Implementation of the EU requirements for tax information exchange

STUDY

EPRS | European Parliamentary Research Service

Editor: Eckhard Binder

Ex-Post Evaluation Unit

PE 662.603 – February 2021

Implementation

of the EU

requirements

for tax

information

exchange

European

Implementation

Assessment

EPRS | European Parliamentary Research Service

Implementation of the

EU requirements for

tax information

exchange

European implementation assessment

In February 2020, the European Parliament's Committee on Economic and

Monetary Affairs (ECON) requested to draw up an implementation report

on the implementation of the EU requirements for tax information

exchange: progress, lessons learnt and obstacles to overcome.

Sven Giegold (Greens/EFA, Germany) has been appointed rapporteur.

This European implementation assessment (EIA) has been prepared by the

Ex-Post Evaluation Unit (EVAL) within the European Parliamentary Research

Service (EPRS) to accompany the scrutiny work of the Committee on

Economic and Monetary Affairs.

AUTHORS

1. The in-house opening analysis was written by Eckhard Binder from the Ex-Post Evaluation Unit, EPRS.

2. The research paper on the implementation of the EU requirements for tax information exchange: progress,

lessons learnt and obstacles to overcome, was written by Patrice Muller, Nicholas Robin and James Forrester

of LE Europe Limited and Amal Larhlid, Vanessa Vazquez-Felpeto, Vasilis Douzenis and Khush Patel of

PricewaterhouseCoopers LLP.

The research paper was written at the request of the Ex-Post Evaluation Unit of the Directorate for Impact

Assessment and European Added Value, within the Directorate-General for Parliamentary Research Services

(EPRS) of the Secretariat of the European Parliament.

To contact the authors, please email: EPRS-ExPostEvaluation@ep.europa.eu

ADMINISTRATOR RESPONSIBLE

Eckhard Binder, Ex-Post Evaluation Unit, EPRS.

To contact the publisher, please e-m

ail EPRS-ExPostEvaluation@ep.europa.eu

LINGUISTIC VERSIONS

Original: EN

Manuscript completed in February 2021.

DISCLAIMER AND COPYRIGHT

This document is prepared for, and addressed to, the Members and staff of the European Parliament as

background material to assist them in their parliamentary work. The content of the document is the sole

responsibility of its author(s) and any opinions expressed herein should not be taken to represent an official

position of the Parliament.

Reproduction and translation for non-commercial purposes are authorised, provided the source is

acknowledged and the European Parliament is given prior notice and sent a copy.

Brussels © European Union, 2021.

PE 662.603

ISBN: 978-92-846-7700-9

DOI: 10.2861/433189

CAT: QA-03-21-028-EN-N

eprs@ep.europa.eu

http://www.eprs.ep.parl.union.eu (intranet)

http://www.europarl.europa.eu/thinktank (internet)

http://epthinktank.eu (blog)

Implementation of the EU requirements for tax information exchange

Table of frequently used abbreviations and acronyms

AEOI

automatic exchange of information

AML

anti-money-laundering

APA

advance pricing agreement

ATR

advance tax ruling

BEPS

base erosion and profit shifting

beneficial ownership

CbCR

country

-by-country report

CCN

common communication network

Commission

European Commission

CRS

common reporting standard

DAC

Directive on Administrative Cooperation

director's fees

directorate

-general

EESC

European Economic and Social Committee

European Commission

ECA

European Court of Auditors

income from employment

EIA

uropean implementation assessment

EOIR

exchange of information on request

European Parliament

EPRS

European Parliamentary Research Service

European Union

EUR

euro

FATCA

Foreign Account Tax Compliance Act

immovable property

LIP

life insurance products

MNE

multinational enterprise

OCTs

overseas countries and territories

OECD

Organisation for Economic Co-operation and Development

pensions

STD

Savings Tax Directive

TIN

taxpayer identification number

United States

EPRS | European Parliamentary Research Service

Table of contents

Table of frequently used abbreviations and acronyms ________________________________ I

PART I. IN-HOUSE INTRODU CTORY AN ALYS IS______________________________________ 1

Executive summary ____________________________________________________________ 1

1. Introduction ________________________________________________________________ 3

1.1. Context of Directive 2011/16/EU ______________________________________________ 3

1.1.1. Amendments to the DAC since 2011____________________________________________ 5

1.1.2. Implementing regulations ___________________________________________________ 7

1.2. Context of the study (DAC1-4) ________________________________________________ 8

1.2.1. Commission evaluation and reports ____________________________________________ 8

1.2.2. European Parliament committees on tax issues ____________________________________ 9

1.2.3. Council of the European Union_______________________________________________ 10

1.2.4. European Court of Auditors' special report ______________________________________ 11

1.2.5. European Economic and Social Committee opinion ________________________________ 11

1.3. Scope and methodology of the study _________________________________________ 11

2. Scope of the DAC and its amendments_________________________________________14

2.1. Forms of income and capital covered by the DAC1-2 _____________________________ 14

2.2. Cross-border tax rulings and advanced pricing arrangements (DAC3) ________________ 14

2.3. Country-by-country reports (DAC4) ___________________________________________ 15

3. Transposition of the DAC1-6 _________________________________________________16

3.1. Transposition by Member States _____________________________________________ 16

3.2. Past and present infringement procedures _____________________________________ 16

4. Data analysis ______________________________________________________________18

4.1. Data availability and limitations ______________________________________________ 18

4.2. National data sources ______________________________________________________ 19

Implementation of the EU requirements for tax information exchange

III

5. Effectiveness of the DAC ____________________________________________________21

5.1. Market reactions to the DAC ________________________________________________ 21

5.1.1. Income categories ________________________________________________________ 21

5.1.2. Circumventions and loopholes _______________________________________________ 22

5.2. Obstacles to the exchange of information______________________________________ 23

5.2.1. Legal obstacle s __________________________________________________________ 23

5.2.2. Practical obstacles ________________________________________________________ 23

6. External coherence of the DAC _______________________________________________25

6.1. Comparison between the DAC2/CRS and the FATCA _____________________________ 25

6.1.1. Global Forum cooperation on the AEOI _________________________________________ 26

6.1.2. Global Forum cooperation on the EOIR _________________________________________ 26

6.2. The DAC and the Savings Tax Directive ________________________________________ 28

6.3. The use of DAC information for other purposes _________________________________ 28

6.3.1. Anti-money-laundering provisions ____________________________________________ 28

6.3.2. Judicial assistance provisions ________________________________________________ 29

7. Efficiency, EU added value and relevance ______________________________________30

7.1. Efficiency ________________________________________________________________ 30

7.2. EU added value ___________________________________________________________ 31

7.3. Relevance _______________________________________________________________ 31

8. Conclusions _______________________________________________________________33

8.1. Main findings on effectiveness _______________________________________________ 33

8.2. Main findings on coherence _________________________________________________ 34

8.3. Recommendations ________________________________________________________ 34

MAIN REFERENCES ___________________________________________________________35

Part II: Implementation of EU requirements for tax information exchange: Progress, lessons

learnt and obstacles to overcome _______________________________________________39

EPRS | European Parliamentary Research Service

Table of figures

Figure 1: Number of countries to which AEOI information was sent, 2017-2019 ____________ 26

Figure 2: Global Forum EOIR – compliance ratings of EU Member States __________________ 27

Table of tables

Table 1: Directive on administrative cooperation (DAC) ________________________________ 7

Table 2: EP resolution of 26 March 2019 and actions taken by the Commission ______________ 9

Table 3: Transposition of the DAC (state of play as of January 2021) ______________________ 16

Table 4: Key findings on the EU added value of the DAC _______________________________ 31

Implementation of the EU requirements for tax information exchange

PART I. IN-HOUSE INTRODUCTORY ANALYSIS

Executive summary

The first part of this European implementation assessment (EIA) starts with an overview of the

context of the Directive on administrative cooperation in the field of taxation (DAC), explaining the

background and history of the legislative framework as well as the subsequent amendments that

have been made to the directive since its adoption in 2011. It also provides a summary of the

research results that are presented in the annexed research paper in the second part of the EIA.

Since the academic literature on the DAC is very limited, a starting point for this EIA was the

evaluation of the directive by the European Commission in 2019. While this evaluation did not

identify major deficiencies in the implementation of the DAC, the European Commission however

recognised that there was very little evidence to prove the effectiveness and efficiency of the

directive. This EIA therefore attempts to provide additional evidence on the implementation of the

directive, with a special focus on its effectiveness and external coherence. Due to the relatively

recent entry into force of the latest two amendments (DAC5-6), this study mainly analyses the initial

directive (DAC or DAC1) and the first three amendments (DAC2-4).

The opinions of other EU institutions and bodies are referred to in separate chapters, highlighting

the needs for reforms as regards administrative cooperation that each of them has identified. In the

case of the European Parliament, its position is exemplified using the report adopted in Ma rch 2019

by the Special Committee on financial crimes, tax evasion and tax avoidance (TAX3). The Council of

the European Union adopted its conclusions with respect to the DAC in June 2020, mentioning

potential areas of improvement of the DAC and its interoperability and convergence with other EU

legislation. In a similar way, the European Economic and Social Committee adopted an own-

initiative opinion in September 2020, asking for further improvements in administrative

cooperation in the field of taxation.

National transposition of the initial directive and the first amendments proved to be quite complex

and led to a relatively high number of cases of late transposition. The first part of the EIA shows

however that most of the infringement procedures have been closed in the meantime, with only

one infringement case currently still open for the DAC4.

Both parts of the EIA provide insights into the achievement of the directive's objectives. They show

that the directive has to some extent proven its usefulness in fighting cross-border tax fraud, evasion

and avoidance, and has a deterrent effect. However, both parts of this EIA point several times to the

fact that there is limited evidence, identifying it as the main obstacle to assessing the directive's

effectiveness. This EIA complements the publicly available evidence with new qualitative and

quantitative information mainly stemming from interviews with tax administrations and a

stakeholder survey.

For the purposes of assessing the directive's effectiveness, the first part of this EIA offers a n o verview

of the different areas that are analysed in detail in the research paper annexed in the second part.

The chapter on market reactions shows how the scope of the directive has changed over time in

order to cover an increasing number of income categories and to close existing loopholes and

prevent circumvention. Another aspect that affects the directive's effectiveness are the legal and

practical obstacles hampering the exchange of tax information. The assessment of these obstacles

has served as grounds for the suggestions made in the annexed research paper on possible

improvements in this area.

EPRS | European Parliamentary Research Service

The assessment of the directive's external coherence has been laid out into two subchapters, the

first one offering a comparison of the different frameworks for tax information exchanges, i.e. the

DAC, the Global Forum on Transparency and Exchange of Information for Tax Purposes (Global

Forum) as well as the bilateral exchanges with the United States. To illustrate how EU Member States

perform with regard to the international exchange of tax information, this subchapter also contains

their ratings within the Global Forum framework. The second subchapter presents the interaction

and convergence between the DAC and other EU legislation in the fields of anti-money-laundering

and judicial assistance.

Despite the fact that there is broad agreement among the different institutions and stakeholders on

the usefulness of administrative cooperation in the field of taxation, there is still scope for

improvement. The analysis of the directive's effectiveness and coherence in this study and the

annexed research paper has helped highlight a number of areas where such improvements could

be made through future amendments. These include increased availability of taxpayer identification

numbers, enhanced due diligence and reporting requirements, and a broader DAC coverage.

The annexed research paper analyses the implementation of the directive based on qualitative desk

research and a literature review, targeted interviews with tax authorities, and a stakeholder survey.

The analysis of quantitative data on tax information exchanges is limited to publicly available

information in the European Commission evaluation and on quantitative information collected from

some Member States. The first part of the study also contains detailed recommendations for further

improvements to the directive's effectiveness and coherence.

Implementation of the EU requirements for tax information exchange

1. Introduction

1.1. Context of Directive 2011/16/EU

The first steps towards cooperation in the exchange of tax information in the field of direct taxation

within the EU were taken in the 1970s. In 1977, the Council adopted the Directive concerning mutual

assistance by the competent authorities of the Member States in the field of direct taxation and

taxation of insurance premiums.

It took another 20 years before tax information exchange became

an important topic in the international tax policy discussions.

In 2003, the Council adopted the EU

Savings Tax Directive. Once it entered into force in 2005,

this directive introduced provisions on the

first mandatory exchange among EU Member States of information of interest payments to

residents of other Member States and 10 dependent and associated territories of the Member

States.

In addition, the Council also concluded agreements on savings taxation with five non-EU

European countries (Andorra, Liechtenstein, Monaco, San Marino and Switzerland).

In 2011, the Council adopted the Directive on administrative cooperation in the field of taxation

(DAC1),

which repealed and replaced the Mutual Assistance Directive of 1977. The DAC1 entered

into force in January 2013 and laid down the rules and procedures for cooperation between EU

Member States on the exchange of foreseeably relevant information

between their national tax

administrations. Besides seeking to enable the Member States to better combat tax evasion, tax

fraud and avoidance linked to aggressive tax planning, the general objectives of the directive

include i) ensuring efficient administrative cooperation between the Member States to help them

overcome the negative effects of the increasing globalisation on the internal market; ii) making tax

systems fairer; and iii) safeguarding Member States' tax revenues.

Going beyond the requirements of the Mutual Assistance Directive of 1977, the DAC1 intr oduced

the requirement for cooperation between tax authorities through different forms of information

exchange and other forms of cooperation, such as simultaneous controls and presence of tax

officers of a given Member State in the administrative offices of another Member State. The DAC1

also introduced a new type of mandatory automatic exchange of information (AEOI) on five specific

categories of income and capital, for which information is available (income from employment,

director's fees, pensions, life insurance products and immovable property).

Council Directive 77/799/EEC of 19 December 1977 concerning mutual assistance by the competent authorities of the

Member States in the field of direct taxation.

M. Keen and J.E. Ligthart, Information Sharing and International Taxation: A Primer, International Tax and Public Finance,

13, pp. 81–110, 2006.

Council Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments

These 10 territories were three United Kingdom-dependent territories (Guernsey, Jersey and the Isle of Man), the five

British Caribbean territories (Anguilla, the Cayman Islands, Montserrat, the Turks and Caicos Islands and the British

Virgin Islands) and the two Dutch Caribbean territories (the Netherlands Antilles and Aruba).

Council Directive 2011/16/EU of 15 February 2011 on administrative cooperation in the field of taxation and repealing

Directive 77/799/EEC.

The importance of the notion of 'foreseeable relevance' in the context of tax information exchange was stressed by the

OECD already in 2010. See: OECD,

Implementing the Tax Transparency Standards: A Handbook for Assessors and

Jurisdictions, OECD Publishing, Paris, 2010. Although the term 'foreseeable relevance' has been used in the DAC

provisions since the adoption of DAC1, the European Commission used its proposal for an amendment made in July

2020 (DAC7) to give a more precise definition of the term and the necessary supporting information that a requesting

competent authority has to provide to demonstrate the foreseeable relevance of the requested information.

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

EPRS | European Parliamentary Research Service

The changes made in the field of tax information exchange between EU Member States over the

past 10 years have also been driven by the call for more transparency and fairness of the taxation

systems within the single market. Starting with the exchange of available information and

information exchange on request, the scope of the directive has evolved over the years to include a

growing number of categories of mandatory automatic exchange. The fight against tax evasion, tax

fraud and tax avoidance has been the guiding principle, reinforced by events such as the LuxLeaks

scandal

in 2014, which triggered the second amendment of the DAC.

Enhanced transparency and fair competition between taxation systems can have an important

economic impact on EU Member States. In 2015, up to 40 % of global profits of multinational

companies – i.e. profits made by multinational companies outside of the country where their parent

is located – were shifted to tax havens.

In this context, 'the governments of the (non-haven)

European Union countries appear to be the prime losers of this shifting'.

Another field closely linked to tax transparency is the fight against money laundering, which has

become more prominent on the political agenda in recent years. This link became visible in the

fourth amendment of the DAC, which gives tax administrations access to beneficial ownership

information as collected under the fifth Anti-money-laundering Directive.

The link between tax

information exchange and the fight against money laundering is also underlined in the European

Commission's Action Plan for a comprehensive Union policy on preventing money laundering and

terrorist financing, adopted in May 2020.

In recent years, the European Parliament has been working extensively on the issues of tax evasion,

tax fraud and tax avoidance. Building on the work done by three special committees and a

committee of inquiry, the Parliament decided in June 2020 to set up a Subcommittee on Tax Matters

(FISC). The DAC provisions deal with the exchange of direct taxation data, i.e. in a field where the EU

has only limited competence. For that reason it was the Council, after consulting with the

Parliament, that adopted the DAC1 and its successive amendments proposed by the Commission.

The roles and positions of the Council and the Parliament are described in greater detail in the

following chapters.

Finally, the developments at the level of the European Union need to be seen in a broader,

international context. Under the patronage of the OECD, the Global Forum on Transparency and

Exchange of Information for Tax Purposes (the Global Forum) has been working on the

implementation of international tax transparency and information exchange standards since 2009.

In November 2014, the International Consortium of Investigative Journalists (ICIJ) disclosed information on a tax ruling

between Luxembourg and more than 340 multinational companies. Through the shifting of profits to Luxembourg,

the companies had been able to reduce their tax payments. For further details, please see:

https://www.icij.org/investigations/luxembourg-leaks/about-project-luxembourg-leaks/

T. Torslov, L. Wier, G. Zucman, The Missing Profits of Nations, World Economy Database, Working Paper N

2020/12, April

2020.

ibid.

Directive 2018/843 of the European Parliament and of the Council of 30 May 2018 amending Directive (EU) 2015/849 on

the prevention of the use of the financial system for the purposes of money laundering or terrorist financing, and

amending Directives 2009/138/EC and 2013/36/EU.

