of use of the tool rather than loopholes in its design.

According to the two diagrams below (see Figures 8 and 9), the number of such exchanges remained low with on average 25.000 to 30.000 pieces of data exchanged per year on non-established taxable persons and 20.0000 on new means of transport. As regards the latter, half of the Member States do not exchange any data at all and the vast majority of information exchanged (on average half of the total) comes from a single country, Belgium.

The United Kingdom notes that non-participation to automatic exchange of vehicle information has a greater negative impact than non-participation to NETP exchange. France declares not being affected by other Member States not participating in automatic exchange of information.

It is surprising to notice in Figure 10 that a large number of Member States considers that the lack of participation of the others in information exchanges without prior request affects at least to some extent their ability to collect VAT while a few Member States seem to be active in this area.

6.1.1.6. Evolution of the total number of feedback requested and provided and qualitative assessment of the perceived effectiveness of this tool

As reported in Figure 4, 17 Member States consider the provision of feedbacks as either a very effective or effective instrument to monitor the correct application of VAT on cross-border transactions and for the control of intra-EU transactions. A feedback can demonstrate the usefulness or accuracy of the information provided. It is also a means for a requested Member State to determine whether its practices to meet the needs of its counterparts are appropriate and if not to improve them. Ultimately, it is also rewarding for officials engaged in administrative cooperation be it at central or field levels.

Feedbacks were introduced in the VAT administrative cooperation framework with Regulation (EU) 904/2010. On average, 2.500 feedbacks were requested and provided by Member States in 2016 (Figures 11 and 12). This is significant increase - +400 % - since only 500 of them were requested and provided in 2012 and still 1.800 in 2014 only. As this tool was new in 2012, it was used in a limited number of instances at the beginning. Since then, a better awareness of it explains this significant increase for the period under review.

The match between the number of feedbacks requested and provided as shown in Figures 11 and 12 demonstrates that when asked to do so, Member States do provide such feedbacks. At the same time, this also means that spontaneous feedbacks are provided in a limited number of instances, even where the requesting Member State is content with the information received as it has been useful and has provided the expected results.

6.1.1.7. Overall assessment of the effectiveness of the different exchange of information programmes

It results from the data presented above that when it comes to effectiveness of exchange of information whichever its form, Regulation (EU) 904/2010 meets its objectives. The stable number of request for information, spontaneous and automatic exchanges since 2013 shows a good appropriation by Member States and a maturity in the use of this instrument:

·with regard to exchange of information on request, Member States are content with the framework created by Regulation (EU) 904/2010 which is, to a large extent, the continuation of Regulation (EC) 1798/2003. This framework seems to be effective although late replies remain a concern for a large majority of them. The total number of late replies while in progress remains high. The three month deadline is still not met in 33% of instances while at the same time Member States recognise that this situation is not appropriate. This issue has always existed. Addressing it would require that Member States implement adequate procedures to collect information and evidences in due time and to properly monitor the deadlines. Such measures should be taken by the Member States in order to meet the requirements provided for by Regulation (EU) 904/2010;

·The stability of the number of spontaneous exchanges demonstrates a good appropriation of the instrument by Member States. However, although a vast majority of Member States considers this instrument to be effective, the total number of such spontaneous exchanges remains relatively low: there are seven times more exchanges on request than spontaneous ones. To be used to a larger extent, such tools require a change in the administrative practices in a spirit of solidarity within the EU: the tax authorities should try to protect the financial interest of the other Member States like they do for their own country.

·The automatic exchange of information was already provided for by Regulation (EC) 1798/2003. Despite this long-standing experience, it seems that Member States do not really invest in this type of instrument while, paradoxically, they consider it to be effective and report negative impacts due to a lack of commitment of the other Member States. Room for improvements exists in this area to implement efficient procedures to collect data for the different categories of automatic exchange of information provided for by Regulation (EU) 904/2010. Member States should actively cooperate in all types of administrative cooperation instrument provided by the Regulation;

·With regard to feedbacks, while the number of exchanges is increasing, Member States are of the opinion that this instrument should be more often and better used. This demonstrates that this is an effective tool. At the same time, if not specifically requested, these feedbacks are not provided. This requires raising awareness in the Member States about the effectiveness of this tool and the existing e-forms that has been drawn up to facilitate its use.

6.1.2. Developments and stakeholder assessment of other forms of administrative cooperation

6.1.2.1. Evolution of the total number of requests for administrative notification and qualitative assessment of the perceived effectiveness of this tool

With a little bit more than 300 requests sent out and received in 2016, the request for administrative notification remain an instrument that is not often used by Member States although the number of such requests made per year doubled between 2012 and 2016 (Figures 13 and 14).

It also results from Member States' answers to the targeted questionnaire that the requests for administrative notification is one of the relatively least appreciated instrument with only 16 Member States viewing it at very effective or effective to monitor the correct application of VAT on cross-border transactions (see Figures 4). One Member States (SI) notes that it is not used much and has a limited impact on VAT assessment due to differences in procedures within the EU. The only problem mentioned in relation to use of this tool is linguistic as requests must often be translated.

6.1.2.2. Evolution of the total number of administrative enquiries and presence of officials in the administrative premises of other administrations and qualitative assessment of the perceived effectiveness of this tool

With regard to presence in administrative offices and presence during administrative enquiries (PAOE), since the entry into force of Regulation (EU) 904/2010, the number of such requests (Figure 15) has increased steadily from 100 for 2012 to 200 for in 2016 with a downward fluctuation in 2015. However, the number of instances where officials participate in foreign administrative enquiries or are present in administrative offices abroad remain low when compared to the 50.000 requests for information sent each year.

When it comes to its effectiveness, as reported in Figure 4, Member States note that PAOE are rarely used due to language barriers, lack of human resources, legislation differences between Member States and lack of awareness as regards its usefulness. The absence of common rules also hampers its use. As a consequence, each Member State approaches the use of these tools differently which creates confusion for the auditors. To deal with these issues, one Member State suggests creating a separate CCN mailbox for PAOEs. A number of Member States are of the opinion that these instruments should be promoted more at national and EU levels for instance by way of workshops, similarly to those organised for simultaneous controls.

One Member State (UK) notes that usually presence in administrative offices is used by Member States to examine records relating to one of their businesses that are kept in another Member State. This approach seems to be beneficial as it speeds up the audit and therefore reduces the administrative burden. However in such cases, the requesting Member State is not conducting the audit itself and must rely on an official from the requested Member State to perform the audit and provide him with the information. It suggests that this practice should be harmonised to allow a Member State to conduct and audit in another Member State (with the consent of the business and the other Member State administration) if the records of the first Member State business are held in the second Member State.

6.1.2.3. Evolution of the total number of the MLCs and effectiveness of this instrument

As for PAOE, the number of simultaneous/multilateral controls ('MLCs') initiated per year remains limited with less than 50 controls in 2016 and 40 controls on average for the period 2012-2016 (Figure 16). After a certain stability with 120 different countries participating in these MLCs during the period 2012-2015 (Figure 17), this number has increased to reach more than 160 countries in 2016. However, this last figure is the direct result of a higher number of MLCs launched in 2016 only; on average three different countries are involved in each MLC, a figure that has been stable since the entry into force of Regulation (EU) 904/2010.

Member States recognize the added value of MLCs as it is considered as effective to collect VAT. However two Member States (LV and NL) note that the underlying procedure is ineffective and needs to be provided with more resources. Mismatches in national legislations create further obstacles for effective application of this tool. It was also noted that such controls often take too long which decreases their effectiveness further. One Member State (LV) mentioned that the option of combining simultaneous controls and PAOE should be envisaged.

The ECA also came to the conclusion in its 2015 report that 27 Member States consider MLCs as a useful tool for combating VAT fraud. The ECA also noted that the instrument was not fully exploited. The same conclusion can also be drawn from the statistics above.

6.1.2.4. Overall assessment of the effectiveness of administrative cooperation programmes other than exchange of information

As regards request for administrative notification, given the low up-take of this instrument, it is difficult to assess the level of its effectiveness.

On PAOE end MLCs, these tools are considered to be effective or very effective by Member States. The MLC instrument is the second best scored by Member States regarding its effectiveness after requests for information. However, no indicators have been put in place by Member States to precisely measure the effect of this tool. It also seems that promotion and better communication are still needed to give this instrument its full potential. This could be the case if all Member States allocated to it more resources and ensured that they either launch or actively participate in them. The use of the PAOE tool could still be improved as it is considered as an appropriate means for tax administrations and taxpayers to save time as common issues can be solved jointly and in a quicker way.

6.1.3. Development and stakeholder assessment of VIES

6.1.3.1. Evolution of the number of queries made on VIES by Member States

The total number of queries made by Member States in VIES is very high (Figure 18): more than 300 million of queries are made on an annual basis. In 2016, this number exceeded 400 million. From the statistics, it appears that this increase is mainly the results of a growing number of requests concerning registration information (Registry Messages) while the number of requests concerning turnover date (TOD message) remains between 50 and 100 million for 2013 to 2016 after reaching a peak at almost 250 million queries in 2012. It has not been possible to provide an explanation to this decrease.

Figure 18: VIES: total number of requests made by Member States per year

6.1.3.2. Qualitative assessment of the use, usefulness and quality (e.g. up-to-datedness, accuracy, reliability, etc.) of data in VIES of the quality of data in VIES

Member States generally find data shared through VIES system rather useful ( Figure 8 19 below). They report that such data is necessary to guarantee the correct application of VAT exemption of intra-EU supplies, to identify potential VAT risks, to minimize the number of information requests and to make these requests more targeted. Some Member States mention that the usefulness of VIES data is undermined by retroactive changes introduced by tax administrations and complain about not being able to view the history of those changes. All Member States either strongly agree or agree that inaccurate VIES data has a negative impact on their ability to collect VAT. They finally mention that inaccurate data create administrative burdens as tax administrations have to conduct additional checks and send requests for information that could have otherwise been avoided.

All Member States find automated access to VIES to be effective ( Figure 10 20) and allow them to control intra-EU transactions and collecting VAT. However some Member States mentioned that the current restrictions listed in Article 21(2)(e) of Regulation (EU) 904/2010 to send the so-called "third Member State requests" into VIES create additional complications (see Chapter 6.4.3.6 for more details).

The issue of the timely cancellation of VAT ID numbers was also stressed in the Commission Report to the Council and the European Parliament on the application of Regulation (EU) 904/2010 128 and again by the ECA 2015 report, in this last instance both from the Member States replies as well as the audits carried out by the ECA itself. It also results from these audits that the error messages received by the Member States are not followed-up in most instances. The ECA came to the conclusion that whilst VIES is a very useful tool for exchanging data on intra-EU supplies between Member States, there are weaknesses in its use which occasionally affect the reliability, accuracy, completeness and timeliness of VIES data and therefore its overall effectiveness in tackling fraud.

6.1.3.3. Overall assessment of the effectiveness of VIES

The growing number of queries made by Member States to the system shows its usefulness for the tax authorities. The same level of appreciation is also reported by the Member States in their replies to the questionnaire. There is a broad consensus amongst Member States to consider the instrument effective or very effective.

However the Member States complain that the accuracy of VIES data is not always ensured. This hampers their ability to control and collect VAT. Member States are not in all instances in a position to maintain their databases up-to-date and some of them update them retroactively (e.g. when cancelling a VAT identification number). These practices jeopardise the legal certainty of the Member States and the businesses (see below). They should be avoided to the best extent possible to maintain the credibility of VIES which it based on its reliability.

Summary of findings: to what extent has Regulation (EU) contributed to a closer cooperation between Member States? (effectiveness)

Many administrative cooperation instruments provided for by Regulation (EU) 904/2010 were already in place under Regulation (EC) 1798/2003. This is the case for exchange of information on requests, spontaneous, automatic, and automated exchange of information, administrative notifications or simultaneous controls.

Member States’ opinion is overall positive as concerns the legal and practical framework implemented with Regulation (EU) 904/2010. The vast majority of them considers that it has contributed to improve the administrative cooperation framework at EU level. From statistics annually provided by Member States, it can be noted that between 2012 and 2016, the number of instances where administrative cooperation took place is more or less stable for the most significant and powerful-scored instruments: exchanges of information, whichever the type of exchanges, and simultaneous controls. This demonstrates a good appropriation of administrative cooperation instruments in the field of VAT by Member States, a maturity in their use and ultimately their effectiveness.

As regards exchanges of information on request, as constantly reported by Member States in surveys or to the Commission, the level of late replies remain significant although their proportion decreased from 43 % in 2012 to 33 % in 2016. This issue questions Member States' capacities to put in place adequate measures to collect information and evidences in due time. Further work in this area at Member States level is still needed.

Whilst a majority of Member States considers the exchange of information without prior request as an effective instrument to combat fraud in the field of VAT, the use of spontaneous and automatic exchanges remains below expectations. At the same time, the Member States also highlight the fact that the lack of commitment of the others Member States to engage in such programmes hampers to a certain extent their capacities to properly fight VAT fraud.

The automated exchange of data by way of VIES is viewed as a very effective tool. At the same time, the Member States report that the way some of them manage their VIES data often negatively affects the reliability, accuracy, completeness and timeliness of VIES data and therefore its effectiveness in tackling fraud.

As regards multilateral controls and the presence in foreign administrative offices and enquiries, these tools are considered to be very effective or effective by Member States. However, the number of instances where they are used remain low, around respectively 50 and 200 instances per year. It is clear that the potential of these tools is not fully exploited in particular when it is viewed together with the total number of exchanges of information on requests made per year – around 49.000.

6.2. To what extent has Regulation (EU) 904/2010 facilitated cooperation between Member States by making it smoother, faster and less burdensome? (efficiency)

6.2.1. Positive development and stakeholder assessment on the use of e-forms

6.2.1.1. Evolution and qualitative assessment of the use of e-forms for EOI on request

The last version of electronic forms (e-forms) for the exchange of information has been used as of 2013. In 2018 a new major evolution will be put in place with the future e-forms Central Application (eFCA) to replace the electronic transmission of XML files 129 . When they were first implemented, e-forms brought a lot of advantages. A single form, used by all Member States, including predefined questions translated into all official languages and exchanged electronically and immediately, was a means to deeply improve the efficiency of the exchanges of information between Member States.

The ECA in its 2015 report mentioned that according to Member States, exchanges of information made through e-forms are the most powerful tool to fight VAT fraud, since replies can be used as evidence before court. E-forms are functioning in a satisfactory manner, leading to speedier processing of requests.

No specific issues in relation to the use of e-forms have been reported by Member States in their replies to the targeted consultation and overall e-forms implemented at EU level meet their needs and is considered as an efficient way to exchange information.