Communication from the Commission C(2020) 2800 final of 7 May 2020 on an Action Plan for a comprehensive Union

policy on preventing money laundering and terrorist financing.

Implementation of the EU requirements for tax information exchange

1.1.1. Amendments to the DAC since 2011

Since its adoption in 2011, the directive has been amended several times. As part of a tax

transparency package, the first amendment of the directive (DAC2)

entered into application in

January 2016 and extended the scope of the mandatory automatic exchange of information to

information on financial accounts. Contrary to the DAC1, the exchange of this information is

mandatory because Member States are already making that information available to the United

States (US) under their bilateral agreements on the Foreign Accounts Tax Compliance Act (FATCA)

with this country. This requirement was included in the amendment to avoid the need of invoking

the most favoured nation clause built into the DAC.

The DAC2 also introduced the EU framework

for the OECD Common Reporting Standard (CRS) and repealed the Savings Tax Directive

that had

required the automatic exchange of information on private savings income between Member States

since 2005.

The LuxLeaks scandal that broke out at the end of 2014 encouraged the European Commission to

propose a further step towards enhanced tax transparency. As a consequence, the second

amendment of the DAC (DAC3),

which entered into application in January 2017, introduced the

automatic exchange of information on advance cross-border rulings and advance price

arrangements.

The third amendment (DAC4),

which was part of an anti-tax avoidance package presented by the

European Commission in 2016, entered into application in June 2017 and introduced the automatic

exchange of information on country-by-country reports (CbCR) for multinational enterprise (MNE)

groups. This amendment sought to combat aggressive tax planning activities by MNE groups by

tackling base erosion and profit-shifting (BEPS) practices.

The fourth amendment (DAC5),

providing access by tax authorities to beneficial ownership

information as collected under anti-money-laundering (AML) rules, entered into application in

January 2018. This provision is thought to help tax authorities to effectively monitor how financial

institutions are applying the due diligence procedures set out in the DAC1, and to combat tax

evasion and tax fraud.

The fifth amendment (DAC6),

which became applicable in July 2020, introduced the requirement

for the automatic exchange of information on tax planning cross-border arrangements to create an

environment of fair taxation in the internal market.

In June 2020, in response to the impact of the Covid-19 pandemic, the Council proposed to

temporarily defer by six months (plus a possible extension by another three months) cer tain time

Council Directive 2014/107/EU of 9 December 2014 amending Directive 2011/16/EU as regards mandatory automatic

exchange of information in the field of taxation.

Under this most favoured nation clause (Article 19 of DAC), Member States are bound to provide any EU partner that

requests it with the same level of information as they provide to third countries, if this is more than provided for under

EU law.

Council Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments.

Council Directive (EU) 2015/2376 of 8 December 2015 amending Directive 2011/16/EU as regards mandatory automatic

exchange of information in the field of taxation.

Council Directive (EU) 2016/881 of 25 May 2016 amending Directive 2011/16/EU as regards mandatory automatic

exchange of information in the field of taxation.

Council Directive (EU) 2016/2258 of 6 December 2016 amending Directive 2011/16/EU as regards access to anti-money-

laundering information by tax authorities.

Council Directive (EU) 2018/822 of 25 May 2018 amending Directive 2011/16/EU as regards mandatory automatic

exchange of information in the field of taxation in relation to reportable cross-border arrangements.

EPRS | European Parliamentary Research Service

limits for the filing and exchange of information on financial accounts, (DAC2), for the disclosure

rules for intermediaries and for tax planning cross-border arrangements (both DAC6). Nevertheless,

again in June 2020, the European Parliament adopted its position to extend these deadlines by only

three months, without the option of a further extension. On 24 June 2020, the Council adopted an

amendment of the DAC,

granting Member States the option to extend the time limit for cross-

border arrangements by six months and by three months for the automatic exchange of information

on financial accounts, including a possible further extension by three months for both categories.

The Commission proposed the latest amendment to the DAC (DAC7) on 15 July 2020,

with two

goals in mind. The first involves widening the scope of the DAC to include reporting on income

earned via digital platforms. This would concern platform sellers, not the taxation of the platform

economy as such. The objective is to ensure fair taxation of income generated through online

platforms in the internal market. The second involves addressing inefficiencies in administrative

cooperation and data exploitation, by means of introducing clarity as regards joint audits and group

requests, on the one hand, and the term 'foreseeable relevance', on the other.

In November 2020, the European Commission published an inception impact assessment

for a

future amendment of the DAC (DAC8). The inception impact assessment mentions the main specific

objectives of the intervention:

to enable tax administrations to obtain information that is necessary to make sure that

taxpayers, in particular taxpayers who earn money via crypto-assets, pay their fair share,

to provide for better cooperation across tax administrations,

to keep business compliance costs to a minimum by providing a common EU reporting

standard

After a planned public consultation phase in the first quarter 2021, the Commission intends to adopt

its proposal for the DAC8 in the course of the third quarter of 2021.

Lastly, the European Commission also proposed a codification of the DAC1-DAC6 on

12 February 2020.

This proposal was transmitted to the Council's preparatory bodies, and the

European Economic and Social Committee issued an opinion endorsing the proposed text in its

552th plenary session on 10 June 2020.

However, in view of the fact that the sixth and seventh

amendments (DAC7-8) were not part of the proposed codification, work on it is currently on hold.

The following table provides an overview of the DAC and its successive amendments.

Council Directive (EU) 2020/876 of 24 June 2020 amending Directive 2011/16/EU to address the urgent need to defer

certain time limits for the filing and exchange of information in the field of taxation because of the COVID-19

pandemic.

Proposal for a Council Directive amending Directive 2011/16/EU on administrative cooperation in the field of taxation,

COM(2020) 314 final, 15.7.2020.

For further details on the proposal and the accompanying impact assessment, see also: Kramer E. and Eisele K., Better

cooperation against tax fraud and evasion, Briefing, European Parliamentary Research Service, Brussels,

November 2020.

European Commission, Inception Impact Assessment for a Proposal for a Council Directive amending Directive

2011/16/EU as regards measures to strengthen existing rules and expand the exchange of information framework in

the field of taxation to include crypto-assets and e-money, 23.11.2020.

Proposal for a Council Directive on administrative cooperation in the field of taxation (codification), COM(2020) 49 final,

12.2.2020.

See Opinion of the European Economic and Social Committee, Proposal for a Council Directive on administrative

cooperation in the field of taxation (codification), adopted on 10 June 2020.

Implementation of the EU requirements for tax information exchange

Table 1: Directive on administrative cooperation (DAC)

DAC1 DAC1 DAC2 DAC3 DAC4 DAC5 DAC6 DAC7

2011/16/EU

NON-AEOI

2011/16/EU

AEOI items

2014/107/

EU AEOI

2015/2376/

EU AEOI

2016/881/

EU AEOI

2016/2258/

EU NON-

AEOI

2018/822/

EU NON-

AEOI

Not yet

adopted

Applies

1/2013

All

exchanges

of info

except Art. 8

Applies

1/2015

exchanges

on 2014 by

30.6.2015

Art. 8

Applies

1/2016

exchanges

on 2016 by

30.9.2017

Art. 8 para 3a

Applies

1/2017

exchanges

by 30.9.2017

Art. 8a

Applies

6/2017

exchanges

on 2016 by

30.6.2018

Art. 8aa

Applies

1/2018

Art. 22, para

Applies

7/2020

exchanges

by 31.8.2020

Art. 8ab and

hallmarks in

Annex 4

Exchanges

on request

Spontan.

exchanges

Presence in

adm.

offices

Simul tan.

controls

Requests

for

notification

Sharing

best

practices

Use of

standard

forms

Automatic

exchange of

information

on 5 non-

financial

categories:

Income from

employment

Director's

fees

Pensions

Life

insurance

products

Immovable

property

(income and

ownership)

Automatic

exchange on

financial

account

information:

Interests,

dividends and

other income

generated by

financial

account

Gross

proceeds

from sale or

redemption

Account

balances

Automatic

exchange of

information

(using a

central

directory as

from 1/2018)

of:

Advance

cross-border

rulings

Advance

pricing

arrange-

ments

Automatic

exchange of

information

on country-

by-country

reports on

certain

financial

information:

Revenues

Profits

Taxes paid

and accrued

Accumulated

earnings

Number of

employees

Certain

assets

Access by tax

authorities

to beneficial

ownership

information

as collected

under AML

rules

Mandatory

disclosure

rules for

intermediari

and

Automatic

exchange of

information

on tax

planning

cross-border

arrange-

ments

AEOI of

income

generated by

sellers on

digital

platforms

and

Explicit

inclusion of

joint audits

and group

requests

1.1.2. Implementing regulations

To ensure the uniform implementation of the DAC, the European Commission adopted several

Implementing regulations that are binding and directly applicable in all Member States. These

implementing regulations deal with the technical aspects of the implementation by the Member

States, specifying:

the standard forms for exchanges on request, spontaneous exchanges, notifications

and feedback; the computerised formats for the mandatory automatic exchange of

information; and the practical arrangements regarding the use of the Common

Communication Network (CCN),

the standard forms, including the linguistic arrangements, for the mandatory automatic

exchange of advance cross-border rulings and advance pricing arrangements, and the

Commission Implementing Regulation (EU) 2015/2378 of 15 December 2015 laying down detailed rules for

implementing certain provisions of Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Implementing Regulation (EU) No 1156/2012.

EPRS | European Parliamentary Research Service

mandatory automatic exchange of information on the country-by-country report

(CbCR),

the form and conditions for the communication of the yearly assessment and statistical

data,

the standard forms, including the linguistic arrangements, for the mandatory automatic

exchange of information on reportable cross-border arrangements.

1.2. Context of the study (DAC1-4)

In recent years, the European Commission and other EU Institutions have presented a number of

evaluations, reports and opinions on the functioning of the DAC. This chapter provides a short

overview of the available documents.

1.2.1. Commission evaluation and reports

Recital 24 and Article 23 of the directive require the European Commission (and Member States) to

evaluate the functioning of the administrative cooperation provided for in the directive. The

European Commission presented a first evaluation of administrative cooperation in direct

taxation,

accompanied by a Commission staff working document,

in 2019. The main conclusions

of this evaluation were that, while the directive seems to be effective and efficient, there is very little

evidence to substantiate this conclusion. The evaluation also stated that the directive is relevant,

appears coherent and has EU added value. The present study therefore examines in detail the level

of effectiveness of the directive and also provides information on the – essentially external –

coherence of the directive. The other three evaluation criteria are summarised in Chapter 7.

Article 27 of the DAC requires the European Commission to report every five years on the application

of the DAC. Accordingly, the European Commission published a first report

and an accompanying

staff working document

on the application of the DAC in 2017. Due to its publication date, this

report mainly covers the application of the DAC1. The key conclusions of this report are that the

implementation of the DAC1 is of limited effectiveness. In the area of exchange of information on

request (EOIR), only 62 % of full replies were given within six months over the 2013-2016 period.

According to the report, the adoption of the DAC3 (on tax rulings and advance pricing

arrangements) led to a significant increase in the number of spontaneous exchanges in these areas

in 2016. This shows that the effectiveness of EOIR had been rather limited in previous years. In the

Commission Implementing Regulation (EU) 2016/1963 of 9 November 2016 amending Implementing Regulation (EU)

2015/2378 as regards standard forms and linguistic arrangements to be used in relation to Council Directives (EU)

2015/2376 and (EU) 2016/881.

Commission Implementing Regulation (EU) 2018/99 of 22 January 2018 amending Implementing Regulation (EU)

2015/2378 as regards the form and conditions of communication for the yearly assessment of the effectiveness of the

automatic exchange of information and the list of statistical data to be provided by Member States for the purposes

of evaluating of Council Directive 2011/16/EU.

Commission Implementing Regulation (EU) 2019/532 of 28 March 2019 amending Implementing Regulation (EU)

2015/2378 as regards the standard forms, including linguistic arrangements, for the mandatory automatic exchange

of information on reportable cross-border arrangements.

Economisti Associati srl, Evaluation of Administrative Cooperation in Direct Taxation, 2019.

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

European Commission, Report from the Commission to the European Parliament and the Council on the application of

Council Directive (EU) 2011/16/EU on administrative cooperation in the field of direct taxation, COM(2017) 781 final,

18.12.2017.

European Commission, Commission Staff Working Document on the application of Council Directive (EU) no 2011/16/EU

on administrative cooperation in the field of direct taxation, SWD(2017) 462 final, 18.12.2017.

Implementation of the EU requirements for tax information exchange

field of automatic exchange of information (AEOI) under the DAC1, the main problem for the

application of the directive was the lack of resources and automated taxpayer identification

processes to make use of the information.

Finally, the European Commission published in 2018 a report on statistics within the AEOI

under

DAC1 to DAC3. In this report, the European Commission raised concerns about the quality of

information and the use of the data received by Member States via the AEOI. This report, which also

provided the data used in the 2019 Commission evaluation, made recommendations for enhancing

the effectiveness of the directive. First, the Commission concluded that the quality of the A EOI could

be improved through a quality review by the sending Member States and by feedback from the

receiving Member States, including feedback to the data suppliers, i.e. financial institutions. In this

context, a broader use of taxpayer identification numbers (TIN) could prove very useful. Second, the

Commission suggested that Member States should work on a common methodology to estimate

the benefits of the AEOI. Beyond possible improvements in the AEOI, Member States should also

make better use of the other formats of administrative cooperation and share best practices on the

better use of data available to them.

1.2.2. European Parliament committees on tax issues

The European Parliament has been very active in the field of tax evasion, tax fraud and tax avoidance

in recent years. Since 2015, three special committees and a committee of inquiry (in chronological

order: TAXE

, TAX2

, PANA

and TAX3

) looked at the different aspects of taxation, including the

exchange of tax information under the DAC. Most recently, the TAX3 resolution of 26 March 2019

highlighted a number of key areas for improvement with respect to the DAC amendments within

the scope of the present study (DAC1-DAC4). The table below shows these key areas and the action

taken and announcements made by the European Commission.

Table 2: EP resolution of 26 March 2019 and actions taken by the Commission

Resolution

Commission action/announcements

(Point 87) Calls on the Commission to assess

and present proposals to close loopholes in

DAC2, particularly by including hard assets and

cryptocurrencies in the scope of the directive,

by prescribing sanctions for non-compliance or

false reporting from financial institutions, as

well as by including more types of financial

institution and types of accounts that are not

being reported at the moment, such as pension

funds;

In 2019, the Commission published a Study on the

Evaluation of the Cooperation in Direct Taxation and a

staff working document

with the results of the

evaluation of Council Directive 2011/16/EU on

administrative cooperation in the field of taxation and

repealing Directive 77/799/EEC.

In the

Action Plan for fair and simple taxation

supporting the Recovery Package, published in

July 2020, the European Commission announced the

extension of automatic exchange of information to

European Commission, Report from the Commission to the European Parliament and the Council on overview and

assessment of the statistics and information on the automatic exchanges in the field of direct taxation, COM(2018)

844 final, 17.12.2018.

Special Committee on tax rulings and other measures similar in nature or effect (TAXE) - From 12 February 2015 to

30 November 2015.

Special committee on tax rulings and other measures similar in nature or effect (TAXE2) - From 2 December 2015 to

2 August 2016.

Committee of Inquiry to investigate alleged contraventions and maladministration in the application of Union law in

relation to money laundering, tax avoidance (PANA) - From 8 June 2016 to 8 December 2017.

Special committee on financial crimes, tax evasion and tax avoidance (TAX3) - From 1 March 2018 to 1 March 2019.

European Parliament, Resolution on financial crimes, tax evasion and tax avoidance, P8_TA(2019)0240, 26 March 2019.

EPRS | European Parliamentary Research Service

Resolution

Commission action/announcements

crypto-assets / e-money (Action 10). A proposal is

planned for the third quarter of 2021.

(Point 88) Reiterates its call for a broader scope

in relation to the exchange of tax rulings and

broader access by the Commission, and for

greater harmonisation of the tax ruling practices

of different national tax authorities;

According to the Commission, the findings of the 2019

evaluation will be followed up by the Member States in

the Working Group on Administrative Cooperation in

the field of Direct Taxation (ACDT), with the aim of

improving the functioning of the automatic exchange

under the DAC. This will include advance cross-border

tax rulings and advance pricing arrangements as well

as, if appropriate, considera

tions for possible future

initiatives to further amend the DAC.

(Point 89) Calls on the Commission to swiftly

release its first assessment of DAC3 in this regard,

looking in particular at the number of rulings

exchanged and the number of occasions on

which national tax administrations accessed

information held by another Member State; asks

that the assessment also consider the impact of

disclosing key information related to tax rulings

(the number of rulings, the names of

beneficiaries, the effective tax rate deriving from

each ruling); invites the Member States to publish

domestic tax rulings;

The 2019 Commission evaluation contains information

that points to an increase in the number of tax rulings

recorded in the central directory. The Commission has

not proposed/promised to take any

further specific

actions.

(Point 91) Reiterates, furthermore, its call to

ensure simultaneous tax audits of persons of

common or complementary interests (including

parent companies and their subsidiaries), and its

call to further enhance tax cooperation between

Member States through an obligation to answer

group requests on tax matters; points out that the

right to remain silent in dealings with tax

authorities does not apply to a purely

administrative investigation and that

cooperation is mandatory;

In its proposal for the DAC7, the Commission tries to

clarify the provisions of the DAC with respect to joint

audits, which so far have been conducted on the basis

of the combined provisions on the presence of foreign

officials in the territory of other Member States and on

simultaneous controls.

On group requests, the proposal

clarifies that

requested authorities would have to inform the

requesting authority of the reasons

in case the

requested authority takes the view that no

administrative enquiry is necessary.