6.2.2. Costs and benefits of administrative cooperation between Member States

6.2.2.1. Costs and benefits associated with the implementation of Regulation (EU) 904/2010

25 out of 27 Member States strongly agree or agree that costs associated with participating in administrative cooperation are proportionate to the benefits ( Figure 14 21). Unfortunately very few Member States provided monetary estimations of the costs and benefits in relation to their participation in administrative cooperation. Member States note that costs and benefits resulting from the use of each specific tool can hardly be estimated as many of tax auditors and officials deal with both European and national enquiries. Only Estonia managed to provide a break-down of (salaries) costs by a specific tool offered by Regulation (EU) 904/2010. Based on this, it can be concluded that the biggest costs are associated with requests for information and administrative enquiries, multilateral controls and Eurofisc operations.

As reported above, the vast majority of Member States have not been able to provide quantitative data precisely detailing the costs incurred by their participation in administrative cooperation and the balance with the benefits achieved. Only few of them have managed to provide some quantification of these benefits. Nevertheless, most estimates cover multiple years and reporting periods vary across Member States. Substantial differences in amounts also hint that Member States seem to be using different methodologies in estimating those benefits. Differences in Member States sizes also make data submitted incomparable. It is therefore not possible to provide any significant results in relation to this matter. From the rather limited information received, it is even more not possible to make comparisons and to extrapolate a robust quantification.

When broken down by specific instrument for administrative cooperation (see Figure 22), automated access to information through VIES scores particularly high with 18 Member States noting that cost-benefit ratio is proportionate to a very high extent. Eurofisc, simultaneous controls and request for information and administrative enquiries also score high. Other tools do not rank as high with feedback being the least proportionate tool when looking at the cost-benefit ratio. Indeed the feedback is useful to the Member States which has provided the information so it is not directly used to assess and recover additional VAT revenues but to improve the administrative practices in dealing with the administrative cooperation instruments in the long run.

6.2.2.2. Perceived administrative burden associated with participation in administrative cooperation viewed by Member States

When it comes to assessing burden in relation to particular tools offered by Regulation (EU) 904/2010 it seems that the largest burdens are associated with tools that require significant human resources such as dealing with requests for administrative enquiries, replying to requests for feedback and participating in multilateral controls.

Member States mentioned additional issues that increase the burden of participating in administrative cooperation:

·language barrier, the need to translate enquiries and often the poor quality of those translations;

·MLCs as well as presence in office and during enquiries can be burdensome as they require more organisation and operational work than using information on request;

·although a large part of Member States does not consider that replying to feedbacks is excessively burdensome, some of them note that feedbacks are deprived of any additional value compared to resources it requires;

·difference in national tax procedure within the EU is an issue that should be addressed to facilitate administrative cooperation;

·insufficient capacity at domestic level to collect data for automatic exchange of information; and

·multiple requests concerning the same taxpayer (e.g. request in relation to different taxable periods) create additional complications.

6.2.2.3. Stakeholder other the Member States' assessment of the costs/benefits for Member States

Stakeholders other than Member States' tax authorities have also been asked as to whether administrative costs borne by Member States to cooperate at EU level are justified by additional VAT revenues (Figure 24). 23 respondents out of 58 considered that the costs are justified by additional VAT revenues against 9 that disagreed. At the same time 15 respondents did not have an opinion.

Summary of findings: to what extent has Regulation (EU) 904/2010 facilitated cooperation between Member States by making it smoother and faster and less burdensome? (efficiency)

Overall the Member States expressed a high level of satisfaction with regard to the efficiency of the administrative cooperation instruments.

Some of the exchanges of information between Member States rely on the use of e-forms and no specific issues in relation to them have been reported by Member States in their replies. E-forms appear to be already an efficient way to exchange information. The implementation of the e-form central application (eFCA) planned for 2018 is expected to improve the situation further.

With respect to ratio cost/benefit of the administrative cooperation, more than half of the Member States consider that costs are to a very high or a high extent proportionate to the benefits achieved. Automated access to information through VIES, Eurofisc, simultaneous controls, request for information and administrative enquiries score particularly high. These instruments are considered as the most efficient instruments by Member States.

However while these instruments are the most cost-efficient, they are also scored as the most burdensome, probably because, apart from MLCs, these are the most used tools and require most involvement from tax authorities in particular in terms of human and IT resources. A number of issues have been reported by Member States but most of them, like languages, legislation issues or multiple requests for a same taxpayer, are well documented and already discussed at EU level.

6.3. To what extent has Regulation (EU) 904/2010 contributed to preventing budget losses stemming from tax avoidance and evasion and from errors in the application of VAT rules? (effectiveness)

6.3.1. Extent to which Regulation (EU) 904/2010 has helped Member States to assess and recover more VAT

6.3.1.1. Amounts assessed and recovered on the basis of Regulation (EU) 904/2010

Very few Member States provided monetary estimations of the costs and benefits in relation to their participation in administrative cooperation. Member States note that costs and benefits resulting from the use of each specific tool can hardly be estimated. For Member States that have been in a position to provide figures, the corresponding additional amount of tax assessed strongly varies among Member States, from EUR 11 million in Slovenia to EUR 400 million in Hungary.

Bulgaria declares that administrative cooperation was used in 207 of its audits in 2016 and contributed to the establishment of about EUR 55millions of additional liabilities. Replies to outgoing requests and incoming spontaneous information have resulted in EUR 21,23 million of additional established liabilities. Amount of additional VAT due to exchanges of information in Slovak Republic is about EUR 150 million. Some Member States note that the requested estimates could be available in the future.

With regard to MLCs, it appears that each single audit led to an average result of EUR 18,5 million additional liabilities (VAT and direct taxes). It should be noted at the same time that MLCs, and other cooperation tools, help to identify and assess the additional amounts of unpaid VAT but their recovery is a matter of further national administrative and legal proceedings. It means that the effective amounts returned to the national budgets will be likely lower than the nominal amounts identified through exchange of information and other tools.

Despite the fact that the Commission has been constantly asking the Member States to provide estimates of the amounts of additional VAT revenues assessed and recovered due to the use of administrative cooperation instruments, these figures are often not available or are not provided. This is mostly the result of a lack of monitoring tools available to Member States to measure in any way the extent to which administrative cooperation has contributed to identifying additional liabilities.

In many instances administrative cooperation is regarded as an additional instrument to national tools that Member States use in the fight against VAT avoidance and evasion. The use of several tools can contribute to the same assessment to an extent which is difficult to evaluate. Therefore most of the Member States do not estimate the specific impact of each instrument used. And any such estimates should always be treated with caution while bearing in mind that Member States measure these differently and depart from different footing.

The ECA itself, in its 2015 report, noted that with the exception of the United Kingdom no monitoring or indicators on intra-EU VAT fraud had been put in place in Member States and consequently that no further impact indicators to monitor progress or results in fight VAT fraud were set up. Furthermore, the same report indicated that, with the exception of the United Kingdom, none of the audited Member States’ tax authorities had set operational targets for collecting additional revenue arising from VAT information exchanges or for the number of assessments carried out or fraud cases identified due to administrative cooperation.

6.3.1.2. Member States assessment on the extent to which delays and irregularities incurred in the process of information exchange are detriment to the correct assessment, control and recovery of VAT

25 Member States note that late replies to request for administrative cooperation have, at least to some extent, negative impacts on their ability to collect VAT ( Figure 6 25). A number of Member States mentioned that due to domestic timelines for conducting audits, replies often arrive when such a timeline is over making the information useless. However it is also noted that quick but of poor quality replies are useless as well. Member States also mentioned that the need to send reminders when the reply is late creates additional burden for liaison officials.

The same happens in relation to automated exchanges made through VIES where irregularities and delays in updating data has a detrimental effect on their ability to assess, control and collect VAT.

All these shortcomings were already stressed by the ECA in its 2015 report.

6.3.1.3. Overall assessment of the extent to which Regulation (EU) 904/2010 has helped Member States to assess and recover more VAT

Few Member States are in a position to provide figures about the VAT assessed and recovered due to the use of administrative cooperation instruments. However, it can reasonably be concluded that administrative cooperation has a positive effect on Member States' capacity to assess and recover VAT.

6.3.2. Extent to which Member States are better equipped to fight VAT fraud thanks to Regulation (EU) 2010/94

6.3.2.1. Assessment of the use, utility and future of Eurofisc

The objective of Regulation (EC) 1798/2003 was to put in place an administrative cooperation framework between Member States in the field of VAT. With the entry into force of Regulation (EU) 904/2010 fighting VAT fraud became another and complementary objective of mutual assistance between Member States. As a result, Eurofisc, the first administrative cooperation instrument dedicated to the fight against the most severe VAT threats was implemented. None of the other instruments provided for by Regulation (EU) 904/2010 has the same unique purpose. Therefore the assessment of Eurofisc is central when assessing the effectiveness of Regulation (EU) 904/2010 in fighting VAT fraud.

A majority of Member States agrees that Eurofisc has a positive impact on the collection of VAT on intra-EU transactions ( Figure 12 26) although not all Member States participate in all working fields. It appears that the two working fields where all Member State participate (Working Field 1 and 4 130 ) are considered to be the most effective by Member States.

However, 0some Member States mentioned that the quality, swiftness and targeting of the information exchanged should be further improved, in particular in relation to Working Fields 2 and 3. The number of information exchanged is viewed as far less useful than the quality and targeting of signals.

Member States also note that the current medium for information exchange – online portal CircaBC 131 – is not user friendly.

Eurofisc liaison officials have extended rights to consult VIES. Contrary to other officials who can only consult intra-EU supplies involving their own country, they can access information on all transactions. These extended rights were introduced by Regulation (EU) 904/2010 to accompany the implementation of Eurofisc. Member States mainly agree that this extended right to consult VIES for Eurofisc officials is useful and contribute to make Eurofisc an effective early warning tool which positively impact collection of VAT on intra-EU transactions as it enhances fast and accurate detection of fraudulent cross-border transactions chains ( Figure 13 27).

This assessment confirms the one made by the ECA in its 2015 report where it was mentioned that 27 Member States consider Eurofisc to be an efficient early warning system for fraud prevention. However the ECA pointed out the following weaknesses, which were also confirmed by the audit tests in Member States: (i) feedback was not frequent enough; (ii) data exchanged was not always well targeted; (iii) not all Member States participated in all Eurofisc working fields; (iv) exchanges of information were not user friendly; and (v) data exchanges were too slow.

Some Member States made suggestions on how Eurofisc can be improved. They are further detailed in Chapter 6.0.

6.3.2.2. Views of stakeholders other than Member States on the effect of administrative cooperation on VAT fraud

A number of questions in relation to the effect of administrative cooperation on VAT fraud have been asked to stakeholders other than Member States through the open public consultation. 27 respondents out of 58 considered that the current instruments against VAT fraud are not effective (Figure 28). 29 of them considered that the current instruments are not effective in the fight against fraud organised by criminal organisations (Figure 29).

28 respondents considered that the current instruments are not adapted to the new business models such as e-commerce or collaborative economy (Figure 30). At the same time 14 did not have an opinion.

These comments came from stakeholders who are not directly involved in the fight against VAT fraud but could be potentially indirectly impacted by VAT fraud, be it at professional or personal level as taxpayers. The relevance and accuracy of this perception according to which the current instruments available to Member States to act in this area are not sufficient, not powerful enough or not used to their full potential is difficult to assess.

6.3.2.3. Overall assessment of the extent to which Member States are better equipped to fight VAT fraud thanks to Regulation (EU) 904/2010

When it comes to Eurofisc which was introduced by Regulation (EU) 904/2010, the level of satisfaction of Member States scores very high: all Member States strongly agree or agree that Eurofisc is an effective tool that improves the control of intra-EU transactions. As for any instrument, in particular when it is relatively new, improvements in relation to the functioning of Eurofisc are still needed. The Member States have for instance identified some shortcomings in the manner in which information is exchanged or signals targeted. The Chair and Coordinators of Eurofisc and Eurofisc working field coordinators are aware of these shortcomings and have been working with Eurofisc liaison officials to improve the situation. As a result, the number of signals exchanged has been decreasing, demonstrating a better targeting. The quick and accurate provision of feedbacks is another area where room for improvement is possible and would help in making Eurofisc more effective.

All this leads to the conclusion that Eurofisc has certainly not reached its full potential yet.

6.3.3. Extent to which Regulation (EU) 904/2010 helped stakeholders other than the Member States to correctly apply VAT rules

6.3.3.1. Evolution of the use of VIES on the web by stakeholders other than the Member States

VIES on the Web (VoW) is the IT tool made available to businesses involved in intra-EU transactions to check the validity of VAT registration numbers attributed by Member States. This is an essential tool for the correct application of VAT in intra-EU transactions.

Between 2012 and 2016, the number of queries made on VoW tripled to reach 1,2 billion (Figure 31). This means that the instrument is needed by stakeholders other than Member States to identify the VAT treatment applicable to their intra-EU transactions.

This is emphasized by the replies to the open public consultation where 41 respondents agreed or strongly agreed that VoW is a useful tool for businesses carrying out intra-EU transactions (Figure 32). 43 consider that the tool is at least satisfactory for their needs (Figure 33)

6.3.3.2. Assessment of stakeholders other than the Member States with regard to their experience with VIES on the web (VoW)

47 respondents report that their registration information was correctly recorded in VoW although only 36 of them indicate that information on their customers was also accurate in VoW (Figures 34 and 35). This is an indication of the fact that VoW does not always provide the expected level of accuracy to these stakeholders. This has direct consequences on them since they are not able to assess with a sufficient level of confidence the VAT status of their business counterparts.

More than half of the Member States have received complaints from traders in relation to VoW data ( Figure 9 36) which are generally related to the accuracy, completeness, and availability of information provided by the system. Three Member States (SE, LT and CZ) mention that retroactive deregistration from VAT registers can cause problems for business too. This is the case when businesses check VAT registration numbers on VoW, these numbers are valid and deregistration takes place at a later time with a retroactive date. Few Member States report discrepancies between data reported by VoW and their national databases. One Member State (UK) stressed that a business not having informed the administration that it has engaged in intra-EU supplies does not appear in the system after registration.

6.3.3.3. Overall assessment of the extent to which Regulation (EU) 904/2010 helped stakeholders other than the Member States to correctly apply VAT rules

VoW is an essential tool for businesses involved in intra-EU transactions. Without this instrument, these businesses are not in a position to meet the requirements set by the Member States to deliver goods or services VAT free to clients located abroad. Although respondents to the open public consultation reported their satisfaction regarding the accuracy of their own data recorded on VoW, it seems that the overall satisfaction level of the instrument is below expectation. The lack of accuracy of the information made available by the Member States has direct impacts on the reliability of data stored in VoW since both systems are linked. There is still room for improvement with regard to the manner in which data recorded in VoW are kept up-to-date by Member States and necessary measures should be taken to make sure that VoW is a reliable instrument that meets the needs of businesses involved in intra-EU transactions.

Summary of findings: to what extent has Regulation (EU) 904/2010 contributed to preventing budget losses stemming from tax avoidance and evasion and from errors in the application of VAT rules? (effectiveness)

It is difficult to ascertain the extent to which Regulation (EU) 904/2010 has contributed to prevent budget losses since only a few number of Member States keep records showing the additional liabilities assessed and collected thanks to administrative cooperation. Nevertheless, as a large majority of Member States does consider that benefits derived from administrative cooperation are proportionate to the costs incurred, it can be considered that this impact is positive.

Many Member States have raised some administrative cooperation shortcomings which impact revenue collection such as late replies and irregularities in VIES data. These two aspects have been reported as having a detrimental effect on Member States' ability to assess, control and collect VAT.

When it comes to Eurofisc, the more prominent instrument to fight VAT fraud, a majority of Member States agree that this new instrument introduced with the entry into force of Regulation (EU) 904/2010 has had a positive impact on the collection of VAT on intra-EU transactions. A very high level of satisfaction is reported by Member States. However, there are still drawbacks regarding its functioning leading to the conclusion that Eurofisc has certainly not reached its potential yet.

Regarding the use of VoW, it results from the open public consultation that this instrument is essential to enabling the businesses to apply properly the VAT rules. Nevertheless, VoW does not always provide the level of reliability these stakeholders expect which demonstrates that there is still a room for improvement in the manner in which data recorded in VoW are kept up-to-date by the Member States.

6.4. To what extent the provisions of Regulation (EU) 904/2010 continue to meet the needs of the Member States? (relevance)

6.4.1. The administrative cooperation instruments meet the needs of Member States

Figure 37 below describes the Member States' general assessment of the relevance of each administrative cooperation instrument made available to them. Some instruments score very high: 22 Member States consider VIES as a relevant tool to a very high extent and 18 do the same for requests for information. If Eurofisc and simultaneous controls rank a little bit behind, overall, almost all Member States consider that these two instruments are relevant to a very high or to a high extent. Feedback and request for administrative notification are again the less best-rated instruments.

6.4.1.1. Appropriation of administrative cooperation instruments by Member States

As described under Chapter 6.1, from the statistics annually provided by Member States, it can be noted that between 2012 and 2016, the number of instances where administrative assistance was provided has been stable for the most significant and effective instruments: exchanges of information, whichever the type of exchanges, and simultaneous controls. This demonstrates a good appropriation of the legal framework provided for by Regulation (EU) 904/2010 by the Member States' tax administrations and a maturity in the use of its instruments. This can be explained as most of these tools were already available to them with Regulation (EC) 1798/2003.

Some administrative cooperation instruments are far less used than others like requests for administrative notification or provision of feedbacks. Nevertheless, this does not mean that they are irrelevant or should be discontinued. The limited use is rather due to various factors such as the particularities of the procedural law of the cooperating Member States as regards the MLCs or the presence of foreign officials. Other instruments could be considered as still gaining traction (e.g. the feedback). Further details on the relevance and use of the different instruments as well as the associated difficulties are described below.

6.4.1.2. Assessment of the use and problems encountered by Member States in using Regulation (EU) 904/2010 for their own needs

One Member State out of two encounters problems either at EU or national levels when using Regulation (EU) 904/2010. At EU level, they include problems related to the lack of commitment or participation of some Member States in respect of some of the instruments, late replies or transmission of incomplete and inaccurate information. Some issues regarding differences between Member States VAT reporting obligation or tax legislation are also raised. A few Member States also insist that only one working language should be used which illustrates some linguistic problems.

At national level, the following problems can be mentioned:

·lack of awareness amongst national auditors with regard to the different administrative cooperation instruments that exist (feedbacks, presence in offices and during enquiries, simultaneous control), their usefulness and their application;

·difficulties to train tax officials to the use of e-forms;

·very slow procedures to carry out request for information and administrative enquiries; and

·lack of human resources and language skills.

No specific serious issue in relation to the implementation of Regulation (EU) 904/2010 or any questioning of its relevance were raised by the Member States in the evaluation process. That implicitly means that the Regulation tools continue to be needed. It is worth mentioning that when a new issue is raised, Member States usually report it to the Commission to address it with the other Member States. Fiscalis activities are specifically designed to address some of the concerns raised, for instance training and information on administrative cooperation tools such as MLCs or the use of e-forms.

6.4.1.3. Qualitative assessment of the problems encountered by tax administrations when exchanging information with other law enforcement authorities

The responsibility to enforce VAT legislation at national level falls in all Member States with national tax authorities. Nevertheless, other authorities and bodies either at national or EU levels are also involved in the enforcement of these obligations or in the fight against VAT fraud. This is typically the case for customs or police authorities, be it at national or EU level.

Member States tax administrations report various experiences regarding the exchange of VAT-related information with other authorities (Figure 38). A large part of them (two thirds) do not exchange tax information at all with OLAF, Europol, and national social security agencies. Such information can be shared sometimes with other financial and criminal national authorities but only if they have expertise in field of VAT.

6.4.1.4. Qualitative assessment of the relevance of VIES for Member States needs and problems encountered when using VIES

Member States are divided as to whether data available through the VIES system meets all their needs ( Figure 11 39 below). The provisions relating to VIES were amended to increase the amount of information stored and exchanged, in particular in relation to the mini one stop shop 132 (MOSS). Member States are usually satisfied with the relevance of this information. Nevertheless, as set out under section 6.4.3.6., Member States also consider that new types of information should be exchanged in an automated manner. 0

Member States report difficulties in using VIES, in particular when it relates to the accuracy of data. Some Member States also consider as creating difficulties the existing restrictions to consult intra-EU supplies where neither the supplier nor customer is registered in the Member State of the official making the enquiry. Currently this access is only granted to Eurofisc liaison officials and under strict conditions.

6.4.1.5. Qualitative assessment of the relevance of automatic and spontaneous exchange of information in confront with other mechanism

Member States are divided as to whether automatic and spontaneous exchange of information could be replaced by other mechanisms ( Figure 18 41). As concluded under Chapters 6.1.1.4. and 6.1.1.5., Member States recognise that spontaneous and automatic exchanges of information can be effective, although these exchanges could be further intensified.

At the same time, it also results from the analysis under Chapter 6.1. that Member States do not fully engage in this type of instruments. This raises questions as to whether these two types of exchanges still meet the Member States' needs.

In addition, half of the Member States could support an alternative approach and mentioned that granting direct access either to tax officials from other Member States or to Eurofisc liaison officials could reduce administrative burden and improve the quality of the information exchanged. The transaction network analysis 133 (TNA) is here viewed as a solution to improve exchange and analysis of information (see section 6.4.3.1. below).

However, 13 Member States disagree with the possibility to implement other access to information mechanisms such as granting direct access to national databases.

6.4.1.6. Qualitative assessment of the extent to which Eurofisc continues to meet the needs of Member States

A majority of Member States agree that Eurofisc has a positive impact on the collection of VAT on intra-EU transactions ( Figure 12 26) although not all Member States participate in all working fields.

However, as reported by the ECA in its 2015 report, the information exchange within Eurofisc is to some extent rudimentary, slow and not user-friendly as it takes place using spreadsheets. The Eurofisc working field coordinators manually compiles and disseminates these spreadsheets among liaison officers of each participating Member State. This runs the risk of transmitting incomplete or wrong information. This audit also demonstrated that there are no common criteria or sources of information to perform risk analysis. Furthermore, feedback on the usefulness of the data exchanged is scarce. As a result, Member States participating in different working fields often exchange information that includes non-dubious traders, thereby wasting resources.

6.4.1.7. Overall assessment as to whether administrative cooperation instruments still fits the needs of Member States

The Member States generally recognise that the administrative cooperation instruments are relevant. Requests for information, Eurofisc, MLCs and administrative enquiries score particularly high. If in a general manner Member States can sometimes be faced with some difficulties when using these instruments, none of them seem to prevent them to engage in administrative cooperation.

With regard to difficulties encountered when exchanging information with other law enforcement authorities, tax administrations in most Member States indicate to have a very limited or no experience. This is particularly the case in relation to Europol or OLAF. It is difficult to rely on this very limited experience to draw conclusions.

It is also difficult to draw conclusions as to whether new ways to access information could be more relevant than spontaneous and automatic exchange of information. There is no clear view shared by all Member States on this. At the same time, most of them also consider that there is room to develop new automated exchange of information. The right balance here would probably be to develop and improve already existing means to exchange information, whichever their format. Direct access to national databases by foreign officials is certainly the limit that a majority of Member States would not accept to cross.

Overall, it seems that the automated exchange of information is viewed as a relevant instrument for a majority of Member States. Despite its imperfections, VIES is considered as a tool which should be further developed with the inclusion of new categories of data or additional information to that already present. Taking into account other comments, if new automated exchange of information was introduced, it would also be necessary to evaluate first whether VIES constitutes the most suitable tool to achieve the objectives. Indeed, new innovative IT solutions have been developed over the last 25 years and this may better suit the Member States' needs.

Finally, although not perfect, Member States consider Eurofisc as a relevant instrument given the objectives for which it was set up i.e. a network of liaison officials for swift and targeted exchanges of information to fight serious VAT fraud. Nevertheless, and according to replies to the questionnaire, after 7 years, most Member States think that Eurofisc is at a turning point and would like to see it modernised (see Chapter 6.4.3.1. below).

6.4.2. Views of stakeholders other than the Member States on VIES on the web

6.4.2.1. Qualitative assessment of the extent to which VIES on the web (VoW) continues to fit the needs of stakeholders other than the Member States

As detailed above under section 6.3.3. of this report, concerns were raised by stakeholders other than Member States regarding the accuracy of data made available to them by VoW. As regards the availability of the instrument (Figure 42), 33 respondents considered it to be sufficient.

35 respondents would like to have an automated notification system that would inform them on technical issues with VIES on-the-web. 46 respondents would prefer to be informed on the changes in their customers VAT details rather than checking themselves.

A large majority of respondent reported that some data are missing from VoW. For instance, name and address of the legal entity concerned or the period of validity of the VAT number should be added. Regarding services proposed by VoW, respondents asked for a batch processing for VAT checks with download possibility and for a reporting functionality in case of invalid VAT number to report potential fraudsters.

Overall, and beyond the question of accuracy and timeliness of information provided on VoW, this tool meets its users' needs and is therefore broadly used. Its relevance was never questioned and as the high number of users indicates, the tool is popular. The results from the open-public consultation and feedback received from the Member States show that adding new functionalities would increase the relevance and utility of VoW further. Therefore, although the relevance of VoW is not questioned, it could be improved to suit its users' needs even better.

6.4.3. Identified areas in need for further changes to Regulation 904/2010

6.4.3.1. Qualitative assessment of the need for change in the area of Eurofisc & Risk analysis

After 7 years of Eurofisc functioning, the areas where processes could be improved are well known. As reported by the ECA in its 2015 report, 27 Member States consider Eurofisc to be an efficient early warning system for fraud prevention, but they still pointed out the following weaknesses, which were also confirmed by the audit tests in Member States: (i) feedback was not frequent enough; (ii) data exchanged was not always well targeted; (iii) not all Member States participated in all Eurofisc working fields; (iv) exchanges of information were not user friendly; and (v) data exchanges were too slow.

Amongst other things, the ECA concluded that the Commission should recommend Member States to introduce a common risk analysis including the use of social network analysis to ensure that the information exchanged through Eurofisc is well targeted to fraud.

It results from the Member States consultation (Figure 45) that most of them support the development of this common risk analysis in Eurofisc. When commenting on common risk analysis within Eurofisc most Member States refer to the TNA as a tool that can facilitate such an approach. Most Member States also agree that TNA should be managed by Eurofisc and are willing to participate. Most Member States also support TNA having access to data necessary to build networks. A majority of Member States also support creating an explicit legal base to implement TNA.

Under current legal arrangements the Commission cannot have access to Eurofisc operational data which, in case of the Commission hosting TNA, will require complicated organisational arrangements and possibly additional development costs. Most Member States support granting the Commission a limited access to these data.

Stakeholders other than Member States that have responded also share the view that an automated joint risk analysis would be of help in fighting VAT fraud (47 strongly agree or agree).

6.4.3.2. Qualitative assessment of the need for change in the area of multilateral activities (MLCs, joint audits, administrative enquiries)

The joint audit is an instrument through which a taxable person is subject to a coordinated audit by a single audit team made up of tax officials of two or more jurisdictions. Regulation (EU) 904/2010 does not provide for joint audits but only for the passive presence of officials from another Member State during an administrative enquiry in the taxpayer premises.

A project group in relation to joint audits took place under the auspice of the Fiscalis 2020 programme. The participants in this project group came to the conclusion that there would be an overall benefit to joint auditing taxpayers. Joint audits could take place when this tool would be considered as more suitable than already existing joint actions. The group recommend the Commission to consider making a proposal to introduce this instrument in Regulation (EU) 904/2010.

According to replies received, only 4 Member States have experience with joint audits but none in relation to VAT. Member States are largely divided (Figure 47) as to whether there were cases in their work where joint audits would have been a more useful tool as compared to the tools currently offered by the Regulation (EU) 904/2010.

When it comes to possible benefits of joints audits (Figure 48) around half of the Member States either strongly agree or agree that joint audit would be a useful and cost effective tool to audit cross-border trader. They also consider that an audit by a single team of auditors would provide more legal certainty and would be more cost efficient for businesses undergoing these joint audits.

However, it is also noted that a number of Member States have no opinion on this and those Member States that are sceptical towards joints audits further explained that:

·the absence of a national and European legal base for joint audits is a concern;

·differences in approach towards audits in Member States may hamper the effectiveness of the instrument;

·asking a Member State of establishment to collect information at the request of another Member States may be more efficient than setting-up a team of auditors; and

·They are not completely sure of the extent to which the report produced within joint audits would be binding for all participating Member States.

Stakeholders other than Member States are also divided as to whether joint audits would be of interest (Figure 49). 19 respondents agreed and 17 disagreed that an active participation of foreign auditors would make it more efficient. 19 respondents could not decide.

19 respondents consider that a joint audit would be less burdensome than being audited at request of different Member States and 19 are neutral neither agreeing or disagreeing or having no opinion on this question (Figure 50). 40 respondents considered that having a single report at the end of the audit would provide taxpayers with more legal certainty (Figure 51).