(Point 95) Calls on the Commission to swiftly

assess the implementation of DAC4 and whether

national tax administrations effectively access

country‑by‑country information held by another

Member State; asks the Commission to assess

how DAC4 relates to Action 13 of the G20/BEPS

action plan on exchange of country

‑

country

information;

The assessment of the DAC4 implementation is part of

the 2019 Commission evaluation, which includes a

reference to the implementation of the standards of

the BEPS action 13.

1.2.3. Council of the European Union

Within the EU, legislation in the field of direct taxation is mainly a competence of the Member States.

The Council of the European Union plays therefore a central role and can adopt legislation in the

Implementation of the EU requirements for tax information exchange

field of direct taxation by unanimity.

The DAC1 and its subsequent amendments were adopted as

Council directives and not under the ordinary (co-decision) procedure, which meant that the

European Parliament was only consulted before their adoption.

In its conclusions of 2 June 2020, the Council expressed the need to update the DAC and to improve

administrative cooperation in order to reduce tax fraud, tax evasion and tax avoidance, es pecia lly in

the area of income generated via digital platforms, as proposed by the European Commission in

July 2020 (DAC7).

At the same time, the Council also invited the European Commission to analyse

possibilities for greater interoperability between administrative cooperation in the field of direct

taxation and anti-money-laundering legislation, and enhanced convergence between the DAC and

the Regulation on combating fraud in the field of value added tax.

1.2.4. European Court of Auditors' special report

The European Court of Auditors is currently working on a special report on the exchange of tax

information in the EU. The publication of the report is planned for January 2021 and will assess the

effectiveness of the system set up by the Commission for the automatic exchange of tax

information. The report will look into the use of tax information received by the tax administrations

in five EU Member States (the Netherlands, Luxembourg, Poland, Spain and Italy).

1.2.5. European Economic and Social Committee opinion

The European Economic and Social Committee (EESC) adopted an own-initiative opinion on the

European Commission's proposal for the DAC7 on 18 September 2020.

The opinion calls for:

the closure of gaps in the DAC concerning missing beneficial ownership information in

certain countries;

extending the scope of the DAC to also cover works of art and other high-value assets

located in free ports and customs warehouses;

public access to country-by-country reports, at least for companies receiving public aid.

As regards the fight against money laundering and tax evasion and avoidance, the EESC considers

the cooperation between the tax authorities in some EU Member States to be still unsatisfactory.

1.3. Scope and methodology of the study

For all legislation, there is a certain time lag between its adoption and the first collection of data on

its implementation. For that reason, the European Commission's Better Regulation Guidelines

recommend carrying out evaluations only once at least three years of a reasonable amount of data

has been collected.

In the case of the DAC, only the initial directive and the first three amendments

fulfil this condition. The main focus of this study is therefore limited to the DAC1-DAC4.

Based on Article 115 TEU, '...the Council shall, acting unanimously...issue directives for the approximation of such laws,

regulations or administrative provisions of the Member States as directly affect the establishment or functioning of

the internal market'.

Council conclusions of 2 June 2020 on the future evolution of administrative cooperation in the field of taxation in the

EU.

Council Regulation (EU) No 904/2010 of 7 October 2010 on administrative cooperation and combating fraud in the field

of value added tax.

European Economic and Social Committee, Opinion on effective and coordinated EU measures to combat tax fraud, tax

avoidance, money laundering and tax havens, ECO/510, 18 September 2020.

European Commission, Better Regulation Guidelines, SWD(2017) 350, 2017.

EPRS | European Parliamentary Research Service

Whereas the European Commission evaluation in 2019

covered all five evaluation criter ia

(effectiveness, efficiency, coherence, relevance and EU added value), the present study and the

annexed research paper look mainly into the effectiveness and coherence of the implementation of

the DAC. A brief summary of the other evaluation criteria, based on the Commission evaluation, can

be found in Chapter 7 of the present study.

As the directive regulates the exchange of administrative taxation data between EU Member States,

the geographical coverage of the study is to a large extent limited to the EU. In addition to the EU

Member States, the annexed research paper also provides a short analysis of the implementation of

current reporting obligations and tax information exchange flows by countries and jurisdictions

previously covered by the EU Savings Tax Directive or under similar obligations in an EU agreement

(Andorra, Cayman Islands, Liechtenstein, Monaco, San Marino and Switzerland). With the exception

of the Cayman Islands, EU Member States' overseas countries and territories (OCTs) are not covered

by the present study. An earlier EPRS study

for the PANA committee has provided detailed

information on OCTs.

Finally, the research paper also looks also into the exchange of tax information at the international

level, namely in the context of the OECD/Global Forum using the Common Reporting Standard

(CRS), and between the EU and the US. Besides describing how tax information is exchanged and

what the potential shortcomings of this process are, the research paper describes the differences

between the US Foreign Account Tax Compliance Act (FATCA) and the DAC, and analyses the

reciprocity including in the intergovernmental agreements (IGA) under the FATCA.

Besides the publicly available information mentioned above, the study's main source of evidence is

the research paper in Part II, which was carried out by a consortium led by LE Europe. This res earch

paper concentrates on three broad areas:

effectiveness of the DAC. What types of income are covered/not covered by the DAC?

Have new types of income appeared after the DAC was first adopted? What are the legal

or practical obstacles to the exchange of tax information? What are the known examples

of circumvention or loopholes for the different types of information exchange under the

DAC? Definition and reporting obligations with regard to financial accounts and

beneficial ownership. This part also contains a short case study on the Cum-Ex tr ading

scandal and the description of a theoretical approach on how to use macroeconomic

data for a plausibility check of DAC2 data. A mapping of bilateral exchange flows for the

different types of exchanges was planned to be done in the context of the study, but

the lack of data rendered this impossible (see Chapter 4 for more details).

evolution of tax information exchange. Here, the intention was to assess the

evolution of the yearly tax data exchange, at least until 2018. This analysis was supposed

to build on the data already included in the Commission evaluation and the underlying

study, both published in 2019, which referred to data until 2017. As the data needed for

this analysis was not accessible to the researcher (see Chapter 4), the analys is presented

in the annexed research paper refers mainly to the Commission study and to data

collected from a few Member States.

coherence. This part of the research paper looks into the differences between the

standards used for under three different frameworks of tax information exchange (the

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

I. Ioannides and J. Tymowski, Tax evasion, money laundering and tax transparency in the EU Overseas Countries and

Territories, EPRS, European Parliament, 2017.

Implementation of the EU requirements for tax information exchange

DAC, the Global Forum's Common Reporting Standard (CRS), and the FATCA). It also

provides a short analysis of the implementation of current reporting obligations and tax

information exchange flows by countries and jurisdictions previously covered by the EU

Savings Tax Directive or under similar obligations in an EU agreement (Andorra, the

Cayman Islands, Liechtenstein, Monaco, San Marino and Switzerland). Finally, the study

gives an overview of the interaction between the provisions of the DAC and the EU r ules

on judicial assistance and anti-money-laundering.

The following methods were used for drawing up the study:

desk research and literature review in all three areas described above;

structured interviews with tax authorities on the effectiveness and coherence of the

DAC. These interviews were conducted with the national tax authorities of 10 EU

Member States and with officials working at the European Commission's Directorate-

General for Taxation and Customs Union. The Member States were chosen to represent

a diversity of sizes (small, medium and large) and geographical locations. The interviews

focused on the legal and practical obstacles to the exchange of tax information, on

circumventions and loopholes, and on the interoperability between the provisions of

the DAC and the EU rules on judicial assistance, money laundering and asset recovery

from a practical point of view. To measure the effect of the repeal of the Savings Tax

Directive, the interviews with national tax administrations included also questions on

the tax information exchange with countries previously covered by the Savings Tax

Directive or under similar obligations in an EU agreements;

a targeted stakeholder survey on the effectiveness and coherence of the DAC. Similar

to the interviews with the 10 Member States' tax administrations, this online survey

focused on the legal and practical obstacles to the exchange of tax information, as well

as on circumventions and loopholes. The survey also included practical questions on

the interaction between the provisions of the DAC and the EU rules on judicial

assistance, money laundering and asset recovery.

EPRS | European Parliamentary Research Service

2. Scope of the DAC and its amendments

This chapter provides a detailed description of the content of the DAC and its amendments, in

addition to the information laid out in Table 1 in Chapter 1.1.1.

2.1. Forms of income and capital covered by the DAC1-2

On 1 January 2015, the provisions on automatic exchange of information, which had been built into

Directive 2011/16/EU (DAC1), entered into force, thereby creating an obligation for Member States

to report with each other on the following available taxable income and capital categories of

residents in other Member States:

income from employment (EI);

director's fees (DF);

life insurance products (LIP) not covered by other EU legal instruments on exchange of

information and other similar measures;

pensions (PEN);

ownership of and income from immovable property (IP).

Since September 2017, a mandatory automatic exchange of information has been taking place as

required under the DAC2. Apart from specifying what kind of identification-related information is

needed for the exchange (account holder's name, account number, name of the reporting financial

institutions, etc.), these provisions also set the requirement for the exchange of the following

financial account information:

account balance or value;

amounts of dividends and interests;

other financial income;

capital gains.

As mentioned earlier, this broadening of the DAC's scope initiated by the European Commission in

order to better fight tax fraud, tax evasion and tax avoidance in EU Member States. At the same time,

this amendment led also to the repeal of the Savings Tax Directive,

under which EU Member States

had exchanged information on private savings income since its entry into application in 2005.

2.2. Cross-border tax rulings and advanced pricing arrangements

(DAC3)

In response to the LuxLeaks scandal, the DAC3 introduced provisions on the automatic exchange

on advance cross-border rulings and advance price arrangements.

The information exchange

covers taxpayers of all business formats (e.g. legal persons such as companies, and legal

arrangements such as trusts and foundations), but excludes rulings relating exclusively to natural

persons. The exchange takes place via a central directory database to which all Member States have

Council Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments.

'An advance cross-border tax ruling is a confirmation or assurance that tax authorities give to taxpayers on how their tax will

be calculated in a cross-border situation before the transaction takes place. Similarly, an advance pricing arrangement

determines in advance of cross-border transactions an appropriate set of criteria between associated enterprises (i.e. group

companies) for the determination of transfer prices or determines the attribution of profit to a permanent establishment.'

Source: https://ec.europa.eu/taxation_customs/sites/taxation/files/council_directive_eu_2015_2376_en.pdf

Implementation of the EU requirements for tax information exchange

access. In contrast the other types of AEOI, shared via the common communication network (CCN),

the European Commission can access DAC3 information in the central directory database only for

monitoring purposes.

2.3. Country-by-country reports (DAC4)

In 2017, the DAC4 introduced the requirement for multinational enterprise (MNE) groups located in

the EU or with operations in the EU to file a country-by-country report (CbCR) if the MNE group's

total consolidated revenue is equal to or higher than €750 million. Such a CbCR has to be sent

automatically by the receiving competent authority of the Member State to any other Member

States in which one or more constituent entities (i.e. companies) of the MNE group are either

resident for tax purposes or subject to tax with respect to the business carried out through a

permanent establishment there.

The CbCR needs to contain the following information:

Aggregate information relating to the:

amount of revenue,

profit (loss) before income tax,

income tax paid,

income tax accrued,

stated capital,

accumulated earnings,

number of employees,

tangible assets other than cash or cash equivalents with regard to each

jurisdiction in which the MNE group operates.

For each constituent entity of the MNE group:

the jurisdiction of tax residence of that constituent entity,

where different from that jurisdiction of tax residence, the jurisdiction under

the laws of which that constituent entity has been established,

the nature of the main business activity or activities of that constituent entity.

EPRS | European Parliamentary Research Service

3. Transposition of the DAC1-6

The following sections give an overview of how the transposition and infringement procedures

under the DAC have evolved with each consecutive amendment of the DAC.

3.1. Transposition by Member States

The table below shows the timetable for the transposition of the different amendments of the DAC

into the Member States' national legislation and for their subsequent entry into application. In the

case of the DAC2, Austria had requested an extension of the deadlines by one year. To cover the

period of the late entry into application of the DAC2, Austria applied the provisions of the Savings

Tax Directive, which was repealed by the DAC2, for an additional year.

Table 3: Transposition of the DAC (state of play as of January 2021)

Directive

Transposition

deadline

Application as of

Infringements

opened

Infringements

still open

DAC1 31/12/2012 01/01/2013 14 0

DAC1 – AEOI

31/12/2012

01/01/2015

DAC2

31/12/2015

01/01/2016*

DAC3

31/12/2016

01/01/2017

DAC4

04/06/2017

05/06/2017

1 (ES)

DAC5

31/12/2017

01/01/2018

1 (IE)

DAC6

30/06/2020

01/07/2020

* 01/01/2017 for Austria.

3.2. Past and present infringement procedures

The varying number of infringement cases (see Table 3) for the respective amendments of the DAC

reflects the complexity of the amendments as well as the difficulties experienced by the Member

States in transposing the directive into their national legislation. In relation to the DAC1, almost all

opened infringement procedures were due to late transposition. Only one infringement procedure

was related to the incomplete transposition of the AEOI provisions by Estonia.

While all infringement cases concerning the transposition of the DAC2 are now closed, the relatively

high number of cases that had been opened can be explained by the complex implementation at

the Member State level. On the one hand, the amendments required the involvement of the

financial sector in the process and the development of new IT structures. On the other hand, for

some Member States, the transposition coincided with the preparation of multilateral exchanges at

OECD level and the first bilateral exchanges with the US under the FATCA.

Late transposition and/or notification of the European Commission were the reasons for the nine

infringement procedures with respect to the DAC3. All infringement procedures related to the

DAC1-3 are now closed.

Economisti Associati srl, Evaluation of Administrative Cooperation in Direct Taxation, 2019.

Implementation of the EU requirements for tax information exchange

All seven infringement procedures that had been opened for late transposition of the DAC4 are now

closed. However, the European Commission opened a new procedure against Spain in 2019. The

current rules in Spain on CbCRs by multinational companies lack a number of elements regarding

the reporting obligations.

Eleven Member States received letters of formal notice for non-compliance with the transposition

deadline for the DAC5. Most infringement cases were closed at an early stage. Ireland's case, initiated

because of the limited access afforded to the tax administration to beneficial ownership

information, is still open.

In the case of the DAC6, the European Commission sent letters of formal notice

to 15 Member

States in January 2020. Most of these Member States notified their response to the Commission's

letters of formal notice with a delay in the course of 2020. Whether these notifications are complete

and allow for closure of the infringement procedures is still being assessed. The infringement

procedures for France, Luxembourg, Poland, Portugal, Romania and the United Kingdom are now

closed, while the ones for Italy, Greece, Sweden, Belgium, Estonia, Spain, Czechia, Latvia and Cyprus

were still open at the time when the present study was being drafted (mid-January 2021).

A letter of formal notice is the first step in an EU infringement procedure, '...giving the State concerned the opportunity to

submit its observations'. (Article 258 TFEU). More information on the different stages of an infringement procedure can

be found on the relevant European Commission website:

https://ec.europa.eu/info/law/law-maki ng-

process/applying-eu-law/infringement-procedure_en.

EPRS | European Parliamentary Research Service

4. Data analysis

The European Commission evaluation and the underlying study, both published in 2019, contain

valuable information on the tax information exchange flows emanating from the different DAC

provisions. However, the time series presented in these documents ended in 2017, which can be

explained by the publication date in 2019. The study also gave an overview of the main sending and

receiving countries of AEOI exchanges under the DAC1-3, but not under the DAC4. A description of

bilateral flows of AEOI and other forms of exchanges was also a part of the study, but only covered

the main flows, without providing a complete picture of the bilateral exchange flows in the EU.

One of the objectives of the present study was to complement the exchange flow time series with

2018 data and to give a better overview of the exchange flows, including under the provisions of

the DAC4, and of all bilateral exchange flows and Member States' replies to the questionnaires on

the yearly assessment of the DAC and its functioning. To this end, the European Parliament

requested to be granted access to DAC data by the European Commission. The chair of the European

Parliament's Economic and Monetary Affairs Committee (ECON) and the director-general of DG

TAXUD signed an arrangement for the examination of confidential documents and information as

regards the implementation of the DAC. As the request concerned confidential data submitted by

the Member States, the European Commission was bound by the Framework Agreement on

relations between the European Parliament and the European Commission

to ask for the Member

States' consent before providing the documents to the European Parliament. The Member States,

represented by the Council, refused to grant the European Parliament access to the data that would

have been necessary for the drawing up of the present European Implementation Assessment (EIA).

In general, countries' – not only EU Member States' – reluctance to provide data on tax information

exchanges is not new, as such information is considered confidential.

Due to the Member States' refusal to grant access to data for the purposes of this study, the

following subchapters and the annexed research paper relied to a large extent on publicly available

information, i.e. the relevant documents that the European Commission has published in recent

years. In addition, the above-mentioned sources – interviews with national tax administrations and

European Commission officials and a stakeholder survey – were used to collect additional

information. Finally, national parliaments also provided useful information on the implementation

of the DAC at the national level in reply to a request sent via the European Centre for Parliamentary

Research and Documentation.

Chapter 3.1.2 of the annexed research paper provides an overview

of these national data sources on information exchange under the DAC1-4.

4.1. Data availability and limitations

One of the main limitations of the 2019 European Commission evaluation of the DAC was that its

conclusions were 'grounded on limited and in some cases very thin evidence, if any at all'.

Any

Framework Agreement on relations between the European Parliament and the European Commission, OJ L 304,

20.11.2010, p. 47–62. In particular, Annex II, paragraph 2.1 of this agreement stipulated that '...Confidential information

from a State, an institution or an international organisation shall be forwarded only with its consent'.

M. Keen and J.E. Ligthart, Information Sharing and International Taxation: A Primer, International Tax and Public Finance,

13, pp. 81–110, 2006.

The ECPRD is a network of the European Parliament and the national parliaments of the Council of Europe countries. For

the purposes of this study, an information request was sent to the EU Member States only. For more information on

the ECPRD, see: https://ecprd.secure.europarl.europa.eu/ecprd/public/page/about

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

Implementation of the EU requirements for tax information exchange

conclusions drawn in the present study have to be seen in this context, especially given the above-

mentioned limitations on access to data. However, the fact that the European Commission and its

external contractor had access to the whole range of data on tax information exchanges under the

DAC allowed for a more complete picture of these exchanges. At the same time, the information

provided by the European Commission reflects the state of affairs in early 2019, i.e. the time when

the external study was finalised.

For the DAC1 AEOI, the European Commission evaluation showed a constant increase in the number

and value of exchanges from 2015 to mid-2017. The main problem with this type of exchange

seemed to be the availability of information on certain types of income and assets in a number of

Member States.

Member States have to send information only for those categories for which

information is already available. The lowest rate of information availability was observed for life

insurance products, for which only eight Member States could provide information in 2017.

According to the Commission's external study, enhanced coverage of income categories could be

achieved without additional information collection, simply by a better 'management and quality of

information already available'.

The DAC2 introduced the requirement for mandatory automatic exchange of all financial account

information, and data availability was therefore less problematic than for the DAC1. Yet, where the

necessary information was not readily available, Member States had to introduce new reporting

obligations. This process happened more or less at the same time as the introduction of multilateral

information exchanges under the Global Forum and the US FATCA.

As DAC2 exchanges were

launched in September 2017, this part of the European Commission evaluation covered only a

seven-month period, which made it impossible to draw any conclusions from the evolution of the

exchanges over time.

Data- and information availability is even worse when it comes to advance cross-border tax r ulings

(ATR) and advance pricing arrangements (APA) under the DAC3. Member States have exclusive

access to the central directory where ATRs and APAs are uploaded, which means that no information

is available on their content to other entities. Although the central directory existed already before

the adoption of the DAC3, most Member States only started using it with the implementation of the

DAC3 in 2017.

The European Commission evaluation did not look into the use of CbCRs under the DAC4;

quantitative information on this amendment is therefore very limited. In the context of the present

EIA, only one national parliament (Germany) informed the researchers about the existence of data

on CbCRs for the years 2016-2018.

4.2. National data sources

Given the absence of access to data and information for the purposes of this study, its authors asked

the Member States' national parliaments to provide them with information on the implementation

of the DAC. In response, they received publicly available information on the DAC's implementation

With respect to the five income and capital categories for automatic exchange under DAC1, Article 8(1) of Directive

2011/16/EU specifies that Member States 'shall ... communicate to the competent authority of any other Member

State, information ... that is available concerning residents of that other Member State'.

Economisti Associati srl, Evaluation of Administrative Cooperation in Direct Taxation, 2019.

ibid.

Deutscher Bundestag, Schriftliche Fragen mit den in der Woche vom 6. Juli 2020 eingegangenen Antworten der

Bundesregierung, Drucksache 19/20953

, 6.7.2020. See also Chapter 3.1.2 of the annexed research paper.

EPRS | European Parliamentary Research Service

from 23 national parliaments. Some Member States were able to provide quantitative information

on the DAC1-4, which is presented in Chapter 3.1.2 of the annexed research paper. Information

sources varied across Member States, the most valuable being replies to parliamentary questions

and reports by the national courts of auditors. While this information provided useful insights into

the implementation and use of DAC data in the 23 Member States, it was not comprehensive

enough to serve the needs of a full-fledged data analysis.

Implementation of the EU requirements for tax information exchange

5. Effectiveness of the DAC

The three specific objectives of the DAC are '(i) an increased ability to fight cross-border tax fraud,

evasion and avoidance; (ii) reduced scope for harmful tax competition, namely through greater

transparency in tax rules; and (iii) enhanced spontaneous tax compliance in a timely manner,

through the 'deterrent effect' resulting from the greater ability to detect cross-border incomes and

assets'.

While the 2019 European Commission evaluation did not detect any major problems in

terms of the DAC's effectiveness, it highlighted the issue of limited evidence as to its effectiveness,

due to the short time that had elapsed between the implementation and the evaluation of the

directive. The present study and the annexed research paper attempt to fill this gap by assessing

three features of the DAC:

comprehensiveness of scope: to what extent do the DAC provisions capture all the

relevant categories of income and capital in order to effectively address tax fraud, tax

avoidance and tax evasion in an evolving globalised context?;

relevance or appropriateness: are the requirements in relation to the information

being exchanged (persons and accounts) and to the role and duties of the reporting

entities and the competent authorities, relevant and appropriate for the purposes of

allowing Member States to assess tax correctly?;

timeliness: are the time limits set out in the provisions for the production and exchange

of information appropriate in order for Member States to be able to assess tax correctly

and in a timely manner, and deter potential tax fraud, avoidance and evasion?

5.1. Market reactions to the DAC

In this part, the annexed research paper looks into the evolution of the types of income, gives an

account of how the DAC provisions are circumvented and what loopholes they have, and to what

extent these three features are addressed by the DAC.

5.1.1. Income categories

In Chapter 2.1.2, the annexed research paper establishes two lists of income categories that are i)

either covered or ii) not (yet) covered by the different DAC amendments, in part because capturing

certain income categories in practice proves to be difficult.

The income categories not (yet) covered by DAC consist of royalties, crypto-currency assets and

related income, income from online platforms and interest on loans through peer-to-peer lending

platforms or crowdfunding platforms. In the meantime, the European Commission has already

proposed to include these income categories in the DAC7 (for royalties, income and interest on

loans from online platforms), or will probably do so in the near future in the DAC8 (for crypto-

currencies).

Other forms of income should in theory be captured by the existing DAC provisions, but there seem

to be difficulties in the distinction from other forms of income. This seems to be the case for

directors' fees and payments in kind, which might however be captured as employment income.

Finally, capital gains related to immovable property or to the sale of financial or other types of assets

are not covered by DAC. The same applies to payments in kind related to financial instruments. In

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

EPRS | European Parliamentary Research Service

the case of capital gains taxes, the limited data availability for different types of income in many

Member States might be a reason for not including capital gains in the DAC provisions. In addition,

capital gains tax regimes differ noticeably between Member States and are often modified.

The

financial impact of these categories is therefore difficult to assess.

Apart from the potential inclusion of capital gains, the research paper does not mention other forms

of income for which it could make sense to widen the coverage under DAC provisions.

5.1.2. Circumventions and loopholes

Based on a literature review they performed, the authors of the research paper detected a number

of circumventions and loopholes; none were detected in the course of the interviews with the

Member States and the stakeholder survey conducted for the research paper. The most problematic

circumventions and loopholes are presented in the conclusions.

Regarding the DAC2, the research paper points to a series of loopholes and circumventions in four

broad areas:

reporting: some types of financial institutions (FI) and accounts are excluded from the

reporting obligations;

tax residence: for different reasons, the information on tax residence held by FIs can be

wrong or outdated. One such example, described in detail, are the Investor Citizenship

and Residence schemes;

beneficial ownership is not reported below certain thresholds or for certain active

entities or trusts;

due diligence procedures and correctness of information: due diligence procedures

often rely on the self-certification of account holders only or on anti-money-

laundering/know-your customer documentation, which is of varying quality in the

Member States.

In addition to these four areas, the research paper also includes a case study on cum-ex/cum-cum

schemes. This case study concludes that the extended scope of the DAC could nowadays prevent

such schemes or at least have a deterrence effect, even if the pace at which the trade under these

schemes took place would make it difficult to detect them on time. In the past, existing measures

were underused and better cooperation, notably a more extensive use of spontaneous exchanges,

could have allowed Member States to sufficiently tackle the schemes.

For the DAC3, the research paper highlights loopholes with respect to pre-2012 ATRs and APAs and

the fact that the DAC3 does not cover ATRs/APAs concerning natural persons or domestic

arrangements, which might however have an impact for another Member State. In addition, it is

possible to circumvent DAC3 reporting obligations by submitting information on a tax ruling or

pricing arrangement only after (and not in advance of) making a transaction, creating a permanent

establishment or filing returns. The research paper also mentions the annual net turnover threshold

of €40 million under which ATRs or APAs for non-financial or investment corporations do not need

to be reported, though the extent and relevance of this loophole seems to be limited.

Concerning the DAC4, one identified loophole refers to the fact that MNE groups with a total

consolidated group revenue of less than €750 million do not have to present CbCRs. In addition,

S. Princen et al., Taxation of Household Capital in EU Member States: Impact on Economic Efficiency, Revenue &

Redistribution, European Commission Directorate-General for Economic and Financial Affairs Discussion Paper 130,

August 2020.

Implementation of the EU requirements for tax information exchange

limited sanctions and penalties in certain Member States can affect the MNE group's willingness to

comply with the DAC4 provisions.

5.2. Obstacles to the exchange of information

In addition to circumventions and loopholes, the annexed research paper offers an overview of legal

and practical obstacles that might hamper the effectiveness of the DAC. Most of these obstacles

influence the effectiveness of the DAC, but can also affect other policy areas, notably when it comes

to disclosure of DAC information for other purposes.

5.2.1. Legal obstacles

The research paper identifies legal obstacles in the following areas:

business secrets, which allow Member States to refuse the exchange of information if

the exchange 'would lead to the disclosure of a commercial, industrial or professional

secret or of a commercial process, or of information whose disclosure would be contrary

to public policy';

public disclosure or more precisely, non-disclosure, e.g. of ATRs and APAs, which can

be perceived as a lack of transparency on tax matters (one of the specific objectives of

the directive);

legislative changes and resulting delays in the exchange;

legal interpretation of DAC provisions and definitions;

legal challenges against DAC provisions;

other challenges, e.g. the unclear definition of the term 'foreseeable relevance'. The

definition of this term is part of the DAC7 proposal adopted by the European

Commission in July 2020;

purpose of the exchange, i.e. the disclosure of DAC information for other purposes.

This point seems however to rather affect other policy areas and not the DAC itself.

5.2.2. Practical obstacles

Practical or technical obstacles can relate to the directive's effectiveness in general or to specific

amendments.

The following practical obstacles of a general nature have been identified:

tax identification numbers (TIN): the absence of TINs in the tax information

exchanged by many Member States hampers its effective and efficient use;

quality of reporting: apart from Member States' concerns about the quality of EOIR

requests, quality of reporting seems to have been a (limited) problem in the early phases

of DAC implementation;

timeliness: for EOIR, only 55 % of replies were given within the six-month deadline as

of 2017.

Even when reporting deadlines are respected by Member States, AEOI (and

to a lesser extent EOIR) cannot detect in a timely manner schemes of the cum-ex or cum-

cum kind due to the high-frequency trading used for such schemes;

technical obstacles: limitations to the technical capacities of (mainly smaller) Member

States.

Article 17(4) of Council Directive 2011/16/EU.

Economisti Associati srl, Evaluation of Administrative Cooperation in Direct Taxation, 2019.

EPRS | European Parliamentary Research Service

The following practical obstacles linked to a specific DAC amendment have been identified:

DAC1: limited availability for some income categories;

DAC2: difficulties in allocating specific individual account holders to joint accounts;

DAC3: the summary information in the central directory is often too brief to be used

without having to request additional information. Tax authorities can decide to

categorise an ATR or APA as being of a cross-border nature and hence not upload them

in the central directory;

DAC4: as there is no common format for filing the CbCRs, this leads to a lower quality

of the CbCRs.

See also Chapter 4.1. on data availability.

Implementation of the EU requirements for tax information exchange

6. External coherence of the DAC

This chapter analyses the exchange of tax information in the international context, that is, beyond

the exchange within the EU. The first part looks into the differences and similarities in the exchange

between EU Member States using the Common Reporting Standard (CRS) for the DAC2 and the US

Foreign Account Tax Compliance Act (FATCA). Beyond the scope of the DAC itself, the chapter also

gives an overview of the tax information exchanged by EU Member States at the international level

within the context of the Global Forum on Transparency and Exchange of Information for Tax

Purposes. It also addresses the question of exchange with countries that were previously covered

by the provisions of the Savings Tax Directive. The second part focuses on the interaction between

the DAC and EU legislation in the field of anti-money-laundering and judicial assistance.

6.1. Comparison between the DAC2/CRS and the FATCA

Although legal definitions and standards for the automatic exchange of information (AEOI) are the

same for the CRS and the DAC2, the latter goes to some extent beyond the obligations under the

CRS, e.g. with respect to wider reporting obligations on the place of birth of reportable persons

under the DAC2 or the exemption for certain forms of insurance under the CRS obligations, which

is not applicable in the EU context.

Apart from these minor differences, the DAC2 and the CRS are

largely coherent.

One – nowadays only theoretical – source of differences between the two frameworks is the

mandatory nature of DAC rules for all EU Member States, whereas membership in the OECD/Global

Forum is not mandatory. However, all EU Member States are also members of the Global Forum and

participate in the EOIR and the AEOI.

The major difference between the provisions of the CRS and the FATCA concerns the group of

persons covered by the provisions. In the case of the FATCA, its provisions cover foreign asset

holdings by US persons and US residents holding financial assets outside the US, whereas the CRS

focuses on all residents of a jurisdiction (from a tax perspective) that hold financial assets in other

jurisdictions that have signed a CRS agreement with the home jurisdiction of the resident.

Applied

in the context of the DAC, the term 'jurisdiction' has been replaced by the term 'Member State' and

there is no need for an agreement between the Member States, as the DAC applies directly within

the confines of the EU.

Another difference between the FATCA and the CRS consists in the thresholds below which

accounts or contracts do not have to be reported. The FATCA has such thresholds, while the CRS

does not. A detailed overview of the differences between the FATCA and the CRS is provided in

Annexes 2 and 3 of the annexed research paper.

See Annex III of European Commission, Implementation of Directive 2014/107/EU, WPIV Meeting on 19 June 2015.

The role of US citizenship in the framework of the FATCA provisions and intergovernmental agreements (IGAs) signed

between the EU Member States and the US can lead to a situation where 'accidental Americans' have to comply with

US tax rules. Such a scenario could not happen under DAC2/CRS provisions, since citizenship is not a determining

factor under these frameworks. The case of 'accidental Americans' who, by accident of birth, have American

citizenship, has also been taken up the European Parliament. In December 2020, the German Council Presidency sent

a letter to ask the US Treasury for a meeting to discuss the issue. See also: European Parliament,

Resolution on the

adverse effects of the US Foreign Account Tax Compliance Act (FATCA) on EU citizens and in particular 'accidental

Americans', P8_TA(2018)0316, 5 July 2018.

EPRS | European Parliamentary Research Service

6.1.1. Global Forum cooperation on the AEOI

The automatic exchange of tax information at the international level uses the Common Reporting

Standard for the automatic exchange of financial account information in tax matters (AEOI) that was

adopted by the Global

Forum on Transparency and

Exchange of Information for

Tax Purposes (the Global

Forum). The OECD, which

developed the standard,

publishes yearly

implementation reports on

the AEOI. Figure 1 shows

that all EU Member States

participate in the AEOI at the

international level. There

are, however, some

differences in the number of

jurisdictions with which EU

Member States exchange

information, ranging from

6O in the case of Czechia to

69 in the case of Belgium , for

data exchanged in 2019.

While the general increase in

the number of jurisdictions

to which AEOI information

was sent is a positive sign, it

has to be made clear that

this does not allow for any

conclusions on the quality of

the information and the use

by the receiving countries.

6.1.2. Global Forum cooperation on the EOIR

In addition to the AEOI, the Global Forum also established a framework for the exchange of

information on request (EOIR). All EU Member States participate in this exchange, yet the

compliance assessment made by the Global Forum in its peer reviews shows some differences

between the EU Member States. The figure below shows the most recent assessments of all EU

Member States. Overall, most are compliant or largely compliant. Malta is the only one that received

a 'partially compliant' assessment in its 2020 peer review,

due to the need to enhance the

supervision and enforcement of its legal requirements on the availability of ownership, identity,

OECD, The 2019 AEOI Implementation Report, OECD, Paris, 2019.

OECD, Global Forum on Transparency and Exchange of Information for Tax Purposes: Malta 2020 (Second Round): Peer

Review Report on the Exchange of Information on Request, Global Forum on Transparency and Exchange of

Information for Tax Purposes, OECD Publishing, 2020.

Figure 1: Number of countries to which AEOI information

was sent, 2017-2019

Source: OECD, AEOI Implementation Report 2019.

Implementation of the EU requirements for tax information exchange

banking, and accounting information. Another issue raised in the peer review is the high number of

inactive companies in the Maltese business and tax registers. In addition, a lack of resources has led

to issues with the timeliness in the replies to the increasing number of received EOI requests by the

Maltese tax administration.

Altogether, the main problem as regards compliance to EOIR requirements seems to be in the field

of ownership and identity information. In total, nine EU Member States received a 'part ially

compliant' or 'non-compliant' assessment in this area in their most recent peer reviews. In the case

of Poland, the 'non-compliant' assessment ownership and identity information (made in 2015) is

due to the lack of such information, e.g. for foreign companies that have tax residence in Poland.

The figure below shows the overall assessment and the assessment of all 10 sub elements (A1-C5)

in the most recent peer review for all EU Member States.

OECD, Global Forum on Transparency and Exchange of Information for Tax Purposes, Peer Reviews: Poland 2015: Phase

2: Implementation of the Standard in Practice, OECD Publishing, 2015.

Figure 2: Global Forum EOIR – compliance ratings of EU Member States

Source: OECD, Global Forum website.

EPRS | European Parliamentary Research Service

6.2. The DAC and the Savings Tax Directive

As already explained in the introductory chapter, the DAC2 repealed the Savings Tax Directive (STD)

that had been adopted in 2003. Overall, the scope and provisions are broader in the case of the

DAC2, with the exception of reporting obligations of paying agents, which were covered by the STD

but not by the DAC2 (the latter refers only to 'reporting financial institutions').