Respondents widely agree that joint audit would constitute a step forward in administrative cooperation and would prevent duplication of tax audits which is burdensome for both tax administration and businesses. However, number of respondents notice that joint audits could be useful as long as a common unique procedure is clearly defined.

6.4.3.3. Qualitative assessment of the need for change in e-forms

Most Member States consider e-forms as relevant and useful. But they also agree that they could be redesigned ( Figure 17 52). Some of the remarks that were made in relation to possible updates:

·allowing request based on more than one company, in order to deal with cases of suspected fraud networks in the requested Member State. (IE)

·identifiers other than VAT registration numbers such as business name, contact details, website should be supported. (IE)

·forms should support EOI for import/export data, which are more and more often related to VAT MTIC fraud. (FI)

·mandatory use of one language in filling the forms should be implemented. (NL)

·forms should be made more user friendly e.g. by allowing corrections without having to fill the whole form from the start (NL), simplifying the form by showing only the field that are relevant to a particular request (IE).

A number of Member States (AT, CY, EL, LT, PT, SK and SI) suggested that a section specific to the MOSS should be added. However one Member State (UK) considers that questions in relation to MOSS can be addressed using free text fields and reports that forms should be kept simple without excessive amount of pre-defined questions. Regarding technical issue, one Member State (EL) suggests to implement an automatic system to cover sensitive data for cases where forms are requested by the Prosecutor’s Office for instance. Member States have high expectations towards future central application (eFCA) and believe that it is going to be a more convenient tool for sending requests.

6.4.3.4. Qualitative assessment of the need for change in the area of automatic exchange of information, including car registration data

Most Member State do not support dropping already existing categories or have no opinion on this matter ( Figure 19 53). Car registration information, data relevant for controlling e-commerce transactions and payment data held by financial institutions are the most popular information that Member States would like to see exchanged in an automatic manner. None of the other categories proposed receive a support of the majority of Member States.

A potential access to car registration data has been discussed several times in Standing Committee on Administrative Cooperation Expert Group (SCAC-EG) meetings. In this context it was agreed to make an in-depth study on the technicalities and related costs of some options to access Eucaris 134 data on cars. This should be considered as a new form of automated access to specific information stored in national databases. A vast majority of Member States consider that access to Eucaris data for tax purposes would be useful to control the VAT treatment of vehicles. Member States mentioned that having access to such data would help to detect of tax frauds with vehicles, in particular the abuse of margin scheme. This would also simplify the work of Eurofisc working field 2.

A number Member States (BG, CZ, SI, PT, AT, RO, FI, UK, IE, NL) would also like access to such application not to be limited only to Eurofisc officials while four (BE, IT, MT, LU) would like access to be limited to Eurofisc officials.

Concerning stakeholders other than Member States, 29 respondents considered that tax authorities should have automated access to data on cars registration from other Member States (Figure 54).

6.4.3.5. Qualitative assessment of the need for change in the area of customs data

In its 2015 report, the ECA mentioned that imports exempted from VAT under the Customs Procedure No 42 (CP 42) may lead to abuses and, consequently, to underpayments of the Member States’ VAT. The audit showed a lack of completeness of VIES data concerning imports under CP 42 and a lack of information sharing between tax and customs authorities. Therefore tax authorities were not able to cross-check customs data on imports under CP 42 with the VAT recapitulative statement and the VAT return submitted by the importer and the VAT return of the recipient of the goods.

Additionally, it was also indicated that with the exception of one Member States, no automatic checking of VAT numbers was available in the customs electronic clearance systems of the visited Member States. The ECA then recommended that The Commission should propose legislative amendments enabling effective cross-checks between customs and tax data.

According to their replies (Figure 38), most Member States exchange information with customs. In most cases Member States encounter problems either to a very low or to a low degree but they have not specified the nature of these problems. Member States note that often there is a single tax and customs authority, which makes it easier to provide VAT-related information to customs. Customs administration in most Member States also have access to data exchanged over VIES. Tax administrations think that information on imports made under CP42 should be made available by customs authorities (Figure 55). Most Member States would like to have automated access to such data.

Stakeholders other than Member States (Figure 56) are also of the opinion that tax administration should have automated access to information on exempt importations from customs authorities (17 respondents strongly agree, 24 agree).

As a whole, it seems that Member States' tax administrations are fully aware of the need to be in position to correctly check importations made under a CP42 procedure. To this end more information should be shared between tax and customs authorities so that each administration would be able to properly enforce the legislation it is responsible for.

6.4.3.6. Qualitative assessment of the need for change in the area of VIES

Access to VIES, in particular for Eurofisc liaison officials (Art. 21(2) of the Regulation

For the moment, VIES data are accessible to tax officials controlling intra-EU transactions involving their country. Only Eurofisc liaison officials have a broader access to VIES. Under strict conditions, this access covers all intra-EU transactions, even when there is no connection with their country According to comments received, a number of Member States (LV, SE, EL, PT, LT, HU, and IE consider that restrictions in access to so called ‘3rd Member State inquiry’ – a right to consult VIES on intra-EU supplies where neither the supplier nor customer is registered in the Member State of the official making the enquiry- is a difficulty they encounter when using VIES.

Type of data to be exchanged in VIES

When they are asked as to whether additional categories of information should be exchanged over VIES ( Figure 21 58), a vast majority of Member States report positives views. The question of car registration and customs data was already discussed in section 6.4.3.4. and 6.4.3.5. above. 14 Member States would also like to be able to check over VIES validity of so-called domestic VAT numbers 135 .

2 Member States have reported other categories of information that could be exchanged over VIES such as detailed reports on intra-EU acquisitions, information on shareholders and managers (LV), non-established taxable persons registration (CZ), information in relation to margin scheme (PT), NACE code (UK), MOSS registration number for third country company (NL), invoice registers data (LT), information on the allocation of VAT identification numbers to taxable persons established in another Member State (BG), total amount of sales towards a Member State in the field of e-commerce (BE). One Member State (IE) noted that all Member States should be required to show standardised data i.e. name, address, activity, registration number and date and also names of officers of the company or sole trade wherever available.

One Member State (HR) indicates that VIES should be modernised before implementing any changes to the system otherwise, additional information must be shared through other channels. Another Member State (FI) notes that VIES might not be the best tool for exchange of information.

Stakeholders other than Member States who have answered support at a vast majority (47 respondents) extending the scope of information exchanged through VIES (Figure 59).

VIES access granted to law enforcement authorities

As detailed by the ECA in its 2015 report, neither Europol nor OLAF have access to Eurofisc data. They also do not have access to VIES. It reduces Europol and OLAF’s ability to help Member States tackling VAT fraud.

When asked about granting access to VIES to Europol and OLAF, only a very limited number of Member States (2) reported that they already provide this access and only 5 and 6 of them would agree to do it in the future (Figure 60).

6.4.3.7. Qualitative assessment of the need for change in the area of strengthening cooperation with other law enforcement bodies

Beyond granting an access to VIES to EU law enforcement authorities, the ECA also recommended in its 2015 report to remove legal obstacles preventing the exchange of information between administrative, judicial and law enforcement authorities at national and EU level. In particular, the ECA recommended that OLAF and Europol should have access to Eurofisc data and Member States should benefit from intelligence information supplied by them. So far, Member States invoke Regulation (EU) 904/2010 and national tax secrecy rules to deny such access.

The ECA notes that it reduces Europol and OLAF’s ability to tackle VAT fraud through the identification and disruption of organised crime groups behind the carousels and even their ability to assess the real impact of intra-EU VAT fraud.

When asked about the sharing of intelligence between Europol and Eurofisc (Figure 61), 12 Member States agree, 4 disagree and 11 have no opinion. When it comes to providing OLAF with such intelligence, 8 Member States agree, 7 are against and 12 have no opinion. In case intelligence is shared with Europol and OLAF, 14 Member States agree that Europol should provide Eurofisc feedback on the use of this information. However, it seems that tax administrations are largely unaware of what information Europol and OLAF have and could not specify what sort of feedback they would expect. Should there be some form of exchange of information between OLAF and Europol on one hand and Eurofisc on the other hand, then Member States would like to see a step by step approach with information exchange happening on case by case basis.

12 Member States agree that it would be useful if Europol participate in Eurofisc working field meetings. 4 Member States are against such participating and 8 have no opinion. 10 Member States would agree with OLAF participation in Eurofisc meeting, 8 are against and 9 have no opinion on the matter. Member States further clarified that such cooperation should be limited to OLAF and Europol participation in Working Field 4 meetings dealing with VAT fraud trends. One country saw a problem in such participation as in their view there cannot be direct cooperation between Europol and Eurofisc as Eurofisc officials are from tax administrations.

As regards stakeholders other than Member States, 32 respondents considered that the current arrangements for cooperation between tax authorities and law enforcement bodies are not effective and 34 respondents consider that the cooperation between tax authorities and law enforcement bodies should be changed. 42 respondents considered that Eurofisc should share relevant information on serious VAT fraud with OLAF and 45 respondents with Europol.

In this context, 28 respondents considered that the fight against VAT fraud would justify more exchange of personal data between tax authorities and law enforcement bodies and 45 that such exchanges should be proportionate to the objectives.

6.4.3.8. Overall assessment of identified areas in need for further changes to Regulation (EU) 904/2010

It results from the consultation of Member States that the introduction of a dedicated legal basis to implement TNA receives a broad consensus with a majority of Member States agreeing to grant the Commission the necessary access rights to perform its duties in the most cost-efficient fashion.

Whilst Member States do not reject the introduction of joint audits in Regulation (EU) 904/2010, they remain divided as to whether such instruments would bring an added-value and question its relevance and the situations in which it could be used.

Overall, Member States are satisfied with the relevance and content of e-forms for the exchange of information although 16 of them consider that they should be redesigned. Obviously, new needs are expressed in particular when new legal provisions are implemented (MOSS) or new fraud trends are encountered.

There is certainly room to further develop automatic and automated exchange of information. Amongst categories of information that Member States consider as the most relevant, car registration information, data relevant for controlling e-commerce transactions and payment data held by financial institutions are the most popular. With regard to VIES, Member States still view it as a relevant and needed instrument. They consider that more categories of information could be exchanged in an automated fashion with broader access to tax officials than today.

Cooperation with customs authorities is viewed as essential and most Member States tax administrations would welcome more exchanges of information with these authorities. They also consider that automated exchange of information would constitute the best instrument to achieve their expectations and be in a better position to fight frauds in relation to CP 42.

Concerning exchange of information with law enforcement authorities, it results from the consultation that Member States are more open to exchanging intelligences with Europol and OLAF than giving them an access to VIES. As regards their participation in Eurofisc, views are more balanced with less Member States rejected upfront the possibility of enhanced exchanges with Europol and OLAF, in particular when it comes to participation in working field 4.

Summary of findings: to what extent the provisions of Regulation (EU) 904/2010 continue to correspond to the needs of the Member States? (relevance)

Statistics provided by Member States and detailed under Chapter 6.1 demonstrate a good appropriation by Member States of the instruments of Regulation (EU) 904/2010 and a maturity in their use.

Experience in exchanging information with OLAF and Europol is rather limited or even absent. It is difficult to draw conclusions as to whether difficulties are encountered by Member States when exchanging information with these law enforcement authorities. Results from the Member States consultation show a limited appetite to grant these two authorities an access to VIES. Participation in Eurofisc working field meetings or exchange of intelligence between Eurofisc, Europol and OLAF seem to be better avenues to improve cooperation between authorities involved in fighting VAT fraud. Member States also consider that more information should be shared with customs authorities so that each administration would be able to properly enforce the legislation it is responsible for.

It appears that automated exchange of information is viewed as a relevant instrument for a majority of Member States. Despite its imperfections, VIES is considered as a tool whose potential may be further developed with the inclusion of new categories of data or additional information to that already present. With regard to VoW, new functionalities would help private stakeholders to perform their VAT obligations in a more efficient fashion. Although the relevance of VoW is not questioned, the instrument could be improved to better suit the users' needs.

Finally, although not perfect, Member States consider Eurofisc as a relevant instrument. Nevertheless, after 7 years of functioning, most Member States think that Eurofisc would like to see it modernised. To this end, the introduction of a dedicated legal basis to implement a joint processing of VAT data receives a broad consensus.

The Member States generally recognise that the existing administrative cooperation instruments are relevant. However, the interest they show in further developing or enhancing some of them demonstrate that they could better meet their needs.

6.5. To what extent are the provisions of Regulation (EU) 904/2010 in line with other policies and priorities of the EU? Could Member States have achieved similar results without acting at EU level? (coherence and EU added value)

6.5.1. Administrative cooperation in VAT is a cornerstone for the proper functioning of the single market and the fight against VAT fraud

6.5.1.1. Qualitative assessment of the level of convergence of the objectives of Regulation (EU) 904/2010 with other Commission's strategies

Traditionally, initiatives for administrative cooperation have always been based on the "mutual" interest of the contracting States. Nevertheless, administrative cooperation in the field of VAT between EU Member States also serves another objective: it contributes to the proper functioning of the EU VAT area that is a key component of the single market. Common EU VAT rules are necessary to allow the free circulation of goods and services and prevent any distortions that would emerge if such rules were not harmonised across Member States. Common EU VAT rules ensure fiscal neutrality for economic decisions of EU businesses and simplify cross-border transactions between Member States and ensure the VAT revenues are protected.

In this sense, administrative cooperation in the field of VAT is not only an instrument whereby Member States can provide mutual assistance to each other, but an EU policy whereby, through a set of rules governing how the Member States exchange information and cooperate on VAT matters, it actively pursues the wider EU objectives of a proper functioning EU VAT area and of tax fairness and fight against tax evasion and avoidance. As such, administrative cooperation organised at EU-level is fully coherent with wider EU strategies, policies and priorities.

Firstly, administrative cooperation in the field of VAT is part of a more global administrative cooperation framework between Member States for taxation matters, direct and indirect. Administrative cooperation in the field of direct taxation is also organised at EU level under the umbrella of the Council Directive 2011/16/EU of 15 February 2011. The same goes in the area of tax recovery assistance on the basis of Council Directive 2014/24/EU of 16 March 2010. In all areas, administrative cooperation and mutual assistance is organised at EU level so that all Member States provide the same level of assistance to each other and share a common objective of safeguarding their tax revenues and protecting market rules. In an increasingly open economy, with new patterns of trade and new trading opportunities, the Member States are drawn into a global 'playground' where interest fuse and become shared. So do, as an undesired consequence, the problems, which can effectively be addressed uniquely by Member States working together. No Member State single-handedly would be able to solve problems related to VAT in a cross-border context.