The STD applied not only to all EU Member States; agreements for equivalent or similar measures

were also signed with five European non-EU countries (Andorra, Liechtenstein, Monaco, San Marino

and Switzerland) and with 10 dependent and associated territories (Aruba, Anguilla, Guernsey,

Jersey, the Isle of Man, the Cayman Islands, the British Virgin Islands, the Netherlands Antilles,

Montserrat and the Turks and Caicos Islands).

All five European non-EU countries applied a

withholding tax that started at 15 % in 2005 and was raised to 35 % in 2011.

Among the other 10

territories, Anguilla, Monserrat, the Cayman Islands and Aruba participated in the information

exchange, while the other six applied a withholding tax regime.

After the repeal of the STD, the agreements with the five European non-EU countries were amended

to align them with the obligations under the DAC2, whereas the other territories committed to

participate in the AEOI within the Global Forum/OECD framework. All 10 territories have signed the

CRS Multilateral Competent Authority Agreement; however, according to the OECD/Global Forum,

the legal frameworks are not in place in Aruba, Curaçao and Sint Maarten (which were both part of

the former Netherlands Antilles),

and there are improvements to be made in the legal frameworks

of the British Virgin Islands and the Turks and Caicos Islands.

6.3. The use of DAC information for other purposes

While the chapter above looked into the coherence between the DAC and similar frameworks for

tax information exchange at the international level, this chapter focuses on the interaction between

the DAC and other EU secondary legislation. The main focus is on anti-money-laundering provisions,

complemented by a short chapter on judicial assistance and asset recovery legislation.

6.3.1. Anti-money-laundering provisions

Beneficial ownership (BO) information is the main feature in the link between the DAC and anti-

money-laundering (AML) legislation. Under the DAC2, financial institutions (FI) must identify the

controlling persons for accounts of passive non-financial entities. To do so, FIs use BO information

obtained under the fourth Anti-money-laundering Directive (AMLD4).

T. Hemmelgarn and G. Nicodème, Tax Co-ordination in Europe: Assessing the First Years of the EU-Savings Taxation

Directive, Taxation Paper No 18, 2009.

T. Rixen and P. Schwarz, How Effective is the European Union's Savings Tax Directive? Evidence from Four EU Member

States, Journal of Common Market Studies, Volume 50 (1), 2012, pp. 151-168.

ibid.

The other three former parts of the Netherlands Antilles (Bonaire, St. Eustatius, and Saba) are now part of the

Netherlands. Source:

https://www.government.nl/topics/caribbean-parts-of-the-kingdom/question- and-

answer/what-are-the-different-parts-of-the-kingdom-of-the-netherlands , accessed on 21 December 2020.

Source: OECD/Global Forum on Transparency and Exchange of Information for Tax Purposes:

http://www.oecd.org/tax/transparency/country-monitoring/

, accessed on 21 December 2020.

Directive (EU) 2015/849 of the European Parliament and of the Council of 20 May 2015 on the prevention of the use of

the financial system for the purposes of money laundering or terrorist financing.

Implementation of the EU requirements for tax information exchange

The fifth AML Directive (AMLD5)

widened the scope for interaction between AML and DAC

provisions through the setting up of interconnected BO registers at the EU level. Thanks to these

registers and the amendments introduced by the DAC5,

national tax administrations will be able

to use the BO information provided by these BO registers. As the AMLD5 had to be transposed by

Member States by 10 January 2020, it is however too early for an assessment of these provisions.

During the research, the following areas with scope for improvement were identified:

the definition of tax crime, as there is no agreed definition of the term at the EU level;

the definition of beneficial ownership, as there are differences between the Member

States' definitions and between the DAC and the AML provisions;

access to and successful exchange of information, depending on the transposition

of the AMLD4 and the AMLD5 into national law by the Member States;

accuracy of information, which differs between Member States and impacts the

effectiveness of the identification of BO under AML provisions. The European

Commission therefore plans to propose the creation of an EU AML supervisor, either

within an existing EU agency or as a new, dedicated body.

A detailed description of each area can be found in Chapter 4.3.1. of the annexed research paper.

6.3.2. Judicial assistance provisions

An increased interaction between the DAC and the use of DAC information in criminal matters could

help improve the efficiency of criminal investigations. Member States can send requests to other

Member States for information on bank accounts or banking transactions by natural or legal persons

who are subject to criminal investigations.

For the time being, the use for non-tax matters is often

not possible as the sending Member State has to give its authorisation of such use. However,

national legislation, which often does not allow for the use of tax-related information for other

purposes, prevents this use.

Directive (EU) 2018/843 of the European Parliament and of the Council of 30 May 2018 on the prevention of the use of

the financial system for the purposes of money laundering or terrorist financing.

Council Directive (EU) 2016/2258 of 6 December 2016 amending Directive 2011/16/EU as regards access to anti-money-

laundering information by tax authorities.

European Commission, Communication from the Commission on an Action Plan for a comprehensive Union policy on

preventing money laundering and terrorist financing, C(2020) 2800 final, 7.5.2020.

Council Act of 16 October 2001 establishing, in accordance with Article 34 of the Treaty on European Union, the Protocol

to the Convention on Mutual Assistance in Criminal Matters between the Member States of the European Union, OJ

C 326, 21.11.2001, p.1-7.

EPRS | European Parliamentary Research Service

7. Efficiency, EU added value and relevance

As explained in the preceding chapters, this European Implementation Assessment (EIA) focuses on

the effectiveness and coherence of the Directive on administrative cooperation. This chapter

provides a brief assessment of other evaluation criteria – efficiency, relevance and EU added value –

as mentioned in the Interinstitutional Agreement between the European Parliament, the Council of

the European Union and the European Commission on Better Law-Making

and in the European

Commission's Better Regulation Guidelines.

7.1. Efficiency

The European Commission evaluation of the DAC, published in 2019,

discussed at length the

efficiency of administrative cooperation under the DAC. Although the evaluation points to concerns

regarding the robustness of its conclusions, the European Commission deems the overall cost-

benefit ratio of the directive to be positive.

Limited data on monetary benefits is the main problem in the evaluation of the efficiency of

administrative cooperation under the DAC. In the context of the European Commission evaluation,

only three Member States (Estonia, Poland, and Slovenia) were able to provide figures on the

additional taxes due to the implementation of the automatic exchange of information under the

DAC1. The total amount of additional taxes assessed in these three Member States increased from

€ 2.9 million in 2016 to € 9.2 million in 2017. Another two Member States (Belgium, Finland)

estimated an increase in the tax base, which could potentially lead to additional tax revenues.

Depending on the share of actual tax revenues, the relatively high amounts of increase in the tax

base stated by Belgium (€ 289.5 million) and Finland (€ 29.0 million) for 2017 could probably lead to

sufficiently high tax revenues to cover the DAC1 compliance costs of the tax authorities in these

countries.

Limited evidence on additional tax revenues is an inherent problem of other forms of administrative

cooperation (mainly exchanges on request, spontaneous exchange, and simultaneous cont r ols) as

well. The Commission evaluation shows an increasing trend (from € 29.3 million in 2014 to

€ 277.1 million in 2017) in additional taxes assessed due to these forms of administrative

cooperation, even though only six Member States provided – in some cases partial – information.

The evidence is not much better when it comes to cost savings for tax authorities, taxpayers and

other reporting bodies. While the standardisation of administrative processes under the DAC has

certainly led to some efficiency gains for tax authorities, it is impossible to quantify it in monetary

terms. Based on data from three Member States (Lithuania, Slovenia, Netherlands), the European

Commission evaluation estimates a yearly burden reduction for taxpayers of € 0.5 million to

1.1 million in these three Member States.

Data availability in terms of cost for the tax authorities is also far from optimal. The Commission

evaluation indicates extrapolated compliance costs up to 2017 for the automatic exchange of

information under the DAC1-DAC3 of € 130 million for all then 28 Member States. Costs due to other

Official Journal of the European Union, Interinstitutional Agreement between the European Parliament, the Council of

the European Union and the European Commission on Better Law-Making, OJ L 123, 12.5.2016, p. 1–14.

European Commission, Better Regulation Guidelines, SWD(2017) 350, 2017.

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

Implementation of the EU requirements for tax information exchange

forms of information exchange, including one-off costs in 2011-2013 for procedures and tools for

exchange on request could not be quantified.

To sum up, total regulatory costs for automatic exchange under the DAC1-3 for Member States and

financial institutions are estimated between € 200 million and 260 million for 2015-2017. According

to European Commission estimates, additional taxes assessed in only six Member States der iving

from the DAC1 amount to € 624 million up to 2017. On the basis of the relatively weak empirical

evidence, the Commission evaluation reached the conclusion that the directive is cost-effective.

7.2. EU added value

The EU added value of the provisions under the DAC1-DAC3 and to some extent under the DAC4,

was assessed in the European Commission evaluation of 2019. First, the mandatory nature of DAC

provisions is a major difference to the corresponding OECD standards. Second, it seems clear that a

common EU framework such as the DAC ensures that tax information exchange between Member

States is carried out in a harmonised manner. The specific findings of the EU added value are

summarised in the table below.

Table 4: Key findings on the EU added value of the DAC

Activities Key EU added value Because of...

Exchange on requests under

DAC2, DAC3, DAC4

Compulsory legal framework for

all Member States

International standards made

mandatory for all Member States

Automatic exchange of

information via DAC1

Compulsory legal framework

No corresponding exchange of

information under international

agreements

Exchange of information

Common secure network to

exchange information (CCN)

Standard forms and

computerised formats

Economies of scale

Lower transaction costs

Raised awareness

Presence abroad and

simultaneous controls

EU budget support to take part

in this activity

EU Fiscalis programme

supporting cooperation between

tax authorities

Feedback

Compulsory for AEOI, to improve

the quality of exchanges

No corresponding requirements

under international agreements

Sharing of best practices DAC-related Fiscalis activities

Directive building trust among

tax authorities

Source: 2019 European Commission evaluation, SWD(2019) 328 final

7.3. Relevance

When looking at tax avoidance and evasion as major problems in the field of taxation, the European

Commission evaluation concludes that the DAC is addressing this problem and is therefore a

relevant intervention. In addition, the directive has evolved to address problems that have been

identified over the past years.

Concerning the provisions on automatic exchange of information under the DAC1, the European

Commission evaluation highlights that the exchanges reflect the intra-EU migration and investment

patterns. Through the information exchanged between Member States, the directive ensures that

tax authorities get a more complete picture of different forms of income per taxpayer. The same

EPRS | European Parliamentary Research Service

reasoning can also be identified in the context of DAC2 exchanges. Within the internal market it is

therefore relevant to make sure that tax authorities receive information on financial assets held

abroad.

DAC3 and DAC4 clearly tackle the issue of tax avoidance and try to reduce opportunities for

aggressive tax planning, and therefore address – as mentioned above – a major problem in the field

of taxation. At the same time, the further development of the directive (DAC5 and DAC6) has shown

that the directive continues to address arising issues and is therefore relevant.

Implementation of the EU requirements for tax information exchange

8. Conclusions

This EIA analyses the implementation of DAC1-4 provisions, focusing predominantly on the

effectiveness and coherence criteria. The directive has been amended several times over the past

years in order to integrate new fields of tax information exchange and to fulfil its general objectives,

i.e. the efficient administrative cooperation between Member States to overcome the negative

effects of the increasing globalisation on the internal market and the safeguarding of Member

States' tax revenues, by introducing common rules, obligations and rights.

The first part of the EIA showed that problems with the transposition into national legislation were

present at least with regard to the initial directive and its first amendment. However, these problems

have been solved to a large extent, open infringement cases are mainly related to the DAC6 and are

still at an early stage of letters of formal notice.

The main problem for the research work consisted in the limited availability of data on tax

information exchanges in the EU. Due to the lack of access to quantitative data, the focus of the

research for this EIA shifted to more qualitative information in the form of interviews of tax

administrations and a stakeholder survey.

8.1. Main findings on effectiveness

As already mentioned in the 2019 Commission evaluation and in the preceding chapters of the

present study, the limited nature of the available evidence poses a problem for the assessment of

the effectiveness of the directive as such. This problem of limited evidence of the directive's

effectiveness might be exacerbated by the fact that the DAC1 had not been accompanied by an

impact assessment, which could have provided a baseline or points of comparison.

However, the

annexed research paper's Chapter 3.2 contains a list of academic literature suggesting that

international cooperation in the field of tax information exchange (such as the FATCA and the CRS)

leads to a reduction in tax evasion, tax avoidance and tax fraud.

By using data from the International Monetary Fund,

the research paper does not es tablish any

notable effect of the entry into application of the DAC on the evolution of foreign portfolio holdings

for the EU as a whole or at Member State level. The research paper reaches the same conclusion

when looking into data on deposits for Luxembourg, the Member State with the largest share of

non-monetary financial institutions' deposits that originate from other EU Member States (20.4 %).

The introduction of the DAC2 did not have any notable impact on the volume of these deposits.

Nevertheless, the evidence gathered for the research paper shows that the directive has achieved

its specific objectives to some extent. The DAC seems to have proven its usefulness to fight cross-

border tax fraud, tax evasion and tax avoidance, as all interviewed Member States reported using

information received under DAC provisions to address these issues. Even if this was not quantifiable,

Member States also reported that the DAC has had some deterrent effect and has led to enhanced

ex-ante compliance. It is very difficult to establish whether the DAC has also led to greater

transparency in tax rules. Although the European Commission published some statistics in the 2019

evaluation, most of the data presented refer to 2017 as the most recent available year. The

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019.

International Monetary Fund, Coordinated Portfolio Investment Survey.

EPRS | European Parliamentary Research Service

assessment of the effectiveness of DAC provisions made in Chapter 5 of this study translates also

into recommendations on further improvements, as described below.

8.2. Main findings on coherence

The 2019 Commission evaluation analysed the internal and external coherence of the DAC and did

not find any major problems in either area. The present EIA focuses on the DAC's external coherence

with other EU legislation and in the international context.

With respect to the information exchange within the OECD framework and the DAC, only minor

differences have been identified. Differences are more significant when comparing the standards

under the DAC2 and the US FATCA, as the DAC2 requirements are broader than the ones under the

FATCA. The fact that the FATCA applies only to US persons and US residents holding financial assets

outside the US above certain thresholds is not comparable to the provisions under the DAC2. DAC2

provisions, focusing on all residents of a Member State (from a tax perspective) that hold financial

assets in other Member States, have to be seen in an internal market context.

Despite the improvements over recent years, the interaction between the DAC and AML legislation

is however somewhat hampered by differences between the Member States as regards certain

definitions (tax crime, beneficial ownership) or between the two legislations. In addition, potential

synergies between the two areas depend on the quality of national transposition and quality of

information provided under AML legislation. An assessment of the effects of the beneficial

ownership registers under the fifth AML Directive is not yet possible due to its recent entry into

force. Finally, the use of DAC information in criminal matters is often not possible as the sending

Member State has to give its authorisation of such use.

8.3. Recommendations

Despite the fact that EU institutions, stakeholders and Member States agree on the usefulness of

administrative cooperation in the field of taxation, they see several areas for potential

improvements to the DAC. Drawing on the analysis made in the annexed research paper, this study

makes recommendations for potential changes to the directive in a number of areas. A full list of the

recommendations (called 'suggestions' in the annexed research paper) can be found in the

respective chapter of the annexed research paper.

Areas of recommendations:

increasing the availability of taxpayer identification numbers;

enhancing the due diligence requirements for DAC2 provisions;

reinforcing and broadening the reporting requirements;

increasing the coverage of income forms, assets, accounts, etc.;

developing a common monitoring and evaluation approach;

simplifying the exchange of information for other purposes and compliance

requirements;

developing guidelines for the standard of foreseeable relevance.

Some of these recommendation will be addressed by the ongoing DAC7 amendment and the

upcoming DAC8, already announced by the European Commission. In any case, the feasibility as

well as benefits of these potential changes to the directive have to be evaluated in the light of

additional costs for tax payers and tax authorities.

Implementation of the EU requirements for tax information exchange

MAIN REFERENCES

Deutscher Bundestag, Schriftliche Fragen mit den in der Woche vom 6. Juli 2020 eingegangenen

Antworten der Bundesregierung, Drucksache 19/20953

, 6.7.2020

Economisti Associati srl, Evaluation of Administrative Cooperation in Direct Taxation, 2019

European Commission, Implementation of Directive 2014/107/EU, WPIV Meeting on 19 June 2015, 2015

European Commission, Better Regulation Guidelines, SWD(2017) 350, 2017

European Commission, Commission Staff Working Document on the application of Council Directive (EU)

no 2011/16/EU on administrative cooperation in the field of direct taxation, SWD(2017) 462 final,

18.12.2017

European Commission, Report from the Commission to the European Parliament and the Council on the

application of Council Directive (EU) 2011/16/EU on administrative cooperation in the field of direct

taxation, COM(2017) 781 final, 18.12.2017

European Commission, Report from the Commission to the European Parliament and the Council on

overview and assessment of the statistics and information on the automatic exchanges in the field of

direct taxation, COM(2018) 844 final, 17.12.2018

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in

the field of taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, 2019

European Commission, Communication from the Commission of 7 May 2020 on an Action Plan for a

comprehensive Union policy on preventing money laundering and terrorist financing, C(2020) 2800 final,

7.5.2020

European Commission, Communication from the Commission to the European Parliament and the

Council of 15 July 2020 on an Action Plan for fair and simple taxation supporting the Recovery Strategy,

COM(2020) 312 final

European Economic and Social Committee, Opinion on effective and coordinated EU measures to

combat tax fraud, tax avoidance, money laundering and tax havens, ECO/510, 18 September 2020.

Hemmelgarn T. and Nicodème G., Tax Co-ordination in Europe: Assessing the First Years of the EU-

Savings Taxation Directive, Taxation Paper No 18, 2009

Ioannides I. and Tymowski J., Tax evasion, money laundering and tax transparency in the EU Overseas

Countries and Territories, EPRS, European Parliament, 2017

M. and Ligthart J.E., Information Sharing and International Taxation: A Primer, International Tax and

Public Finance, 13, pp. 81–110, 2006

Kramer E. and Eisele K., Better cooperation against tax fraud and evasion

, Briefing, European

Parliamentary Research Service, Brussels, November 2020

OECD, Implementing the Tax Transparency Standards: A Handbook for Assessors and Jurisdictions, OECD

Publishing, Paris, 2010.

OECD, The 2019 AEOI Implementation Report, OECD, Paris, 2019

OECD, Global Forum on Transparency and Exchange of Information for Tax Purposes, Peer Reviews:

Poland 2015: Phase 2: Implementation of the Standard in Practice, OECD Publishing, 2015

OECD, Global Forum on Transparency and Exchange of Information for Tax Purposes: Malta 2020 (Second

Round): Peer Review Report on the Exchange of Information on Request, Global Forum on Transparency

and Exchange of Information for Tax Purposes, OECD Publishing, 2020

Official Journal of the European Union, Interinstitutional Agreement between the European Parliament,

the Council of the European Union and the European Commission on Better Law-Making, OJ L 123,

12.5.2016, pp. 1–14

EPRS | European Parliamentary Research Service

Princen S. et al., Taxation of Household Capital in EU Member States: Impact on Economic Efficiency,

Revenue & Redistribution, European Commission Directorate-General for Economic and Financial Affairs

Discussion Paper 130, August 2020

Rixen T. and Schwarz P., How Effective is the European Union's Savings Tax Directive? Evidence from Four

EU Member States, Journal of Common Market Studies, Volume 50 (1), 2012, pp. 151-168

Torslov T., Wier L., Zucman G., The Missing Profits of Nations, World Economy Database, Working Paper

2020/12, April 2020

LEGISLATIVE INSTRUMENTS

Directive 2011/16/EU and amendments:

Council Directive 2011/16/EU

of 15 February 2011 on administrative cooperation in the field of taxation

and repealing Directive 77/799/EEC; DAC1

Council Directive 2014/107/EU of 9 December 2014 amending Directive 2011/16/EU as regards

mandatory automatic exchange of information in the field of taxation; DAC2

Council Directive (EU) 2015/2376 of 8 December 2015 amending Directive 2011/16/EU as regards

mandatory automatic exchange of information in the field of taxation; DAC3

Council Directive (EU) 2016/881 of 25 May 2016 amending Directive 2011/16/EU as regards mandatory

automatic exchange of information in the field of taxation; DAC4

Council Directive (EU) 2016/2258 of 6 December 2016 amending Directive 2011/16/EU as regards access

to anti-money-laundering information by tax authorities; DAC5

Council Directive (EU) 2018/822 of 25 May 2018 amending Directive 2011/16/EU as regards mandatory

automatic exchange of information in the field of taxation in relation to reportable cross-borde r

arrangements; DAC6

European Commission proposal for DAC7 (not yet in force):

Proposal for a Council Directive amending Directive 2011/16/EU on administrative cooperation in the

field of taxation, COM(2020) 314 final, 15.7.2020, DAC7

Amendment on deferral:

Council Directive (EU) 2020/876 of 24 June 2020 amending Directive 2011/16/EU to address the urgent

need to defer certain time limits for the filing and exchange of information in the field of taxation because

of the COVID-19 pandemic

European Commission proposal for a codification of DAC1-DAC6 (not in force):

Proposal for a Council Directive on administrative cooperation in the field of taxation (codification),

COM(2020) 49 final, 12.2.2020

Implementing regulations in force:

Commission Implementing Regulation (EU) 2015/2378

of 15 December 2015 laying down detailed rules

for implementing certain provisions of Council Directive 2011/16/EU on administrative cooperation in

the field of taxation and repealing Implementing Regulation (EU) No 1156/2012

Commission Implementing Regulation (EU) 2016/1963 of 9 November 2016 amending Implementing

Regulation (EU) 2015/2378 as regards standard forms and linguistic arrangements to be used in relation

to Council Directives (EU) 2015/2376 and (EU) 2016/881

Implementation of the EU requirements for tax information exchange

Commission Implementing Regulation (EU) 2018/99 of 22 January 2018 amending Implementing

Regulation (EU) 2015/2378 as regards the form and conditions of communication for the yearly

assessment of the effectiveness of the automatic exchange of information and the list of statistical data

to be provided by Member States for the purposes of evaluating of Council Directive 2011/16/EU

Commission Implementing Regulation (EU) 2019/532 of 28 March 2019 amending Implementing

Regulation (EU) 2015/2378 as regards the standard forms, including linguistic arrangements, for the

mandatory automatic exchange of information on reportable cross-border arrangements

Implementing regulations no longer in force:

Commission Implementing Regulation (EU) 1156/2012

of 6 December 2012 laying down detailed rules

for implementing certain provisions of Council Directive 2011/16/EU on administrative cooperation in

the field of taxation; repealed by Commission Implementing Regulation (EU) 2015/2378

Commission Implementing Regulation (EU) 1353/2014 of 15 December 2014 amending Implementing

Regulation (EU) No 1156/2012 laying down detailed rules for implementing certain provisions of Council

Directive 2011/16/EU on administrative cooperation in the field of taxation; date of end of validity:

31/12/2015, implicitly repealed by Commission Implementing Regulation (EU) 2015/2378

Savings Tax Directive (no longer in force):

Council Directive 2003/48/EC

of 3 June 2003 on taxation of savings income in the form of interest

payments

Mutual Assistance Directive (no longer in force):

Council Directive 77/799/EEC

of 19 December 1977 concerning mutual assistance by the competent

authorities of the Member States in the field of direct taxation

EPRS | European Parliamentary Research Service

Implementation of EU

requirements for tax

information exchange:

Progress, lessons

learnt and obstacles

to overcome

(Directive 2011/16/EU and its amendments)

Research paper

This research paper has been prepared in support of a European

Implementation Assessment assessing the implementation of tax

information exchange requirements under Council Directive 2011/16/EU

(DAC1) and its first three amendments (DAC2 to DAC4). The paper firstly

focuses on the effectiveness of DAC by analysing the market reactions to

DAC, identifying obstacles to the exchange of information between tax

authorities and examining the use of such information and DAC’s potential

fiscal benefits. Next, the paper analyses the pan-European volume of

information being exchanged under DAC and the evolution of within-EU

cross-border holdings of financial assets by EU residents. Finally, the paper

compares the main provisions of the OECD Common Reporting Standard

(CRS), which DAC2 incorporated into EU law, to the US Foreign Account Tax

Compliance Act (FATCA), and assesses the coherence of DAC with other EU

legislation such as the Anti-Money-Laundering Directive.

AUTHOR(S)

This study has been written by Patrice Muller, Nicholas Robin and James Forrester of LE Europe Limited with

research support on the literature review and stakeholder interviews from Vanessa Vázquez Felpeto, Vasilis

Douzenis and Khush Patel of PricewaterhouseCoopers LLP at the request of the Ex-Post Evaluation Unit of the

Directorate for Impact Assessment and European Added Value, within the Directorate-General for

Parliamentary Research Services (EPRS) of the Secretariat of the European Parliament.

This document has been prepared for the European Parliament on the basis of existing literature, interviews

with Member States and stakeholder consultation. It does not reflect the opinions of the authors, and the

authors cannot be held responsible for any use which may be made of the information contained therein.

ADMINISTRATOR RESPONSIBLE

Eckhard Binder, Ex-Post Evaluation Unit

To contact the publisher, please e-mail EPRS-ExPostEvaluation@ep.europa.eu

LINGUISTIC VERSIONS

Original: EN

Manuscript completed in February 2021.

DISCLAIMER AND COPYRIGHT

This document is prepared for, and addressed to, the Members and staff of the European Parliament as

background material to assist them in their parliamentary work. The content of the document is the sole

responsibility of its author(s) and any opinions expressed herein should not be taken to represent an official

position of the Parliament.

Reproduction and translation for non-commercial purposes are authorised, provided the source is

acknowledged and the European Parliament is given prior notice and sent a copy.

Brussels © European Union, 2021.

PE 662.603

ISBN: 978-92-846-7700-9

DOI: 10.2861/433189

CAT: QA-03-21-028-EN-N

eprs@ep.europa.eu

http://www.eprs.ep.parl.union.eu (intranet)

http://www.europarl.europa.eu/thinktank (internet)

http://epthinktank.eu (blog)

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Executive summary

The study

Council Directive 2011/16/EU, Directive on Administrative Cooperation (DAC or DAC1) and its

subsequent amendments aim to combat tax fraud and tax evasion by facilitating the exchange of

information between the tax authorities (TAs) of EU Member States. This Paper assesses the

effectiveness of DAC and its first three amendments (DAC2 to DAC4)

, analyses the evolution of tax

information exchanges and discusses the coherence of DAC with other international legislation on

automatic exchange of information (Foreign Account Tax Compliance Act (FATCA) and the OECD

Common Reporting Standard (CRS) and with other EU legislation, in particular the Anti-Money-

Laundering (AML) legislation.

DAC1 requires Member States to automatically exchange information of income from employment,

directors’ fees, life insurance products not covered by other Union legal instruments on exchange

of information and similar measures, pensions and the ownership and income from immovable

properties. The scope of automatic exchange of information was extended under DAC2 to financial

assets and income accruing to such assets, and under DAC3 to advance rulings by TAs on tax

arrangements and pricing arrangements. The ambit of the automatic exchange was further

broadened by DAC4 which provided for the preparation and exchange of country-by-country

reports by multinationals.

Effectiveness of DAC

In order to assess the effectiveness of DAC1-4, the research focused on the following three key pillars

to identify and assess the issues that have arisen:

1 Pillar 1: Understanding market reactions to DAC

2 Pillar 2: Obstacles in the exchange of information

3 Pillar 3: Information exchange, usage and fiscal impacts

Under Pillar 1, the research highlighted how DAC has evolved over time, and its scope has expanded

with subsequent amendments to cover new and emerging types of income, by broadening the

definition of income in the Directive. However, DAC has not always kept pace with the global

economic environment, resulting in two categories of issues which impact the Directive’s

effectiveness. The first, around income not covered at all within the provisions, for example

cryptocurrencies, and the second, around the exploitation of circumventions and loopholes in the

provisions in areas such as:

Reporting;

Tax residence;

Beneficial ownership;

Due diligence procedures and correctness of information; and

Tax evasion and tax avoidance schemes – e.g. cum-ex, cum-cum, bilateral agreements etc.

Pillar 2 analysis identified a wide range of legal and practical obstacles that have reduced the

effectiveness of DAC1-4. These have had a range of impacts with certain obstacles affecting multiple

1 DAC2: Council Directive 2014/107/EU of 9 December 2014 amending Directive 2011/16/EU as regards mandatory

automatic exchange of information in the field of taxation; DAC3: Council Directive (EU) 2015/2376 of 8 December

2015 amending Directive 2011/16/EU as regards mandatory automatic exchange of information in the field of

taxation; and DAC4: Council Directive (EU) 2016/881 of 25 May 2016 amending Directive 2011/16/EU as regards

mandatory automatic exchange of information in the field of taxation.

EPRS | European Parliamentary Research Service

DAC amendments or even the Directive as a whole, including compliance costs for DAC1 to DAC4.

When focusing on legal obstacles it becomes clear that many of the issues identified arise through

conflicting obligations from other directives or legislation, while practical issues are impeding the

exchange or the processing of information itself.

Pillar 3 established the extent to which the Directive was being used and whether there had been

any discernible fiscal impacts. In the absence of the latest data, the research made use of previous

work and sought qualitative evidence. Member States reported that DAC did have a deterrence

effect with an increased level of ex-ante compliance perceived.

Evolution of tax information exchanges

Unfortunately, publicly available data on the volume of exchanges of information (automatic and

other) and the value covered by such exchanges is very limited, especially regarding bilateral flows

of information between the TAs of Member States. However, the limited information that is available

shows that the volume of information exchanges has increased markedly, and the bulk of the

exchanges occurs mainly between a limited number of Member States, often, but not always,

between neighbouring Member States.

The limited academic literature on the impact of the various international mechanisms to fight tax

avoidance and evasion related to cross-border financial investments (OECD Common Reporting

Standard (CRS), FATCA and DAC2) finds that these arrangements have reduced to some extent

investments into financial off-shore centres and so-called tax havens. In contrast, a review of various

data from the Bank for International Settlements (BIS), the European Central Bank (ECB) and the

International Monetary Fund (IMF) on within EU cross-border financial asset holdings does not show

any impact of DAC2 on the level of such holdings.

Coherence

The OECD CRS, which is incorporated into EU law by DAC2, builds on the approach taken by FATCA

but is not exactly identical. The scope of the CRS is much broader in the sense that it aims to cover

cross-border financial asset holdings by taxpayers of as many countries as possible and not just US

citizens and residents. Moreover, under the CRS, controlling persons of a non-financial entity (NFE)

are reportable irrespective of whether they are resident in the same jurisdiction as the NFE while, in

the case of FATCA, only US controlling persons of a passive NFE are reportable. Reportable assets

under FATCA include investments in real property. Such investments are not covered by the CRS

(but are covered by DAC1). In addition, FATCA specifies thresholds above which accounts are

reportable while no such thresholds exist in the CRS.

Although significant steps have been made to enhance cooperation and integration between the

provisions of the DAC and anti-money laundering (AML) provisions, some challenges still remain as

the 5th Anti-Money Laundering Directive (AMLD5) only entered into force from January 2020. In

particular, challenges surrounding the definition of beneficial ownership (BO) have been raised and

both national and international fora. Currently, the definition of beneficial ownership is within the

discretion of Member States although there is a significant alignment with the recommendations

set out by the Financial Action Task Force (FATF), in particular within the AML provisions. With

regards to the interaction between DAC and AML provisions, beneficial ownership (BO) challenges

primarily concerns DAC5 which is out of scope for this Paper. In addition, there is also a challenge

surrounding the supervisory element of the Directive, which ensures that the information held is

accurate and complete so that any exchange of information is relevant, useful, and adds value to

the Member States receiving the information.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

The 5th anti-money laundering Directive (AMLD5)

, whose rules apply since January 2020, aims to

make the structures of economic entities more transparent by requiring corporate and other legal

entities incorporated within Member States to hold up-to-date information on their beneficial

ownership and to make such publicly available. Moreover, the fifth amendment of DAC (i.e. DAC5)

gives TAs access to beneficial ownership information collected under the AML legislation. However,

DAC5 is outside the scope of the present study.

The interactions between DAC and a) the Protocol established by the Council in accordance with

Article 34 of the Treaty on European Union to the Convention on Mutual Assistance in Criminal

Matters between the Member States of the European Union and b) the Regulation (EU) 2018/1805

of the European Parliament and of the Council of 14 November 2018 on the mutual recognition of

freezing orders and confiscation orders were examined and no inconsistencies between DAC and

the judicial assistance were identified.

However, it was noted that the process of finding out whether a person under criminal investigation

has bank accounts in one or several Member States is facilitated if financial account information

received by TAs can be used in criminal investigations. At the present time, the use for non-tax

matters of the information having been exchanged between TAs requires the authorisation of the

Member State(s) having sent the information. Such authorisation is not always granted, often

because the national laws of the sending Member State(s) do not permit the use of tax related

information in non-tax matters.

Suggestions

The research has identified six areas which may improve the effectiveness of DAC. These areas are

presented along with associated, specific suggestions in the table below. Whilst potential benefits

may be derived from undertaking changes to the provisions, these should be balanced against other

considerations such as an increase in compliance costs, given the compliance burden already

present for taxpayers, obliged entities, and tax administrations. A full cost-benefit analysis of each

suggestion would therefore be needed to thoroughly assess their effectiveness, which is out of

scope for this paper.

Table 1: Areas where DAC effectiveness can be improved and associated suggestions

(please see footnotes for original source of suggestion)

# Area Suggestion Basis

1 Availability of TIN data

1.1: Introduce mandatory TIN

requirements to improve data

matching and identification

across Member States. This is

currently considered in DAC7

proposals for exchanges related

to DAC1 and DAC2 requirements.

Ongoing work for TIN verification,

such as 'TIN on Europa' or 'TIN on

Based on literature

review, further details

available in section 2.2.2

2 Directive (EU) 2018/843 of the European Parliament and of the Council of 30 May 2018 amending Directive (EU) 2015/849

on the prevention of the use of the financial system for the purposes of money laundering or terrorist financing, and

amending Directives 2009/138/EC and 2013/36/EU.

3 Council Directive (EU) 2016/2258 of 6 December 2016 amending Directive 2011/16/EU as regards access to anti-money-

laundering information by tax authorities.

EPRS | European Parliamentary Research Service

# Area Suggestion Basis

the Web', should be leveraged, to

the extent possible.

Due diligence related to DAC2

requirements

2.1: Enhanced due diligence

requirements for tax residence

and beneficial ownership for all

accounts, including those where

account holders state residence

in countries with golden visa

schemes.

2.2: Introduce regular checks to

ensure information held is

accurate instead of relying on

checks only when there is a

change in circumstances.

2.3: Where a Member State is not

compliant with FATF

recommendations, FIs should not

be able to solely rely on AML

information and should instead

be required to conduct their own

independent checks.

Based on the literature

review, further details

available in section 2.1.3.

3 Reporting requirements

3.1: Under DAC1: Remove the

availability criterion and

introduction of reporting

requirement for 5 out of 5

categories.