Secondly, on 7 April 2016, the Commission presented a Communication to the European Parliament, the Council and the European Economic and Social Committee on an action plan on VAT 136 setting out ways to modernise the VAT system so as to make it simpler, more fraud-proof and business-friendly. As described under Chapter 2 of this report, improving administrative cooperation in the field of VAT and Eurofisc has been considered as an essential instrument to achieve these objectives and further action in that area was explicitly called for.

Thirdly, 20 Member States have recently reached a political agreement on the establishment of the new European Public Prosecutor's office under enhanced cooperation. The European Parliament will have to give its consent. Once in place, the independent EU public prosecutor will be equipped with the power to investigate and prosecute criminal cases affecting the EU budget, such as corruption or fraud with EU funds, or cross-border VAT fraud. It will be a strong, independent and efficient body specialised in fighting financial crime across the EU. The same goes for the adoption of the Commission proposal for a Directive on the fight against fraud to the Union's financial interests by means of criminal law to cover serious EU-wide VAT fraud.

It results from the above that administrative cooperation is a fundamental aspect of the common EU VAT system and the single market in general. Furthermore, there are at the moment several initiatives whose purpose is to put the emphasis on fighting the most serious VAT threats. Many of them are cross-border offences, exploring the loopholes of the current VAT system. Improving VAT administrative cooperation between Member States and in particular Eurofisc is completely coherent with these initiatives.

6.5.1.2. Qualitative assessment of the inter-dependence of provisions governing administrative cooperation in the VAT with other Commission's strategic initiatives and programmes (e.g. the developments in the VAT area towards the definitive VAT system, Fiscalis activities)

The first Administrative Cooperation Regulation in the field of VAT was introduced along with the abolition of the physical borders across the EU for the trade of goods. Further, implementing all VAT rules, including administration cooperation provisions, was a prerequisite for each country outside the EU willing to join it.

Each time the VAT system was improved or expanded, this was accompanied by additional administrative cooperation instruments to support and permit these developments. Can be mentioned here the modernisation of cross-border VAT rules to business to business transactions on services, the creation of the mini one stop shop, the adoption - under way - of new rules in relation to e-commerce, and at a later stage the definitive VAT regime. It is not conceivable to implement new VAT rules at EU level if improved cooperation rules between Member States creating a deterrent effect to this set of common rules are not adopted at the same time.

6.5.2. Administrative cooperation in VAT contributes to the creation of a single EU VAT area

6.5.2.1. Qualitative assessment of the Regulation (EU) 904/2010's contribution to the creation of a common system for cooperation between the Member States

Providing mutual assistance and exchange information in the field of VAT is in the interest of all EU Member States (and of their citizens and companies), despite the unequal use of the assistance framework. As described above, the problems related to VAT evasion and avoidance are becoming increasingly cross-border, demanding a coordinated approach of the countries involved. Many a time it will not be possible to say at the first glance which country's interests are at stake because the fraud schemes are getting ever more sophisticated, cutting across territories and markets. In this sense, it is the interest of all Member States to be equally involved in active exchange of information and cooperation, with the view not only to react on fraud detected, but also, and more importantly, prevent future attempts. That common interest should be an incentive for all Member States to provide sufficient resources for administrative cooperation in the EU and adhere fully to the principle of reciprocity.

The extent to which the EU VAT administrative cooperation framework is used differs from one Member State to another. Moreover, due to the market patterns, geographical location, socio-demographic factors (including cultural) or size of economies, some Member States are more often requested to provide assistance than others.

Even more, it results from figures presented under chapter 6.1. of this report that the most recent Member States are those that have viewed the most important increase in the use of Regulation (EU) 904/2010 over the period 2011-2016. This demonstrates that this Regulation is not only a means to sustain the single market and the free circulation of goods and services but a necessary instrument to properly enforce VAT legislation and liabilities in each Member State. This cooperation is best organised at EU level since it would otherwise be difficult for Member States to reach similar convergences of objectives and efforts on the basis of bilateral agreements or non-EU multilateral instruments.

Furthermore, all authorities can use the common electronic request forms and uniform instruments (uniform notification form and uniform instrument permitting enforcement in the requested Member State). The use of common rules and common forms – with an automated translation – considerably facilitates the work of the authorities dealing with administrative cooperation. This constitutes a major advantage for the cooperation between the Member States' tax authorities. Taking this simple example of any common EU e-form, one could imagine a situation where every country, or group of countries, would come up with their own form, or other means, to ask each other for assistance. Each of these countries would then need to make such requests somehow communicating to their national systems in order to pull the asked for data. The result would be an incredibly dense and burdensome web of non-inter-operable, linguistically and methodologically challenging, non-uniform means through which Member States would be attempting to assist each other. The overall effect is not difficult to picture and the effectiveness of cooperation based on such approach would be completely disproportionate to the effort put it (assuming Member States do actually bother), showing the great added value of agreeing on the rules governing such cooperation and providing for common instruments to do so as effectively and efficiently as the circumstances permit. What is important to note is that any such forms are actually discussed and agreed on amongst the Member States, which makes them fitting their needs, administrative and technological capacities as best as possible. This is explained in further detail under section 6.5.3.2.

The Commission services are currently building a central application aimed at encompassing all electronic forms for all taxation domains. This central application will allow further streamlining and rationalisation of electronic forms, ensuring quick modification at Commission level, tackling new challenges in the field of exchange of information, while reducing drastically the deployment costs at EU and Member States levels. This development will allow to simplify the lay-out of the e-forms and to make them still more user-friendly. All Member States have been consulted on the development of the central platform design for the request forms. 137

6.5.3. Joint EU approach in administrative cooperation in VAT has advantages over other forms of national and international forms of tax cooperation

6.5.3.1. Cases of cross-border fraud (e.g. MTIC), MLC indicator, VAT gap)

The EU is the only area with an integrated VAT system worldwide. Moreover, it results from discussions in fora outside the European Union, at OECD level for instance, that the EU is the only area with an in depth and long-standing close cooperation in the field of VAT. When other organisations only start to think about exchanging information on VAT or developing other forms of close cooperation, the EU can already rely and gain from a 40 year experience with administrative cooperation between its Member States. This has a great value and has no comparison worldwide.

Hence, if with respect to administrative cooperation in the field of direct taxation several mutual assistance instruments can be used, this is far less the case in the field of VAT. Bilateral double tax conventions are almost not used to this end and the joint OECD/CoE multilateral convention, if promising, is only an emerging instruments that is not really used by countries with a long-standing and established tradition of administrative cooperation. It results from this that if VAT administrative cooperation would not be organised at EU level, Member States would not be in a position to individually achieve the same objectives since they could not rely on any other agreements, be it bilateral or multilateral.

When it comes to cross-border transactions, an area where VAT fraud is severe, mutual assistance between Member States is essential to achieve tangible results. National tax authorities must enforce their national VAT legislations. In relation to cross-border transactions they cannot do it without the support of their EU counterparts because many of the problems they aim to tackle are not restricted to their own territory. VIES was first implemented to enable national administrations to determine whether goods have been introduced VAT free onto national markets. Further, it has been considered that more powerful and targeted administrative cooperation instruments should be introduced between Member States to put an end to VAT fraud and close, at least to some extent, the VAT gap. This is precisely why Member States have decided to create Eurofisc and more recently to reinforce it by setting up two new working fields and implementing the transaction network analysis. All this needs to be implemented at EU level in a coordinated manner as otherwise it would create loopholes that could be used by fraudsters to abuse the system.

Over the course of the open public consultation, views of stakeholders other than Member States tax administration on administrative cooperation were sought (Figure 62). It results from this that 49 respondents agree or strongly agree that the EU is the best placed to provide Member States with common rules to allow them to work more closely together in order to monitor intra-EU trade for VAT purpose and ensure the correct application of VAT. Furthermore, 53 of them consider that the EU should assist the Member States so that they use these tools to the largest extent possible and 89 % that should ensure that the Member States use these tools to the largest extent possible.

6.5.3.2. Exchange of best practices

The EU tradition of providing different programmes of administrative cooperation is also long-standing. Beyond operational exchange of information instruments, the EU also provides a framework where tax authorities can exchange best practices or meet to discuss specific issues and try to find responses or develop common approaches.

To this end, and amongst other things, the Fiscalis programme 138 was developed. Different types of cooperation can take place under Fiscalis such as project groups, workshops or seminars. In particular, a number of recent initiatives answering needs expressed by Member States in the field of VAT administrative cooperation have been developed:

·Eurofisc organised under 7 different Fiscalis project groups, a group for coordinators and 6 for each working field;

·The multilateral control platform to improve the functioning and use of MLCs and to improve their overall quality;

·control of e-commerce whose purpose was to find new sources of information and new ways of cooperation to detect frauds in this sector;

·joint audit to analyse in what circumstance this instrument could be useful and would it have an added value compared to other tools and analyse the legal and organisational obstacles to organise a joint audit;

·transaction network analysis to define the user needs and requirements, specify the exact operational processes, security plan and terms of reference, including roles and responsibilities of stakeholders;

·domestic VAT listings : VAT listing allow tax authorities to collect data in addition to what is being reported in VAT returns in particular detailed data on incoming and outgoing invoices for a specific reporting period. The project group is looking at potential benefits and threats of such a system.

In addition, seminars are also organised, in particular Head of Central liaison office seminars where cross-cutting issues in managing VAT administrative cooperation can be discussed amongst Member States.

Summary of findings: To what extent are the provisions of Regulation (EU) 904/2010 in line with other policies and priorities of the EU? Could Member States have achieved similar results without acting at EU level? (coherence and EU added-value)

Administrative cooperation is essential for the good functioning of the EU VAT area which itself is a cornerstone of the single market. Common VAT rules have been accepted by the Member States because they had first been supported by reciprocal, binding and evolving administrative cooperation rules. As such, administrative cooperation is an essential component of the free circulation of goods and services which is one of the four freedoms enshrined in the EU.

Administrative cooperation is also coherent with other EU policies such as the fight against the most serious threats to the financial interests of the EU and of the Member States and ensuring tax fairness for the European businesses. Improvements in that area have been explicitly called for in the VAT Action Plan as one of the key action to be taken at the EU level. Well-functioning administrative cooperation goes in the same direction as the initiatives to create a European Public Prosecutor's Office or to implement rules to protect the EU financial interests.

It results from this evaluation report that the EU is the most VAT integrated area worldwide with a long-standing and unique experience in providing administrative cooperation. All details provided above demonstrate that Regulation (EU) 904/2010 is widely used by all Member States. Furthermore the EU has been in a position to create a unique, binding and appropriate administrative cooperation landscape to fit the needs of tax authorities to properly enforce the VAT legislation. In addition, other forms of cooperation taking place under the Fiscalis programme also demonstrate the need, for Member States, to use the EU level as the platform of reference to discuss and develop administrative cooperation in the field of VAT.

7. CONCLUSIONS

1. Member States’ opinion is overall positive on the legal and practical framework implemented with Regulation (EU) 904/2010. The vast majority of them consider that it has contributed to improve the administrative cooperation between Member States. Amongst the instrument provided by the Regulation, exchange of information on request, automated exchanges (VIES), Eurofisc and multilateral controls are viewed as the most effective instruments in the field of administrative cooperation.

2. Drawbacks in the manner in which administrative cooperation takes place are however reported. Late replies to requests for information are the most significant one. Although in decrease, the number of instances where it happens remains significant - 33% - and Member States should work to remedy it. The accuracy of data exchange through VIES is a domain where concerns are expressed by both the Member States and the businesses. Businesses need accurate and up-to-date information to properly apply the VAT rules to their intra-EU transactions.

3. As regards the efficiency of the administrative cooperation instruments, most Member States consider that costs incurred by participating in the administrative cooperation in the field of VAT are to a high extent proportionate to the benefits achieved. Automated access to information through VIES, Eurofisc, simultaneous controls, request for information and administrative enquiries score particularly high. These instruments are considered as the most efficient instruments by the Member States.

4. However, the Member States consider some instruments as rather burdensome. This is the case for exchanges of information on request and multilateral controls. This is the result respectively of the large number of instances where this tool is used and high resources required from tax authorities to carry out such controls.

5. While Member States report a high level of satisfaction with regard to the effectiveness and efficiency of the Regulation (EU) 904/2010 instruments, it has not been possible to ascertain the extent to which they have contributed to prevent budget losses. Nevertheless, Member States consider that the costs incurred by administrative cooperation are proportionate to the benefits. They also consider that inaccuracies in the information exchanged or a lack of commitment of their counterparts can have a detrimental effect on the ability to assess VAT. Nevertheless, it seems that thanks to Regulation (EU) 904/2010, they are able to collect additional VAT with positive effects on the prevention of budget losses.

6. Eurofisc has been introduced in Regulation (EU) 904/2010 as an early warning mechanism to facilitate multilateral cooperation to fight VAT fraud. A vast majority of Member States agrees that this new instrument has had a positive impact on the collection of VAT on intra-EU transactions and Member States report a high level of satisfaction. However, there are still drawbacks regarding its functioning leading to the conclusion that Eurofisc has certainly not reached its potential yet.

7. The Member States consider that all administrative cooperation instruments in the field of VAT are relevant. Some instruments score very high such as VIES and requests for information. If Eurofisc and simultaneous controls rank a little bit behind, overall, almost all Member States consider that these two instruments are relevant to a high extent. Feedbacks and requests for administrative notification are the less best-rated instruments.

8. Developing new instruments or new ways of cooperating seems to be needed. In particular, Eurofisc appears to be at a turning point as Member Sates particularly support the possibility to implement a joint process of VAT data. There is also room to further develop automated exchange of information or access to new sets of data. In this context, Member States are particularly interested by an access to customs data or cars registration information. The best way to access it still needs to be discussed, although Member States are more in favour of automated than automatic exchanges.

9. Exchange of information with EU law enforcement authorities remains a sensitive area. Participation in Eurofisc working fields meetings or exchange of intelligence between Eurofisc, Europol and OLAF seem to be better avenues to improve cooperation between authorities involved in fighting VAT fraud at EU level than granting these two authorities an access to VIES.

10. Improving administrative cooperation in the field of VAT is fully coherent with other EU policies currently under development. This is particularly the case with regard to mutual assistance in the field of direct taxation where a number of improvements took place over the recent years. It is also coherent with other EU initiatives underway such as the draft Directive to protect the EU financial interests or the European Public Prosecutor's office. All this demonstrates that there are several common initiatives taken at EU level with similar objectives: improving cooperation between law enforcement authorities and finding new way of fighting the most severe threats to tax revenues.