3.2: Under DAC2: Where there is

no information to report by a

reporting FI it should then file nil

returns.

3.3: Under DAC2: Remove the

thresholds on the value of

accounts that are required to

report information.

3.4: Under DAC2: Include a

marker to signal joint ownership

of different account holder to

avoid duplicate reporting and to

Reporting Requirement

suggestion 3.1 is based

on the literature review,

further details available in

section 2.1.2.

Reporting Requirement

suggestions 3.2, 3.3, 3.7

and 3.9 are based on the

literature review and

cross-analysis of DAC and

AML legislation, further

details available in section

2.1.3.

Reporting Requirement

suggestions 3.4 to 3.6 are

based on the Member

State interviews, further

4 See European Commission, Directorate-General Taxation and Customs Union, Evaluation of Administrative Cooperation

in Direct Taxation, Final Report prepared by Economisti Associati in collaboration with ECOPA and Oxford Consulting,

April 2019.

5 See Andres Knobel, Reporting taxation: Analysing loopholes in the EU’s automatic exchange of information and how to

close them, A Report Commissioned by the Greens/EFA Group in the European Parliament, October 2018.

6 As above.

7 As above.

8 See Economisti Associati 2019.

9 See Andres Knobel, 2018.

10 As above.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

# Area Suggestion Basis

facilitate accurate identification

of account balances.

3.5: Under DAC2: In addition to

having a flag for joint accounts,

entities could record the

ownership share of each account

holder and flag when an account

is held by owners from different

jurisdictions.

3.5: Under DAC3: The structure

and presentation of information

provided by Member States

should be standardised.

3.6: Under DAC4: Extend the

scope of information provided by

MNE that own several entities

within the same jurisdiction

beyond providing only

aggregate-level information.

3.7: DAC can improve

transparency beyond the

information that is made

available to TA, by publishing

more information on statistics,

non-compliance, rulings and

corporate disclosures.

3.8: Requirement to establish

ownership for active entities as

well as passive NFEs (e.g. by

applying the look through

principle) at both the entity level

and the Beneficial Owner level.

3.9: Lower the 25 % ownership

threshold for classifying

Beneficial Owners.

details available in section

2.1.3.

4 Coverage

4.1: Under DAC2: When assets are

sold, the sales and purchase price

should be reported, to the extent

possible. At the present time it is

not possible to determine

whether the seller made any

All Coverage suggestions

are based on the

literature review and

analysis of DAC

Based on responses during the Member State interviews.

As above.

See Andres Knobel, 2018.

As above.

EPRS | European Parliamentary Research Service

# Area Suggestion Basis

capital gains (which would be

taxable in some jurisdictions).

4.2: Under DAC2: Include FIs that

are currently excluded within

scope, e.g. electronic money

institutions.

4.3: Under DAC2: Include

cryptocurrencies and e-accounts

under scope - currently in the

DAC 8 proposals.

4.4: Under DAC2: Commodities

including precious metals should

be included as financial assets if

held through an account at a

broker or trading platform.

4.5: Under DAC2: Accounts at

larger peer-to-peer lending

platforms, crowdfunding

platforms, and similar, etc. should

be covered by DAC2.

4.6: Under DAC3: Include non-

cross-border rulings and APAs

that are likely to be of interest to

other Member States.

4.7: Under DAC3: Include natural

persons within scope.

4.8: Under DAC3: Include ruling

and APAs that are not

advanced.

4.9: Under DAC3: Include rulings

or APAs issued, amended or

renewed before 2012 but still

valid.

legislation, further details

available in section 2.1.3.

5 Monitoring and evaluation

5.1: Develop a common or

recommended approach for

assessing/evaluating the impact

of the different DACs.

Based on multiple

responses to the Member

State interviews.

As above.

See Andres Knobel, 2018.

As above.

Based on responses during the Member State interviews.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

# Area Suggestion Basis

6 Simplicity and certainty

6.1: Standardise the approach to

the foreseeable relevance

standard by introducing

guidelines that can facilitate

Member States to exchange

information.

6.2: Simplify the exchange of

information for other purposes,

like AML, by setting timelines

around providing authorization

to share this type of

information.

6.3 Simplify the compliance

requirements for individual

taxpayers and companies/MNEs,

in order to increase compliance

and manage the compliance

burden.

All Simplicity and

certainty suggestions are

based on the Member

State interviews and the

stakeholders survey, as

well as cross-analysis of

the DAC legislation, see

section 2.2.1 for further

details.

As above.

Based on responses to the stakeholder survey.

EPRS | European Parliamentary Research Service

List of abbreviations

ACBR

Advance Cross-Border Rulings

AEFI

Expert Group on Automatic Exchange of Financial Account Information

AEOI

Automatic Exchange of Information

AML

Anti-money Laundering

AML/CTF

Anti-money Laundering/Counter-Terrorist Financing

AMLD5

5th Anti-Money Laundering Directive

APA

Advance Pricing Agreement

BEPS

Base Erosion and Profit Shifting

BIS

Bank for International Settlement

Beneficial Ownership

BoP

Balance of Payments

Competent Authority

CbCR

Country-by-Country Reporting

CBTR

Cross-border Tax Ruling

CPIS

Coordinated Portfolio Investment Survey

CRS

Common Reporting Standard

CRS MCAA

CRS Multilateral Competent Authority Agreement

DAC

Directive on Administrative Cooperation

Directors’ Fees

DG EPRS

Directorate-General European Parliamentary Research Service

DG TAXUD

Directorate-General Taxation and Customs Union

European Commission

ECB

European Central Bank

Employment Income

EOI

Exchange of Information

EOIR

Exchange of Information on Request

European Parliament

European Union

EU - 27

27 Member States of the European Union as of 1

January 2020

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

FATCA

Foreign Account Tax Compliance Act

FATF

Financial Action Task Force

FIs

Financial Institutions

HIRE

Hiring Incentives to Restore Employment Act

HNWI

High Net Worth Individuals

IMF

International Monetary Fund

IGA

Intergovernmental Agreement

Immovable Property

IRS

Internal Revenue Service

Information Technology

KYC

Know Your Customer

LIP

Life Insurance Products

MEPs

Members of European Parliament

MNE

Multi-National Enterprise

MoF

Ministry of Finance

NFE

Non-financial entity

OECD

Organization for Economic Co-operation and Development

PEN

Pensions

QFD

Questionnaire on the Functioning of the Directive

SEOI

Spontaneous Exchange of Information

SME

Small and Medium-sized Enterprises

STD

Savings Tax Directive

Tax Administration

TIEA

Tax Information Exchange Agreement

TOR

Terms of Reference

TIN

Tax Identification Number

United States

VAT

Value Added Tax

WHT

Withholding Tax

EPRS | European Parliamentary Research Service

Table of contents

Executive summary ____________________________________________________________ 41

The study __________________________________________________________________ 41

Effectiveness of DAC _________________________________________________________ 41

Evolution of tax information exchanges__________________________________________ 42

Coherence _________________________________________________________________ 42

Suggestions ________________________________________________________________ 43

List of abbreviations ___________________________________________________________ 48

1. Introduction________________________________________________________________ 55

1.1. Introduction to the Directive ________________________________________________ 55

1.1.1. Context of the Directive on Administrative Cooperation ________________________ 55

1.1.2. DAC1 and its subsequent amendments _____________________________________ 55

1.2. Objectives of the Research Paper _____________________________________________ 60

1.3. Structure of the Research Paper ______________________________________________ 60

1.4. Analytical approach and methodology ________________________________________ 61

1.4.1. Q1. Effectiveness of DAC _________________________________________________ 61

1.4.2. Q2. Analysis of the evolution of tax information exchanges _____________________ 62

1.4.3. Q3. Coherence _________________________________________________________ 63

2. Effectiveness of DAC _________________________________________________________ 64

2.1. Pillar 1: Understanding market reactions to DAC_________________________________ 64

2.1.1. Evolution of the different features of DAC ___________________________________ 64

2.1.2. Evolution of income_____________________________________________________ 66

2.1.3. Circumventions and loopholes ____________________________________________ 70

2.2. Pillar 2: Obstacles in the exchange of information________________________________ 80

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

2.2.1. Legal obstacles_________________________________________________________ 80

2.2.2. Practical obstacles ______________________________________________________ 83

2.3. Pillar 3: Information exchange, usage and fiscal impacts __________________________ 86

2.3.1. Using macroeconomic data to monitor the plausibility of DAC2 information exchanges87

3. Quantitative analysis of the DAC information _____________________________________ 91

3.1. Evolution of the information exchanges under DAC ______________________________ 91

3.1.1. Key findings from the Economisti Associati study _____________________________ 91

3.1.2. Findings from selected Member States ______________________________________ 94

3.2. Impact of exchange of tax related information _________________________________ 104

3.2.1. The IMF Coordinated Portfolio Investment Survey (CPIS) and DAC2 ______________ 105

3.2.2. The deposits of monetary financial institutions in the euro-zone and DAC2 ________ 106

3.2.3. The BIS locational banking statistics and DAC2 ______________________________ 107

4. Coherence with other provisions ______________________________________________ 108

4.1. FATCA and CRS __________________________________________________________ 108

4.1.1. FATCA_______________________________________________________________ 108

4.1.2. CRS _________________________________________________________________ 109

4.1.3. Comparison of the provisions of FATCA and the CRS __________________________ 109

4.1.4. DAC2 and the CRS _____________________________________________________ 110

4.2. EU Savings Tax Directive___________________________________________________ 111

4.3. Other EU provisions ______________________________________________________ 113

4.3.1. AML provisions________________________________________________________ 113

4.3.2. Judicial assistance provisions ____________________________________________ 117

5. Conclusions _______________________________________________________________ 119

5.1. General conclusions on the implementation of DAC ____________________________ 119

EPRS | European Parliamentary Research Service

5.1.1. Effectiveness of DAC ___________________________________________________ 119

5.1.2. Evolution of information exchanges _______________________________________ 121

5.1.3. Coherence ___________________________________________________________ 121

5.2. Suggestions ____________________________________________________________ 122

5.2.1. Costs of compliance and other considerations _______________________________ 122

6. References ________________________________________________________________ 126

Appendix 1: Analytical approach and methodology _________________________________ 129

Appendix 2: Comparison of key features of FATCA and CRS ___________________________ 137

Appendix 3: DAC2 and the CRS__________________________________________________ 153

Appendix 4: Stakeholder interview guides and survey questionnaire____________________ 166

Appendix 4.1: Interviews of tax officials - short interview guide _______________________ 166

Appendix 4.2 Interviews of tax officials - long interview guide ________________________ 168

Appendix 4.3: Questionnaire used by online survey ________________________________ 172

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Table of figures

Figure 1: Effectiveness of the DAC and Golden Passports / Residency Visas ________________ 72

Figure 2: Case study on cum-ex/cum-cum schemes __________________________________ 75

Figure 3: Number of Member States with information available in 2017 by DAC1 category ___ 92

Figure 4: Germany: number of DAC1 records sent to other EU Member States and received from

other EU Member States ________________________________________________________ 96

Figure 5: Germany: number of DAC2 records sent to and received from other EU Member States97

Figure 6: Germany: Share of the total value of the financial accounts records received by Germany

from EU Member States under DAC2 AEOI in 2018- Member States which account for 1 % of the total

value _______________________________________________________________________ 99

Figure 7: Germany: Average value (in €) of financial accounts records received by Germany from EU

Member States under DAC2 AEOI in 2018 _________________________________________ 100

Figure 8: Germany: Share of the total value of the financial accounts records sent by Germany to EU

Member States under DAC2 AEOI in 2018- Member States which account for 1 % or more of the total

value ______________________________________________________________________ 100

Figure 9: Germany: Average value (in €) of financial accounts records sent by Germany to EU Member

States under DAC2 AEOI in 2018_________________________________________________ 101

Figure 5: Luxembourg - number of DAC messages sent to other EU-27 Member States and received

from other EU-27 Member States ________________________________________________ 103

Figure 5: Luxembourg - records sent to EU-27 Member States and the US, and received from EU

Member States and the US as a percentage of total number of records sent and received ___ 103

Figure 6: Foreign portfolio holdings of EU-27 FIs (other than central banks), non-financial

corporations and households (as % of EU-27 GDP) __________________________________ 105

Figure 7: Deposits on the books of MFIs in Luxembourg (million EUR) ___________________ 107

Figure 8: Loan and deposit liabilities of Luxembourg banks to non-bank EU counterparties (million

EUR) _______________________________________________________________________ 107

EPRS | European Parliamentary Research Service

Table of tables

Table 1: Areas where DAC effectiveness can be improved and associated suggestions (please see

footnotes for original source of suggestion) ________________________________________ 43

Table 1: Income categories and their potential sources________________________________ 66

Table 2: Income categories and their potential sources________________________________ 68

Table 3: Income categories not covered by DAC provisions ____________________________ 68

Table 4: Summary of the assessment of the four potential databases_____________________ 88

Table 5: Belgium: Number of records received under AEOI _____________________________ 95

Table 6: Number of DAC1 records received and sent by Germany under AEOI in per cent of total

number of records received / sent ________________________________________________ 96

Table 7: Share (in %) of Member States which account for 5 % or more of DAC2 messages received

and sent by Germany in 2018 ____________________________________________________ 98

Table 6: Number of DAC1 reports received by Luxembourg under AEOI by tax year ________ 102

Table 7: Amending Protocols for third countries ____________________________________ 113

Table 8: Areas where DAC effectiveness can be improved and associated suggestions (please see

footnotes for original source of suggestion) _______________________________________ 123

Table 9: Information sources to be used for Q1- Effectiveness of DAC ___________________ 129

Table 10: Information sources to be used for Q2- Analysis of the evolution of tax information

exchanges __________________________________________________________________ 132

Table 11: Information sources to be used for Q3 - Coherence __________________________ 133

Table 12: List of stakeholders invited to participate in the on-line survey _________________ 134

Table 13: Reporting FIs ________________________________________________________ 137

Table 14: Types of financial accounts which are reportable if held by one or more Reportable Persons

or by a Passive Non-Financial Entity with one or more Controlling Persons that is a Reportable Person

___________________________________________________________________________ 142

Table 15: Reportable Persons (natural or legal) whose account(s) are reportable___________ 145

Table 16: Information to be reported and exchanged ________________________________ 149

Table 17: DAC2 and the CRS ____________________________________________________ 153

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

1. Introduction

1.1. Introduction to the Directive

1.1.1. Context of the Directive on Administrative Cooperation

According to the Council Directive 2011/16/EU, Directive on Administrative Cooperation (DAC or

DAC1), the key reason

for its introduction and adoption is that, in the years preceding DAC, the

mobility of taxpayers, the number of cross-border transactions and the internationalisation of

financial instruments had grown so rapidly that it became difficult for Member States to assess taxes

due properly. The risk of tax fraud and tax evasion had grown considerably as a result of these

developments which affected the ‘functioning of taxation systems and entailed double taxation

while the powers of control remain at the national level’.

Such developments were viewed as being

detrimental to the well-functioning of the internal market.

1.1.2. DAC1 and its subsequent amendments

DAC1

While an existing instrument, the Council Directive 77/799/EEC of 19 December 1977 concerning

mutual assistance by the competent authorities of the Member States in the field of direct taxation

and taxation of insurance premiums, existed at the time that DAC1 was proposed, the 1977 Directive

was generally viewed as being no longer adequate to allow the TAs to meet successfully the new

challenges posed by the rapidly growing mobility of capital and a fully new approach and

instrument was viewed as necessary.

To address these issues, DAC1 aims to develop exchanges between relevant national TAs of tax-

related information.

First, DAC1 imposes an obligation on relevant national TAs to exchange information concerning

particular cases when requested by another Member State (i.e. the so-called exchange of

information on request (EIOR)) and to make the necessary enquiries to obtain such information.

Second, and more importantly, DAC1 also imposes a mandatory automatic exchange of information

(AEOI) without preconditions as the ‘most effective means of enhancing the correct assessment of

taxes in cross-border situations and of fighting fraud’

The intention is that exchanges of information between national TAs should cover ‘all taxes of any

kind levied by, or on behalf of, a Member State or the Member State’s territorial or administrative

subdivisions, including the local authorities’ but not a) ‘value added tax and customs duties or excise

duties covered by other Union legislation on administrative cooperation between Member States’;

See Recital 1 of the DAC1.

Recital 1 of DAC1.

Recitals 4 to 6 of DAC1.

See Articles 5 and 6 of the DAC1.

See Recital 10 of the DAC1.

An automatic exchange of information had already been implemented by a precursor of the first amendment of DAC,

namely the Directive 2003/48/EC of 3 June 2003 on taxation of savings income in the form of interest payments (i.e.,

the Savings Tax Directive).

EPRS | European Parliamentary Research Service

and b) ‘compulsory social security contributions payable to the Member State or a subdivision of

the Member State or to social security institutions established under public law’.

A step-by-step approach to the AEOI was considered the best way forward and DAC1 implements

AEOI for five categories

, namely:

1 income from employment (EI);

2 directors’ fees (DF);

3 life insurance products not covered by other Union legal instruments on exchange of

information and other similar measures (LIP);

4 pensions (PEN);

5 ownership of and income from immovable property (IP).

This obligation is subject to the availability of the required information in Member States.

DAC1 also provides for a) the spontaneous exchange of information (SEOI) between Member

States

, b) the possibility of officials of the tax administration of one Member State to be allowed to

be present in the territory of another Member State

and c) the carrying out, by mutual agreement

and on voluntary basis, simultaneous control of one or more persons liable to tax established in

several Member States.

All the DAC1 provisions for exchange of information except the AEOI became effective in

January 2013. The AEOI provisions became effective in January 2015.