11. Administrative cooperation is essential for the proper functioning of the EU VAT area which itself is a cornerstone of the single market. Common VAT rules have been accepted by the Member States because they were supported by reciprocal, binding and evolving administrative cooperation rules. As such, administrative cooperation is an essential component of the free circulation of goods and services which is one of the four freedoms enshrined in the EU. Other forms of cooperation between Member States also demonstrate the need to use the EU level as the platform of reference to discuss and develop administrative cooperation in the field of VAT.

9.4. Annex 4: Consultation of Members States

Introduction

In the context of evaluating current arrangements for administrative cooperation and fight against fraud in the field of VAT the Commission sent out a targeted consultation to Member States' tax administration. The consultation sought to gather Member States' views on:

-effectiveness of the current arrangements for administrative cooperation to assess to what extent the objectives of the intervention have been achieved;

-efficiency of the current arrangements for administrative cooperation to assess to what extent the costs borne are proportionate to the benefits;

-relevance of the current arrangements for administrative cooperation to assess to what extent the objectives of the intervention are still corresponding to the needs of the Member States;

-coherence of the current arrangements for administrative cooperation to assess how its various internal components operate together to achieve the objectives of the intervention;

-the EU added value to assess to what extent having common rules and tools at EU level makes the difference compared to what Member States could achieve at a national level.

The Commission also sought Member States opinion on the possibility to introduce additional instruments and enhance existing ones for cross-border administrative cooperation: the possibility to expand data sources available to tax administrations, possible ways to ensure that law enforcement authorities have access to relevant VAT data in order for them to perform their tasks and, in particular, to carry out criminal investigations when necessary.

This report summarises the replies provided by the Member States to that consultation. Chapter 1 makes general conclusion on Member States opinions on currently available tools for administrative cooperation and fight against fraud. Chapter 2 goes more in depth and aims at evaluating efficiency, usefulness and relevance of current intervention. Chapter 3 looks at burden and costs associated with participation in administrative cooperation and fight against fraud in the field of VAT. Chapter 4 summarises Member States opinions on possible changes to current framework for administrative cooperation and fight against fraud in the field of VAT.

The consultation was made available to Member States on 2nd of March 2017. 27 Member States provided their replies. Germany is yet to provide the replies to the consultation and therefore Germanys' views are not taken into account in this report.

1. General conclusions of the targeted consultation

Current arrangements for administrative cooperation and fight against fraud meet Member States' needs either to a very high or to a high extent. Member States mostly either strongly agree or agree that new provisions introduced into the framework of administrative cooperation or fight against fraud in the field of VAT under Regulation (EU) 904/2010 have improved their ability to monitor cross-border transactions and to collect VAT, improved quality, reliability and timeliness of information exchange as well as contributed to increased legal certainty for traders.

Majority of Member States are of the opinion that cost of participating in administrative cooperation is proportionate to the benefits achieved although opinions vary across the range of tools offered under current legal framework.

Automated access to information is highly valued by the Member States. Ability to check information over VIES system is considered to be very relevant for control of intra-EU transactions and efficient in collection of VAT. This tool also offers the biggest benefits in relation to costs. Extended rights to consult VIES for Eurofisc officials are also highly valued by the Member States. However many Member States consider that limiting those rights only to Eurofisc officials along with additional conditions is an ineffective out of date arrangement that hinders efficient fight against fraud.

Eurofisc is considered to be an effective mechanism for fight against fraud by the majority of Member States. However not all Eurofisc Working Fields are equally effective – while Working Field 1 and 4 rank high then in relation to Working Field 2 and 3 further actions are needed to improve targeting of signals. Issues with data quality, timeliness of feedback and targeting of early warning signals remain. Majority of Member States have high expectation towards TNA and are looking forward to having this tool at their disposal.

Collaboration with law enforcement remains limited with only few Member States exchanging VAT relevant data. Very few Member States would like to see granting law enforcement agencies access to VAT relevant data.

Cooperation with customs is of growing importance. Some Member States mention that VAT frauds often involve transactions across the external border making information exchange with customs important. Majority of Member States seem to exchange VAT relevant information with customs and would like to have access to data on importations using Customs Procedure 42.

Member States would like to have access to car registration data to better fight against VAT fraud.

Member States rather agree than disagree that joint audits could potentially be a useful tool in auditing companies involved in intra-EU trade. However a number of issues were raised in relation to rights of the auditors participating in such audits and extent to which the decision of the audit is binding.

2. Efficiency, usefulness and relevance of current intervention

Member States are fairly satisfied with administrative cooperation and fight against fraud in the field of VAT instruments provided by Regulation (EU) 904/2010. 26 Member States have replied that the Regulation meets their needs to a very high or to a high extent ( Figure 2 ).

Member States’ opinions are overall positive regarding changes introduced to administrative cooperation framework in 2010 ( Figure 3 ). One Member States (CZ) thinks that the quality of information exchange has improved mainly due to introduction Eurofisc. Another Member State (IT) notes that in relation to Eurofisc the quality of information remains an issue though signals received led to numerous follow up actions. Three Member States (SE, FI and LU) insist on the growing need for the exchange of information to be faster.

2.1. Requests for administrative enquiries, exchange of information without prior request, feedback, presence in administrative offices and during administrative enquiries, simultaneous controls

A majority of Member States find that tools provided by Regulation (EU) 904/2010 are either very effective or effective when it comes to monitoring the correct application of VAT on cross-border transactions and ensuring their ability to collect VAT ( Figure 5 ). Member States all declare that request for information and administrative enquiries (art 7), spontaneous exchange of information without prior request (art 15), automated access to data through VIES (art 21), simultaneous controls (art 29) and Eurofisc (art 33) are overall relevant for the control of intra-EU transactions and the collection of VAT ( Figure 4 ).

2.11. Request for information and administrative enquiries (art 7) and feedback (art 16)

Member States consider request for information and administrative enquiries to be effective and relevant for control of intra-EU transactions and collection of VAT. However Member States consider replying to such requests to be relatively burdensome ( Figure 15 ) but the burden seems to be proportionate to the benefits achieved ( Figure 16 ). Member States note that inaccurate/incomplete requests, multiple requests in relation to one trader, problems with interpretation of open-field questions create additional burden. One Member State (IE) mentioned inability to send integrated requests on multiple connected/related entities as a drawback of current implementation. Chapter 0 lists possible improvements to the forms used by Member States to send requests.

25 Member States noted that late replies to request for administrative cooperation at least to some extent have negative impact on their ability to collect VAT ( Figure 6 ). A number of Member States mentioned that due to domestic deadlines for conducting audits, replies often arrive past such a deadline making the information useless. However it was also noted that quick but of poor quality replies are useless as well. Member States also mentioned that the need to send reminders when the reply is late creates additional burden for liaison officials. One Member State (CZ) noted that sending and treatment of reminders varies from amongst Member States as this procedure is not covered by the Regulation (EU) 904/2010. Another Member Stated (IE) noted the complexity of the forms used for sending requests and lack of awareness of the existence of the tool admitting though that it is up to Member States themselves to educate their staff and popularise the tools.

Member States note that often there are difficulties with obtaining the feedback which can be explained that under current rules providing feedback in not obligatory. Chapter 0 describes Member State views on what can be changed in relation to feedback mechanism.

2.1.2. Exchange of information without prior request

Member States consider automatic and spontaneous exchanges of information to be effective and relevant for collection of VAT from intra-EU transactions.

However under current rules Member States can abstain from such exchange of information where collection of necessary information would require imposition of new obligations on persons liable for VAT or would impose a disproportionate administrative burden on the Member State. More than half of the Member States are affected by Member States abstaining from exchanging information without prior request at least to some extent ( Figure 7 ) which negatively affects Member States' ability to collect VAT. Member States note that the impact on VAT collection of Member States abstaining from exchanging information is hard to evaluate as Member States do not know precisely what is not being exchanged. A number of Member States (BG, LV, HR, LT, PT and UK) insist that all Member States should participate in automatic exchanges of information and provide complete, accurate and properly targeted information. Some Member States note that non-participation to exchange of information hinders administrative cooperation. However two Member States (AT and LU) believe that automatic exchange of information is hardly useful and the cost/benefit ratio is low. One Member State (FI) noted that when Member States send information spontaneously once year in bulk, it can decrease the usefulness of such information for the receiving Member State. Another Member State (UK) stressed the importance of participation in such information exchange claiming that non-participation to automated-exchange of vehicle information has a greater negative impact than non-participation to non-established taxable person (NETP) information exchange.

2.1.3. Request for administrative notification

Request for administrative notification seems to be the least relevant and effective tools for ensuring Member States' ability to collect VAT. One Member States (SI) notes that it is not used much and has a limited impact on VAT assessment due to differences in procedures within the EU. The only problem mentioned in relation to use of this tool is linguistic as requests often must be translated.

2.1.4. Presence in administrative offices, presence during administrative inquiries and simultaneous controls

Member States note that the presence in administrative offices and presence during administrative inquiries (PAOE) are rarely used due to language barriers and lack of awareness as regards to the usefulness of the tool. Absence of common rules in relation to usage of those tools hampers their use. As a consequence, each Member State approaches these tools differently which creates confusion for auditors. To deal with these issues, one Member State suggests creating a separate CCN mailbox for PAOE’s. Some Member States are of the opinion that these tools should be promoted more actively in particular at national levels. At EU level workshops could be held to promote these tools, similarly to those organised for simultaneous controls.

One Member State (UK) noted that currently presence in administrative offices is used by Member States to examine records relating to one of their businesses that are held in another Member State. This approach seems to be beneficial as it speeds up the audit and therefore reduces the administrative burden. However in such cases the requested Member State is not conducting an audit itself and therefore the official from the requesting Member State is not present during the administrative enquiry. It was suggested that this practice can be harmonized by allowing a Member State to conduct and audit in another Member State (with the consent of the business and the other Member State administration) if the records of the first Member State business are held in the second Member State.

In relation to simultaneous controls Member States are of the opinion that the tool is relevant for control intra-EU transactions and effective in ensuring their ability to collect VAT. However two Member States (LV and NL) note that procedure for simultaneous control is ineffective and needs to be provided with more resources. Mismatches in national legislation create further obstacles for the effectiveness of this tool. It was also noted that such controls often take too long which decreases their efficiency. One Member State (LV) mentioned that the option of combining simultaneous controls and PAOE should be foreseen.

2.2. VIES

Member States generally find the data shared through VIES system rather useful ( Figure 8 ) mentioning that such data is necessary to guarantee the correct application of VAT exemption in intra-EU supplies, to identify potential VAT risks, to minimize the number of information requests and to make those requests more targeted. Some Member States mention that usefulness of VIES data is undermined by retroactive changes and complain about not being able to view the history of those changes. All Member States either strongly agree or agree that inaccurate VIES data has a negative impact on their ability to collect VAT. Member States mention that inaccurate data creates administrative burden as tax administration have to conduct additional checks and send request for information that could have otherwise been avoided.

Member States are split on the question if data available through VIES system meets all their needs ( Figure 11 ). Member States made a number of suggestions on what additional data could be made available through VIES that are further explained in Chapter 0.

All Member States find automated access to VIES to be effective ( Figure 10 ) and agree to a very high or high extent that it is relevant ( Figure 4 ) for controlling intra-EU transactions and collecting VAT. However multiple Member States mentioned that current restrictions to send so-called "third Member State requests" into VIES listed in article 21.2(e) of the Regulation (EU) 904/2010 create additional complications (see Chapter 0 for more details).

More than half of the Member States have received complaints from traders in relation to Vies-on-the-Web data ( Figure 9 ) which are generally related to accuracy, completeness, and availability of information provided by the system. Three Member States (SE, LT and CZ) mention that retroactive deregistration from VAT registers can cause problems for business too. Few Member States report discrepancies between data reported by VoW and national databases. For example in one Member State (LV) VoW reported the VAT number to be valid despite it being deregistered from national database. One Member State (UK) brings to attention the fact that a business not having notified administration that it has engaged in intra-EU supplies does not appear in the system after registration.

2.3. Eurofisc

A majority of Member States agree that Eurofisc has a positive impact on the collection of VAT on intra-EU transactions ( Figure 12 ) although not all Member States participate in all working fields. It seems that the Working Fields where all Member State participate (Working Field 1 and 4) are considered to be the most effective by Member States.

Some Member States mention that the quality, swiftness and targeting of information should be further improved, in particular in relation to Working Field 2 and 3. Member States also note that the current medium for information exchange – online portal CircaBC – is not user friendly. A number of Member States made suggestions on how Eurofisc can be improved that are further described in Chapter 0.

Member States mainly agree that extended right to consult VIES for Eurofisc officials is useful and an effective early warning tool positively impacting collection of VAT on intra-EU transactions for it enhances fast and accurate detection of fraudulent cross-border transactions chains ( Figure 13 ). The only Member State that disagrees that extended rights of Eurofisc officials to consult VIES (FI) explains its' assessment with limitations contained in article 21 paragraph 2 point "e" mentioning that such limitations are out-of-date, old-fashioned compromise by the Member States and do not support fight against fraud (see Chapter 0 for more details).

3. Burden and costs associated with participating in administrative cooperation

A majority of Member States agree that participating in administrative cooperation is not very burdensome – to some extent at least ( Figure 15: Burden in relation to participating in administrative cooperation. Figure 15 ). Most Member States strongly agree or agree that costs associated with participating in administrative cooperation are proportionate to the benefits ( Figure 14 ).

When it comes to assessing burden in relation to particular tools offered by Regulation (EU) 904/2010 it seems that the largest burden is associated with tools that require manual work on individual cases: dealing with request for administrative enquiries, replying to request for feedback and participating in multilateral controls ( Figure 15 ).

Member States note that incomplete, inaccurate and unclear request and replies as well as multiple requests for a same taxpayer in a short period of time cause additional burden for tax administrations.

Although a large part of Member States does not consider that replying to feedbacks is excessively burdensome, some of them note that feedbacks are deprived of any additional value as regard to the resources it requires. Some Member States note that simultaneous controls as well as presence in office and during enquiries can be burdensome for they require more organisation and operational work than using usual request for information tool.

Some Member States report lack of internal resources and not very effective national procedures which impact quality and timeliness of their replies and limits their scope of action. One Member States (SK) suggested that European Commission could support those Member States that face high number of request though the content of such support was not specified.

Member States mentioned additional issues that increase the burden of participating in administrative cooperation:

-language barrier, need to translate enquiries and often poor quality of those translations,

-difference in national tax procedure within the EU is an issue that should be addressed to facilitate administrative cooperation,

-Insufficient capacity at domestic level to collect data for automatic exchange of information,

-Multiple requests concerning the same taxpayer (e.g. request in relation to different taxable periods) create additional complications.

Unfortunately very few Member States provided monetary estimations of the costs and benefits in relation to participation in administrative cooperation. Member States note that costs and benefits resulting from the use of each specific tool can hardly be estimated as many of tax auditors and tax officials deal with both European and national enquiries. Only Estonia managed to provide a break-down on (salaries) costs by a specific tool offered by Regulation (EU) 904/2010. Based on the costs estimations provided by Estonia it can be concluded that the biggest costs are associated with requests for information and administrative inquiries with second being multilateral controls and third Eurofisc.

More Member States have managed to quantify benefits of administrative cooperation. However most estimations cover multiple years and periods vary across Member States. Substantial differences in amounts also hint that Member States seem to be using different methodologies in estimating those benefits. Different sizes of Member States also make data submitted incomparable. Therefore for evaluation of proportionality of costs in relation to benefits of administrative cooperation the data reported in Figure 16 will be used.

Costs per tool offered in the framework of administrative cooperation and fight against fraud in VAT are considered to be proportionate to the benefits achieved to a very high or high extent by at least half of the Member States. Automated access to information through VIES scores particularly high with 18 Member States noting that cost-benefit ratio is proportionate to a very high extent. Eurofisc, simultaneous controls and request for information and administrative enquiries also score high. Other tools do not rank as high with feedback being the least proportionate tool when looking at the cost-benefit ratio.

4. Possible modifications to the regulatory framework for administrative cooperation and fight against fraud in the field of VAT

4.1. Possible modifications to currently existing tools for administrative cooperation

4.1.1. Standard forms

Most Member States agree that standard SCAC request form should be redesigned ( Figure 17 ). Some of the remarks that were made in relation to possible updates:

-allowing request based on more than one company, in order to deal with cases of suspected fraud networks in the requested Member State. (IE)

-identifiers other than VRN such as business name, contact details, website should be supported. (IE)

-forms should support exchange of import/export data, which are more and more often related to VAT MTIC fraud. (FI)

-mandatory use of one language in filling the forms should be implemented. (NL)

-forms should be made more user friendly e.g. by allowing corrections without having to fill the whole form from the start (NL), simplifying the form by showing only the field that are relevant to a particular request (IE).

A number of Member States (AT, CY, EL, LT, PT, SK and SI) suggested that a section specific to MOSS should be added. However one Member State (UK) thought that questions in relation to MOSS can be addressed using free text fields and mentioned that forms should be kept simple without excessive amount of pre-defined questions. Regarding technical issue, one Member State (EL) suggests to implement an automatic system to cover sensitive data for cases when forms are requested by the Prosecutor’s Office for instance. Member States have high expectations towards future central application (eFCA) and believe that it is going to be a more convenient tool for sending requests.

4.1.2. Automatic and spontaneous exchange of information

Member States are split on the question whether automatic and spontaneous exchange of information could be replaced by some other mechanism ( Figure 18 ). Those Member States that could support an alternative approach mention that granting direct access either to tax officials from other Member States or to Eurofisc liaison officials could reduce administrative burden for tax administrations and improve the quality of the information exchanged. One Member State (SE) suggests that this new tool could rely on an online platform – such as TNA for instance – providing direct information to designated officials both in requested and requesting countries and while other Member State (IE) would like to have an automated response tracking, statistics gathering and an analysis dashboard. In addition to information currently being exchanged without prior request, Member States mentioned legal persons' registration data as well as information relevant to control of VAT refunds as potentially interesting categories for exchange of information.

13 Member States disagree with the proposition, mainly because they believe that granting direct access to national database could increase the risk of missing relevant fraud-related information by impeding possibility of cross-checking received information with data held by national tax administration of the requesting Member State. Technical issues were also mentioned as a potential obstacle as national databases were not designed to allow external access. In addition official accessing information held in another Member State directly would have to understand the data and language of that Member State.

When it comes to categories for automatic exchange of information then most Member State do not support dropping already existing categories or have no opinion on the matter ( Figure 19 ). However Many Member State would support adding categories to the list of categories of information exchanged without prior request. In particular car registration data could be added (see Chapter 0) as well as data relevant for control of e-commerce held by Internet Platforms. Payment data held by financial institutions seems to be less relevant. When it comes to other data then the following was mentioned: data on whether the threshold for distance sales was exceeded, data held by Payment Service Providers, data on intra-EU acquisitions for those Member States that collect such data, legal persons' registration data and business activity indicators such as turnover amounts, employees and VAT payment information.

4.1.3. Feedback

Most Member States do not support mandatory replying to the feedback request ( Figure 20 ). To improve Member States ability to provide feedback on request for administrative cooperation some Member States would like to see a more automated-system with feedbacks sent via e-forms central application and automatic reminders transmitted until the feedback is submitted. One Member State (FI) asks for more guidelines and training for EOI functions and another Member State (UK) recommends that Member States should use Fiscalis working visits to learn on how other Member States implement feedbacks internally. One Member State (IE) insists that feedbacks are mandatory and replies are due within normal deadlines. Another Member State (NL) asks for introduction of deadline for providing feedbacks. Two Member States (BE and UK) believe that feedbacks are useful but can hardly be made mandatory (disproportionate workload for CLO, Member State often have nothing to pass on as they rarely collect relevant information from their own officers). One Member State (DK) specifies that feedback is manageable as far as it relies only on informing whether the information provided was useful. It was suggested that feedbacks must be provided if a Member State requires it, even if the reply is “we do not have a result yet” (UK) and that a mandatory motivation field in the request for feedback could be added in order to explain why this information is needed (NL).

4.1.4. VIES

When it comes to adding categories of information to be exchanged over VIES then it seems that most Member States would like to see data necessary for control of Customs Procedure 42 as well as data on car registration to be exchanged over VIES ( Figure 21 ). 14 Member States would also like to be able to check over VIES validity of so-called domestic VAT numbers 139 .

Majority of Member States believe that data in relation to Customs Procedure 42 could enhance the detection of risky imports. One Member States (BG) suggested to add an index box in VIES to indicate whether the transaction was preceded by Customs Procedure 42. Such solution would require minimum changes to VIES. According to the same Member State a more complete solution would see VAT identification number of the importer, client, value, commodity code to be made available over VIES. However another Member States (UK) noted that adding Customs Procedure 42 information to VIES would be very complex. It recommended having customs system checking the validity of VAT number at the time of importation and then having the customs system supply relevant information to Member States involved via automatic monthly supply of information.

If EUCARIS data to be shared through VIES one Member State (SE) would like information exchange to be clearly defined and closely controlled. In addition experts such as liaison officials from Eurofisc Working Field 2 should be consulted before. Another Member State (UK) believes that such information should not be shared through VIES.

One Member State (EE) welcomed the idea of sharing data on the identity, activity, legal form and address of taxable persons who have not been allocated an EU VAT identification number mentioning that such data could be very useful in order to identify the person when no EU VAT number is available. However another Member State (UK) noted that it could not be able to provide this information at it does not keep this kind of information in any database.

12 Member States would like to see other types of information to be exchanged over VIES. There are Member States that require traders to provide detailed reports on intra-EU acquisitions. Few Member States (BG, CR, RO, EE) suggest that such information should be shared through VIES in order to cross-check them with the information on intra-EU supplies. It was also suggested to share information on shareholders and managers (LV), NETP registration (CZ), information in relation to margin scheme (PT), NACE code should be added to Level 2 enquiries (UK), EU-MOSS registration number for third country company (NL) and invoice registers data (LT). One Member States (LT) noted that it would be valuable to see VAT registration number of a particular company in all Member State in order to verify correct tax compliance. One Member State (BG) suggests sharing information on the allocation of VAT identification numbers to taxable persons established in another Member State (Regulation 79/2012 (Art 3.1)). Another Member States (BE) also suggested to add information regarding the total amount of sales towards a Member State in the field of e-commerce. One Member States (DK) mentioned that if harmonized excise duty is involved, than it could be indicated in VIES (for instance a tick box). One Member State (IE) noted that all Member States should be required to show standardised data i.e. name, address, activity, registration number and date and also names of officers of the company or sole trade wherever available.

In view of one Member State (HR) VIES should be modernized before implementing any changes to the system. Otherwise, additional information must be shared through other channels. Another Member State (FI) notes that VIES might not be the best tool for exchange of information and that Eurofisc should be enhanced.

At least 13 Member States would like to see rules for automated access to VIES changed ( Figure 22 ). According to the comments a number of Member States (LV, SE, EL, PT, LT, HU, IE) would like to have restrictions mentioned in Art 21 para 2 p "e" sub points "i", "ii" and "iii" revised. Those sub points restrict access to so called "3rd Member State inquiry" – a right to consult VIES on intra-EU supplies where neither the supplier nor customer is registered in the Member State of the official making an enquiry. Currently such access is restricted to Eurofisc liaison officials who hold personal user identifications for the electronic systems allowing access to this information. Access must be in connection with an investigation into suspected fraud and is only granted during general working hours. In addition two Member States (SE and RO) mentioned that the requirement for Eurofisc liaison official to hold personal user identification for the electronic systems to gain access to "3rd Member State inquiry" should be clarified as presently Member States have different views on what is meant by this requirement allowing some Member State not to grant such access. Furthermore one Member State (FR) thinks that broader access to VIES data should be allowed, in particular for social security agency. Another Member State (NL) thinks that all competent authorities for VAT should have access to VIES.

As to other improvements to VIES Member States suggested the following:

-It seems that retroactive deregistration and registration create problems for Member States. One Member State (CZ) suggested that retrospective registration could be harmonised in the EU while another three Member States (PT, LT and UK) would like to prevent or forbid retroactive deregistration. Should there be changes to registration information then one Member States (CZ) would like to see what those changes were. Another Member State (UK) mentioned that all VAT Registered businesses should be automatically entered on VIES when they are registered for VAT.

-Accuracy of data should be ensured. In that context one Member State (EE) suggested that there should be regular monitoring of VIES and VAT declaration data for discrepancies. Another Member State (IE) mentioned that corrections and adjustments to VIES values made by the trader should always be reflected correctly in the data.

-According to one Member State (CZ) the rules for declaring supplies in recapitulative statement should be more unified. Some commodities e.g. investment gold are declared differently in some Member States and therefore there are discrepancies between VIES data and declared acquisitions of goods.

4.1.5. Eurofisc

Eurofisc is a network of liaison officials for swift and targeted exchange of information to facilitate multilateral cooperation in the fight against VAT fraud. The legal base for Eurofisc was introduced in 2010. According to the replies to the questionnaire after 7 years of Eurofisc functioning in its current format most Member States would like Eurofisc to evolve.

Most Member State support developing common risk analysis in Eurofisc ( Figure 23 ). Out of 3 Member States (CZ, UK and MT) that would not like to develop a common risk analysis within Eurofisc only one (CZ) provided a follow up comment explaining that already existing common risk indicators that are used in Eurofisc Working Field 1, 2 and 3 are sufficient.

Those Member States that support common risk analysis within Eurofisc mentioned that:

-Common risk analysis could significantly improve Eurofisc network as compared to current decentralized approach where each Member States develop their own risk criteria to select cases to be put under monitoring in Eurofisc. (LV)

-One Member State would like to see common analytical team on EU level. (PL)

-One Member State would like Eurofisc to more actively monitor possible threats, possibly in real time. (FI)

When commenting on common risk analysis within Eurofisc most Member States refer to TNA as a tool that can facilitate such an approach. Transaction Network Analysis – or TNA – is a tool currently being developed by the Commission to support information exchange within Eurofisc. It will use relevant VAT information (data on intra-EU supplies) and Eurofisc data to build, visualise and prioritize on the level of risk potentially fraudulent cross-border networks. According to the replies to the questionnaire most Member States are strongly in favour of TNA and support its' development ( Figure 24 ). Most Member States also agree that TNA should be managed by Eurofisc and are willing to participate in information exchange by using TNA. Most Member States also support TNA having access to data necessary to build networks (UK against). Majority of Member States also support creating an explicit legal base for TNA. The only country that does not support creating such legal base commented that according to their interpretation current rules are already sufficient to implement TNA and therefore no further action is necessary. TNA is supposed to be developed and hosted by the Commission with Eurofisc officials as users of the system. Under current legal arrangements the Commission cannot have access to Eurofisc operational data which, in case of the Commission hosting TNA, will require complicated organisational arrangements and possibly additional development costs. To simplify hosting arrangements the Commission could get access to Eurofisc data in so far as it is necessary for care, maintenance and development similarly to those that currently exist for CCN/CSI that is also facilitated by the Commission. Most Member States support granting the Commission such access. Out of two Member States (BE and UK) that do not support the Commission having such rights only one (UK) clarified that such access could be only in relation to the volumes of exchanges whilst taxpayer specific information should not be accessible.

In the replies to the open question in relation to TNA a number of Member States expressed their support to the project. Some Member States mentioned that current legal base in their view is sufficient for implementation of TNA. At least one (RO) Member State mentioned that in case the Commission will make a proposal for an explicit legal base, then the proposal should not be too inflexible not to make impossible future improvements to TNA.

In relation to other possible improvement to Eurofisc for it to better serve the needs of the Member States the following points were mentioned:

-Current arrangements for exchanging information via Excel spreadsheets uploaded to CircaBC are not optimal (LT, LU). Better IT solutions that would check the validity of data exchanged by Eurofisc officials (AT, LV) would improve the quality of data and the speediness of data exchanged eliminating the need for the working field coordinators to compile into one filed data uploaded by different Member States.

-The feedback should be provided quicker (UK, HR).

-Eurofisc reports should focus more on demonstrating the benefits of Eurofisc instead of describing the volumes of information exchange (UK).

-It should be clarified if the external organisation (e.g. Europol, Eurojust) or non-EU countries could participate in Eurofisc meetings where no sensitive information is exchanged. (SE)

-In addition sharing information with non-EU countries should be clarified when information is received through the administrative cooperation (SE) where traders in that non-EU country are involved in fraud.

-Customs could participate in Working Field on e-commerce to monitor distance sales from non-EU countries. (FR)

-Working Field 1 could use data on intra-EU supplies that follow importation to better detect MTIC fraud with the interposition of non-EU transactions (FR) and Eurofisc in general could also exchange information on imports and exports (PT).

4.2. Collaboration with law enforcement authorities

The VAT action plan, the European Parliament, the European Court of auditors and the Council stress the need for deeper cooperation between different authorities to prevent and fight VAT fraud. In particular customs and tax administrations lack access to each other's data. Customs has often no access to VIES data when carrying out checks and tax administrations do not have systematic access to information on imports using the customs 42 procedure. Fighting organised crime networks engaged in carrousel fraud requires joint efforts between tax administrations, law enforcement bodies in the Member States and at EU level.

Member States have different experiences regarding exchange of VAT-related information with other authorities. A large part of them however do not exchange information at all with OLAF, Europol, and Social Security Agencies, unless for cases falling within their abilities. Such information can be shared sometimes with other financial and criminal national authorities but only if they have expertise in the relevant field.

From the comments to the open question the following replies stand out:

-One Member State (ES) has a special regulation for this purpose, few problems are encountered. Another Member State (LU) has no legal basis for cooperation with these entities.

-Another Member State (DK) notes that all information can be shared with foreign police authority only through the justice department.

-Another Member State (FI) notes that VAT-related cases often might come up often with other non-tax issues that require permission from the competent authority of the other Member State. As definition of crime differs within the EU, permission could be refused sometimes. It would be advisable thus to allow the use of the information of some of the most common purposes without special permission.

According to the replies tax administration in most Member States do not exchange VAT related information with Europol. Neither do most Member States provide Europol access to data exchanged over VIES not think that Europol should have access to VIES citing the lack of legal basis and tax secrecy. Member States that are encountering problems with information exchange with Europol to a very high or high degree (BE, AT, ES, DK and LU) further explained that information exchange happens through the police or justice department meaning that also in those cases there is no direct information exchange between tax administration and Europol. Replies on cooperation with OLAF largely follow the same pattern. Most Member States do not exchange information with OLAF and think that OLAF should not have access to VIES.

12 Member States agree, 4 disagree and 11 have no opinion on the question whether Europol should receive targeted intelligence from Eurofisc ( Figure 27 ). When it comes to providing OLAF with intelligence from Eurofisc then 8 Member States would agree with such arrangement, 7 are against and 12 have no opinion. In case such intelligence is shared with Europol and OLAF 14 Member States agree that those entities should provide Eurofisc feedback on the use of this information. However it seems that tax administrations are largely unaware of what information Europol and OLAF have and could not specify what sort of feedback they would expect. One Member State (CZ) mentioned that such exchange could be useful but the absence of legal base is an obstacle. Should there be some form of exchange of information between OLAF and Europol on the one hand and Eurofisc on the other then Member State would like to see a step by step approach with information exchange happening on a case by case basis.

12 Member States agree that it would be useful if Europol participate in Eurofisc Working Field meetings with 4 Member States against such participating and 8 having no opinion. 10 Member States would agree with OLAF participation in Eurofisc meeting, 8 are against and 9 have no opinion on the matter. Member States further clarified that such cooperation should be limited to OLAF's and Europol's participation in Working Field 4 meetings and not operational Working Fields. One country saw a problem in such participation as in their view there cannot be a direct cooperation between Europol and Eurofisc as Eurofisc officials are from tax administrations.

According to the replies most Member States exchange information with Customs. In most cases Member States encounter problems either to a very low or to a low degree ( Figure 25 ) when exchanging information with customs but Member States did not specify the nature of those problems. Member States note that as often their tax and customs administrations are a single authority, VAT-related information is available to customs. Customs administration in most Member States also have access to data exchanged over VIES ( Figure 26 ). Tax administrations think that information on imports where Customs procedure 42 is used should be made available to them by customs authorities. Most Member States would like to have automated access to such data.

4.3. Administrative cooperation with 3rd countries

There is evidence that fraudsters are utilizing non-EU countries to enable VAT frauds. The ECA's report states that cooperation with non-EU countries should be strengthened. The Commission is currently negotiating an agreement on administrative cooperation in the field of VAT with Norway. To understand further needs the Commission asked Member States to list countries where similar agreements could be beneficial. According to the replies top five countries are Switzerland, China, US, Russia and Norway.

4.4. EUCARIS

The potential use of the EUCARIS system has been discussed several times during the meetings of the Standing Committee on Administrative Cooperation Expert Group (SCAC-EG). In this context it was agreed to make an in-depth study on the technicalities and related costs of some options to access EUCARIS data on cars. This should be considered as a new form of automated access to specific information stored in national databases.

All Member States consider that access to the EUCARIS database on cars for tax authorities would be useful to control the VAT treatment of vehicles. Member States mentioned that having access to such data would help to detect of tax frauds with vehicles, in particular the abuse of margin scheme.

All Member States but one (RO) agree that having access to such data would reduce the administrative burden of tax administrations in exchanging information without prior request and simplify the work of Eurofisc Working Field 2. Member States further explained that direct access to such data should increase the speediness of the detection of fraud and could decrease the amount of requests for administrative cooperation.

Most Member States would like to see an automated access to EUCARIS data. A number of Member States (BG, CZ, FR) mentioned the VISTA application as the preferred solution – a central application that can be developed by the Commission.

A number Member States (BG, CZ, SI, PT, AT, RO, FI, UK, IE, NL) would also like access to such application not to be limited only to Eurofisc officials. Four Member States (BE, IT, MT, LU) however would like access to be limited to Eurofisc officials.

In terms of necessary data then those can be split in the following categories:

-Data that allows to identify the vehicle: VIN, license plate number;

-Data that allows to identify the owner: name (surname), driving license number and for the companies VAT number, Member State of registration;

-Registration history: information on former owners with dates of registration and deregistration.

4.5. Joint audits

The joint audit is an instrument through which a taxable person is subject to a coordinated audit by a single audit team made up of tax officials of two or more jurisdictions. Regulation (EU) 904/2010 does not provide for joint audits but for only the passive presence of officials from another Member State during an administrative enquiry in the taxpayer premises.

According to the replies only 4 Member States have experience with joint audits. Out of those 4 Member States 2 specified that this experience relates to direct taxes. According to the replies none of the Member State have experience with joint audits in relation to VAT. Member States are largely split as to whether there where cases in their work where joint audits would have been a more useful tool as compared to the tools currently offered by the Regulation (EU) 904/2010 ( Figure 30 ).

When it comes to possible benefits of joints audits around half of the Member States either strongly agree or agree that joint audit would be a useful and cost effective tool to audit cross-border trader as well as that an audit by a single team of auditors would provide more legal certainty and would be more cost efficient for the businesses as to being audited multiple times by different countries ( Figure 31 ). A number of Member States had no opinion on this question.

Those Member States that are sceptical towards joints audits further explained that:

-The absence of a national and European legal base for joint audits is a concern as well as differences in approach towards audits in Member States might hamper setting up of joint audits (BE).

-One Member State did not see a difference between asking a Member State of establishment to collecting the information at the request of another Member States and sending a team of auditors to the Member State of establishment. In view of that Member State the cost-benefit of joint audits would not be any different from currently existing simultaneous controls and Member States already have an option to be passively present in other Member States during audits activities. In addition legal certainty for companies is not evident as it is not clear on how to proceed in case of appeals to tax assessments. (AT)

Some Member States are not completely sure of the extent to which the report produced within joint audits is binding (FI, EE, UK). If it is not binding then it does not increase the legal certainty for companies (FI). There must be clear rules of procedures for acting during joint audits according to which officials can act in other Member States (SI). Participating in joint audits should be on voluntary basis (FI) and auditors should opt for this solution only when currently existing tools are not sufficient (EE). One Member State mentioned there should be a clear legal base for joint audits (EE). When talking about creating a legal base at least one Member State (FI) would like to see a legal base for joint audits both in VAT and in the field of direct taxation as not aligned legal base for different types of taxes could impede the use of the tool. Few Member States commented that the tool is likely to be useful (UK) in particular for the e-commerce sector (EE). One Member State (UK) would like to see a gradual approach towards joint audits remarking that there will be a list of issues to solve, in particular questions in relation to national competences: use of information powers, powers to assess outside of the Member State of an auditor, permissions for a joint audit team to operate and whether the presence of the home auditor is necessary. One Member State (IT) expressed a clear support to introduction of joint audits into the list of tools for administrative cooperation in the field of VAT.

9.5. Annex 5: Consultation of stakeholders other than Member States

1. Background

In March 2017, the Commission (Directorate General Taxation and Customs Union) launched an open public consultation on the functioning of the administrative cooperation and fight against fraud in the field of Value Added Tax (VAT).

This administrative cooperation in the field of VAT is governed by Council Regulation (EU) No 904/2010 of 7 October 2010.

The purpose of the consultation was:

·to gather views from stakeholders about their experience of the current rules governing administrative cooperation and fight against cross-border fraud in the field of VAT;

·to bring new insights for the on-going evaluation of Regulation (EU) 904/2010;

·to provide information about possible improvements including ‘VIES on-the-web’; and

·to collect data on possible reduction or increase of regulatory costs/benefits (administrative burden and/or compliance costs) for businesses (in particular SMEs) although in practice, such information was actually not provided in the answers received.

This report gives an overview of the responses to the public consultation.

2. Respondents

The Commission received 58 individual replies through the on-line survey tool which was available until 31 May 2017. Though the public consultation was announced in several fora, publicly announced on Commission websites including ‘VIES on the web’ page and being made available in all EU official languages bar the Gaelic, there was only a limited number of responses to this public consultation (58). Respondents were mainly professionals responding on behalf of their organisation (74%). These professionals generally worked for a private enterprise (75%) and were established in 12 different Member States.

It is important to note that since only 58 replies from 12 Member States were received over the course of the open-public consultation, these responses are not statistically representative of the target population. Answer ratios must therefore be interpreted with care without any possibility to draw any general conclusions from these replies.

3. Replies

3.1. General remarks

49 respondents agreed or strongly agreed that there should be EU common rules for cooperation between Member States in the field of VAT. 54 respondents considered that there is a role for the EU to assist Member States in their cooperation in the field of VAT and 50

considered that the role of the EU should go beyond support and assistance and should make sure that the cooperation takes place to the largest extent possible.

27 respondents considered that the current instruments against VAT fraud are not effective. As regards the fight against VAT fraud perpetrated by criminal organisations, 29 respondents considered that the current instruments are not effective.

28 respondents considered that the current instruments are not adapted to the new business models such as e-commerce or the collaborative economy. At the same time 14 respondents did not have an opinion.

23 respondents considered that the costs for cooperation can be justified by additional VAT revenues against 9 who disagreed. At the same time 15 respondents did not have an opinion.

13 respondents considered that manual exchange should remain the main way of cooperation in the field of VAT against 22 who disagreed. 47 respondents were in favour of sharing more data through VIES when only 4 disagreed.

In particular, 29 respondents considered that tax authorities should have automated access to data on cars registration from other Member States. 41 respondents considered that the automated access should cover also exempted importation.

36 respondents were in favour of sharing VAT data with other national authorities.

28 respondents considered that a more intense exchange of personal data would be justified by the fight against VAT fraud. At the same time, 45 respondents considered that this exchange should be strictly proportionate to the objectives.

Open text comments

Additionally, respondents frequently emphasised that the exchange of information should be highly-secured, framed and used only for the purpose of fighting VAT fraud. Apart from this objective, the exchange of information would be irrelevant and unwanted. On possible improvements and extensions of the current system, respondents insisted that EU should concentrate on making existing systems efficient and performant before thinking of creating new entities. Respondents also frequently noted that all improvements should be designed with a great carefulness taking into consideration their impact on businesses.

Respondents widely approve the idea that VAT fraud is an international issue that must be dealt jointly. Administrative cooperation is thus valuable in their opinion.

Although automated exchange of information is very useful as it helps reducing delays in processing VAT information, a majority of respondent believes that some improvements should be made. A large part of respondent asks for more harmonization between Member States of the information collected and provided. Respondents also report a lack of information available in VIES on-the-web and suggest that at least name and address of businesses could be provided. Some of them also suggest making local tax numbers together with EU VAT numbers in other Member States if applicable and their validity period available in VIES and VIES on-the-web.

On services proposed by VIES on-the-web, respondents suggest to add a storage option of previous VAT number checks and a report functionality in case of invalid VAT check – in order to report easily potential fraudsters.

One respondent suggest to link VIES to other existing channels such as registrations in professional order or at Chamber of commerce in order to provide more complete and accurate information.

Nonetheless, respondent recall that exchange of information should be closely controlled, in order to guarantee data protection and human right of privacy. Two respondents also observed that although the exchange of information in the field of the fight against VAT fraud is valuable, other data exchanges could be irrelevant.

Some respondents also pointed out that EU should be working on improving the current system before considering creating a new database or increasing the number of requests sent - and by doing so should be careful not to increase burdens for businesses.

3.2. Joint audits

21 respondents agreed that auditors from other Member States should be involved only in a passive way during an audit. At the same time, 12 disagreed and 22 did not have an opinion.

19 respondents agreed and 17 disagreed that an active participation of foreign auditors would make it more efficient. 19 respondents could not decide.

19 respondents considered that the international audits for VAT purposes are burdensome.

In comparison, 21 respondents consider that a joint audit would be less burdensome.

40 respondents considered that having a single report at the end of the audit would provide taxpayers with more legal certainty.

Open text comments

Respondents widely agreed that joint audits would constitute a step forward in administrative cooperation regarding fight against VAT fraud and would prevent duplication of tax audit which is burdensome for both tax administration and businesses.

However, some respondents observed that joint audits could be useful as long as a common unique procedure is clearly defined and relies on a legal basis. Joint audits outcomes should be directly shared with other Member States as well.

It was emphasized that as the expertise of tax officials matters more than their nationality, joint audits could be valuable if they are led by highly-qualified auditors knowing all Member States’ tax regulation.

Respondents also observed that language barriers are an issue to be considered.

Few respondents disapprove the idea as they believed that cooperation should rely on mutual trust. Thus, national audits should be sufficient.

Two respondents insisted that joint audits could be very useful in the field of transfer pricing, by preventing double taxation and legal disputes.

3.3. Eurofisc

47 respondents considered that an automated joint risk analysis would help in the fight against VAT fraud.

36 respondents considered that Eurofisc is the right structure to coordinate and carry out the joint risk analysis against VAT fraud.

43 respondents saw a role for Eurofisc to coordinate cross-border administrative enquiries.

32 respondents considered that the current arrangements for cooperation between tax authorities and law enforcement bodies are not effective. Only 9 considered them as being effective.

34 respondents considered that the cooperation between tax authorities and law enforcement bodies should be changed.

42 respondents considered that Eurofisc should share relevant information on serious VAT fraud with OLAF.

45 respondents considered that Eurofisc should share relevant information on serious VAT fraud with Europol.

In this context, 24 respondents considered that the fight against VAT fraud would justify more exchange of personal data between tax authorities and law enforcement bodies.

45 respondents considered that these exchanges should be proportionate to the objectives.