DAC2

To combat tax evasion on income from financial assets, the United States (US) adopted in 2010 the

Foreign Account Tax Compliance Act (FATCA) which generally requires that foreign financial

Institutions and certain other non-financial foreign entities report on the foreign assets held by their

US account holders or be subject to withholding on withholdable payments.

To implement the

reporting provisions of FATCA, the US government signed a number of intergovernmental

agreements (IGAs) under which the TAs of the non-US party to the IGA undertook to either collect

the required information from the relevant financial institutions in their country and automatically

send this information to the US TAs or to require the relevant financial institutions to the send the

information directly to the US TAs.

In 2012, five European countries (France, Germany, Italy, Spain and the United Kingdom) agreed

with the United States on a reciprocal exchange of FATCA information under IGAs concluded

between the United States and each of the five countries.

Following the implementation of FATCA, countries throughout the world expressed a growing

interest in adopting an approach similar to FATCA (i.e. AEOI) to reduce tax evasion and protect their

tax revenues. In September 2013, the G20 formally requested the OECD to develop a CRS for such

AEOI. By June 2014 the full CRS was approved by the OECD and subsequently endorsed by the G20

Finance Ministers and Central Bank Governors in September 2014. This automatic exchange of

information is implemented through the CRS Multilateral Competent Authority Agreement (CRS

MCAA) which is a bilateral agreement between two countries. As of August 2020, over 4 200 bilateral

See Article 2 of the DAC1.

See Article 8 of the DAC1.

See Article 9 of the DAC1.

See Article 11 of the DAC1.

See Article 12 of the DAC1.

FATCA also requires US citizens and residents to report their non-U.S. financial assets annually to the US TAs.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

exchange relationships had been activated with respect to more than 100 jurisdictions committed

to the CRS.

In order to ensure a common and consistent approach to the implementation by EU Member States

of the OECD CRS and avoid distortions that would be detrimental to the smooth functioning of the

internal market, an expanded automatic information exchange on the basis of a Union-wide legal

instrument was considered desirable by EU policy-makers.

To that end, the Council Directive 2014/107/EU amending Directive 2011/16/EU as regards

mandatory automatic exchange of information in the field of taxation (DAC2) was adopted on

9 December 2014 and applies from January 2016 onwards. Essentially, DAC2 transcribes into EU law

the various elements of the OECD CRS.

DAC2 requires the TAs of Member States to automatically exchange information on financial

accounts and related income, namely:

1 interest, dividends or other income generated by the financial accounts

2 gross proceeds from sale or redemption of assets held in the financial accounts

3 financial account balances

A more detailed analysis and comparison of the DAC2 provisions with those of the OECD CRS and

the US FATCA is provided in Chapter 4.

As the scope of DAC2 encompasses the exchange of information between TAs foreseen by the EU

Savings Tax Directive (STD), the latter was repealed at the same time.

DAC3

While Member States were implementing the various processes and procedures to allow for the

AEOI on DAC2 items, DAC1 was further amended in December 2015 by the Council Directive (EU)

2015/2376 of 8 December 2015 amending Directive 2011/16/EU as regards mandatory AEOI in the

field of taxation (DAC3).

The purpose of DAC3 is to bring within the scope of DAC automatic exchange of information

’advance cross-border rulings and advance pricing arrangements, issued, amended or renewed to a

particular person or group of persons upon which that person or group of persons is entitled to rely,

should cover any material form (irrespective of their binding or non-binding character and the way

they are issued)'.

Prior to DAC3, such information was shared between Member States through the SEOI foreseen by

DAC1, and Member States decided with which other Member States to share the information. To

address the fact that information had rarely been shared with the Member States affected by rulings

concerning tax-driven structures and that such rulings had, ‘in certain cases, led to a low level of

taxation of artificially high amounts of income in the country issuing, amending or renewing the

advance ruling, and left artificially low amounts of income to be taxed in any other countries

involved, an increase in transparency was viewed as essential’.

See International Framework for the CRS, OECD website.

See Recital 8 of the DAC2.

See Recital 5 of the DAC3.

See Recital 1 of the DAC3.

EPRS | European Parliamentary Research Service

The DAC3 provisions of AEOI on advance cross-border rulings and advance pricing arrangements

apply from January 2017, and from January 2018 such information is to be uploaded to a central

platform maintained by the European Commission (EC) (DG TAXUD).

DAC4

The DAC was further amended in May 2016 by the Council Directive (EU) 2016/881 of 25 May 2016

amending Directive 2011/16/EU as regards mandatory automatic exchange of information in the

field of taxation (DAC4) to require Member States to collect country-by-country information from

their multinational enterprise (MNE) groups and automatically exchange such information with

other Member States.

The policy rationale for such a requirement is that as ‘MNE groups are active in different countries,

they have the possibility of engaging in aggressive tax-planning practices that are not available to

domestic companies. When MNE Groups do so, purely domestic companies, normally small and

medium-sized enterprises (SMEs), may be particularly affected, as their tax burden is higher than

that of MNE Groups. On the other hand, all Member States may suffer revenue losses and there is

the risk of competition to attract MNE Groups by offering them further tax benefits’.

‘Member

States' TAs need comprehensive and relevant information on MNE Groups regarding their structure,

transfer-pricing policy and internal transactions in and outside the Union. That information will

enable the TAs to react to harmful tax practices by making changes in legislation or by undertaking

adequate risk assessments and tax audits, and to identify whether companies have engaged in

practices that have the effect of artificially shifting substantial amounts of income into tax-

advantaged environments’.

DAC4 requires an Ultimate Parent Entity of an MNE Group that is resident for tax purposes in a

Member State

, to file with the TAs of that Member State a ‘country-by-country report (CbCR) with

respect to its reporting fiscal year within 12 months of the last day of the reporting fiscal year of the

MNE Group’.

The country-by-country report includes the following information

‘aggregate information relating to the amount of revenue, profit (loss) before income tax,

income tax paid, income tax accrued, stated capital, accumulated earnings, number of

employees, and tangible assets other than cash or cash equivalents with regard to each

jurisdiction in which the MNE Group operates;

an identification of each Constituent Entity of the MNE Group setting out the jurisdiction of tax

residence of that Constituent Entity and, where different from that jurisdiction of tax residence,

the jurisdiction under the laws of which that Constituent Entity is organised, and the nature of

the main business activity or activities of that Constituent Entity’.

It is important to note that DAC4 covers only the exchange between the tax administrations of

Member States of CbCRs of MNE groups from the Member States. The EC also put forward in 2016 a

proposal that would oblige multinational companies to set out publicly, on a country-by-country

basis, in which national jurisdictions they earn their profits and where they pay their tax in the EU.

The interinstitutional negotiations on this proposal have not yet concluded.

See Recital 2 of the DAC4.

See Recital 3 of the DAC4.

Or any other Reporting Entity in accordance with Section II of Annex III of the DAC4.

See Article 1 (2) of the DAC4.

See Article 1 (2) 3 of the DAC4.

European Commission, Proposal for a Directive of the European Parliament and of the Council amending Directive

2013/34/EU as regards disclosure of income tax information by certain undertakings and branches, (Text with EEA

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

The provisions of the DAC4 AEOI apply from June 2017.

DAC5

The DAC1 was further amended by the Council Directive (EU) 2016/2258 of 6 December 2016

amending Directive 2011/16/EU as regards access to anti-money-laundering information by tax

authorities (DAC5) to give TAs access to the AML information, procedures, documents and

mechanisms for the performance of their duties in monitoring the proper application of Directive

2011/16/EU and for the functioning of all forms of administrative cooperation provided for in that

Directive.

AML information is obtained pursuant to Directive (EU) 2015/849 of the European Parliament and

of the Council by the financial institutions (FIs) for the identification of the beneficial owners. To

ensure effective monitoring of the application by FIs of the due diligence procedures set out in

DAC2, the TAs need access to AML information. In the absence of such access, those authorities

would not be able to monitor, confirm and audit that the FIs are applying DAC2 by correctly

identifying and reporting on the beneficial owners of intermediary structures.

DAC5 applies from January 2018.

DAC6

DAC1 was further amended in May 2018 by Council Directive (EU) 2018/822 of 25 May 2018

amending Directive 2011/16/EU as regards mandatory automatic exchange of information in the

field of taxation in relation to reportable cross-border arrangements (DAC6).

The main aim of the Directive is to provide TAs with an early warning mechanism on new risks of tax

avoidance and thereby enable them to carry out audits more effectively. Prior to the Directive, such

cross-border arrangements did not need to be reported under EU legislation. DAC6 introduced

mandatory disclosure rules for cross-border arrangements. Persons subject to these disclosure rules

include any intermediary that designs, markets, organises or makes available for implementation or

manages the implementation of a reportable cross-border arrangement.

An intermediary can be

either an individual or a company (i.e. accountants, advisers, lawyers, banks, etc.). Intermediaries,

who sell reportable cross-border tax arrangements to their clients, should report information on the

arrangement to the TAs of their home Member State. An intermediary is also any person that

provides, directly or by means of other persons, aid, assistance or advice with respect to designing,

marketing, organising, making available for implementation or managing the implementation of a

reportable cross-border arrangement.

The information on cross-border arrangements received by TAs of a Member State is to be

automatically exchanged with the TAs of the other Member States.

DAC7

Discussions are currently on-going a further amendment to DAC1 which would:

1 bring royalties within the scope of categories of income subject to mandatory automatic

exchange between the Member States; and

relevance), {SWD(2016) 117 final}, {SWD(2016) 118 final}, Strasbourg, 12.4.2016, COM(2016) 198 final, 2016/0107

(COD).

European Commission (DG TAXUD), Note on Council Directive 2018/822/EU of 25 May 2018 available at

https://ec.europa.eu/taxation_customs/business/tax-cooperation-control/administrative-cooperation/enhanced-

administrative-cooperation-field-direct-taxation_en.

EPRS | European Parliamentary Research Service

2 specify reporting rules for platform operators whose information will be also subject to an

automatic exchange of information obligation.

1.2. Objectives of the Research Paper

On 5 February 2020, the European Parliament’s Committee on Economic and Monetary Affairs

(ECON) was authorised to draw-up an implementation report on EU requirements for exchange of

tax information. In this context, the ECON Secretariat has requested the Ex-Post Evaluation Unit of

the European Parliamentary Research Service to assist in the preparation of the implementation

report with a European Implementation Assessment assessing the implementation of the

obligations of information exchange under Council Directive 2011/16/EU (DAC1) and the Directive’s

first three amendments (DAC2 to DAC4).

A European Commission (EC) evaluation in 2019

and underlying work by consultants

found that

there is limited evidence of the effectiveness of DAC and that there remains scope to enhance the

provisions. Overall, the evaluation work was based mainly on data covering a period of 5 years, from

2013 to 2017.

The main objective of this Research Paper is to assess the implementation of the obligations under

DAC, focusing on the initial Directive (DAC1) and the first three amendments (DAC2 to DAC4),

building on the 2019 EC evaluation. It also compares the arrangements under DAC with other similar

international instruments such as FATCA and CRS, and arrangements with countries previously

covered by the EU Savings Tax Directive (STD) herein referred to as the Savings Directive.

Specifically, the main objectives of the Research Paper are:

to assess the effectiveness and evolution of DAC

to analyse the differences between DAC and other similar international arrangements; and

to provide insights into potential areas for improvement.

1.3. Structure of the Research Paper

The Research Paper is structured as follows:

Section 1 provides the background information on the DAC and describes the analytical and

research approach.

Section 2 presents the research’s findings on the effectiveness of DAC.

Section 3 presents first quantitative information on the evolution of information exchanges

under DAC and examines at a high level whether any changes on cross-border holdings within

the EU can be observed following the adoption of DAC2.

Section 4 focuses on the coherence of DAC with similar international legislation (FATCA and

OECD CRS) and with other EU legislation such as the AML, Judiciary Assistance and the Financial

Assets Recovery legislation.

Section 5 sets out key conclusions and suggestions.

European Commission, Proposal for a Council Directive amending Directive 2011/16/EU on administrative cooperation

in the field of taxation, {SEC(2020) 271 final} - {SWD(2020) 129 final} - {SWD(2020) 130 final} -{SWD(2020) 131 final},

15.7.2020 COM(2020) 314 final.

European Commission, Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field of

taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, European Commission, September 2019.

European Commission, Directorate-General Taxation and Customs Union, Evaluation of Administrative Cooperation in

Direct Taxation, Final Report prepared by Economisti Associati in collaboration with ECOPA and Oxford Consulting,

April 2019.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

1.4. Analytical approach and methodology

The research approach is a mixed approach combining, reviewing and assessing qualitative and

quantitative information.

The gathering of the qualitative information involved:

A literature review;

Interviews with officials from DG TAXUD and Member States; and

An online survey of selected stakeholders.

The research plan used for addressing the specific research questions under each of the three broad

headings is described in the next three subsections. Further details of the various methods are

provided in Appendix 1.

1.4.1. Q1. Effectiveness of DAC

In evaluating the effectiveness of DAC from a qualitative perspective, the focus was on identifying

areas of discrepancy between the intended and achieved outcomes, extending beyond the

European Commission’s recent timeframe of evaluation. The approach relied on three pillars which

are described below.

Pillar 1: Understanding market reactions to DAC

The priority was to understand how the behavioural response of market participants could have

impacted the effectiveness of DAC captured by two main areas, namely:

evolution of income definition; and,

circumventions and loopholes.

Evolution of income definitions

In both cases, the first step was to identify the types of conduct that have arisen and then classify

them across a number of relevant categories, in order to facilitate structured comparisons. The

following aspects were considered among others: type of income, type of request, motivation

driving behaviour (e.g. compliance costs, lack of understanding, incentives for accurate reporting).

More specifically, the following areas were investigated: the evolution of income definitions across

time, how this was connected to successive DAC amendments and whether certain types of income

remain outside DAC’s scope, either because they were not within scope when DAC and its

amendments were adopted, or as a result of a shifting income definition landscape.

Circumventions and loopholes

One of the main challenges in effective implementation of the Directive is ensuring that the

mechanism cannot be circumvented and that all target information is captured and exchanged

where relevant. Loopholes and circumventions in the market, such as avoidance schemes and

structures which escape reporting requirements, go against the spirit of the Directive, and impact

the effectiveness of the Directive, and expected benefits for the EU and its tax revenues.

Therefore, a high-level assessment of the circumventions that arose as the result of the introduction

of each of the iterations of the Directive was conducted.

Examples of areas which were considered include:

Golden visa and other residency/citizenship schemes;

EPRS | European Parliamentary Research Service

Non-financial assets not covered by DAC;

Lack of beneficial ownership information; and

Excluded accounts.

In order to shed light on some aspects of the issue of circumventions and loopholes, a short case

study on the cum-ex / cum-cum scandal has been developed through desk research. This case study

is set out in Section 2.1. The scheme took advantage of the European rules of the taxation of

dividends. It raises the question of whether a more effective exchange of information mechanism

would have made transparent the complex transactions that allowed ownership to be held by

various parties and avoided the tax refund process altogether.

Therefore, an assessment was undertaken on whether an improved exchange of information

framework could prevent these types of scandals in the future.

Pillar 2: Obstacles in the exchange of information

This second pillar focuses on identifying legal, practical or other types of obstacles or limitations

linked to the implementation of the Directives. These were identified both by looking at inherent

framework design issues and exploring areas that have been observed to be problematic to

implement.

Issues that were considered include among others:

The legal design of the conditions for automatic exchange: are the right definitions (for e.g.

financial accounts and beneficial owners) in place and are the resulting reporting requirements

clear?

Lack of information in requested countries: where this is a significant issue, insights were gained

by looking at the particular types of income that are involved and identifying any underlying

trends. Moreover, the alleged reasons for non-availability were examined.

Administrative issues such as lack of tax authority capacity and capability.

Issues in relation to the information systems used, such as compatibility across Member States.

For example, the lack of Tax Identification Number (TIN) was identified as an issue by the 2019 DAC

evaluation.

Pillar 3: Examination of bilateral exchange data

No comprehensive quantitative information beyond that shown in the DAC evaluation undertaken

by Economisti Associati

was available and, therefore, no new analysis of bilateral information flows

(such as, for example, number of data points and value of income/financial assets) could be

undertaken.

However, some information was collected on how and the extent to which the data exchanged

between TAs are used by Member States and whether this information has had any fiscal impacts.

1.4.2. Q2. Analysis of the evolution of tax information exchanges

As access to the information pertaining to the data exchanged between the Member States’ TAs was

not granted for the purposes of the Research Paper, it was not possible to undertake a

comprehensive analysis of the evolution of tax data exchanges over time, in aggregate and between

pairs of Member States for each of the income categories covered by DAC1 and DAC2. Some

information was available for selected Member States. The main sources of such information were

Economisti Associati 2019.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

responses by governments to questions from members of national parliaments, reports from

national public audit organisations and ad-hoc studies.

Various data sources (BIS, ECB, IMF, Eurostat) providing information on cross-border holdings of

financial assets by households, non-financial corporations and FIs were also examined in detail with

a view to:

1 providing a high-level picture of cross-border financial asset holdings by EU persons (natural

and legal); and,

2 assessing whether any shift in location and volume of cross-border asset holdings can be

identified post implementation of DAC2.

1.4.3. Q3. Coherence

While the 2019 EC evaluation did not reveal fundamental coherence concerns for the Directive, a

review of how the Directive interacts with other pertinent legal frameworks while focusing on

facilitating AML information exchange was conducted.

First, DAC was compared to the US FATCA and the OECD’s CRS standard based on publicly available

information discussing framework design and structural arrangements. The analysis focused on key

areas such as types of income covered, types of exchanges (spontaneous, automatic, etc.) and

reporting requirements.

Second, an analysis of how DAC2 and the STD differ was also conducted.

Third, the research examined how DAC interacts with relevant EU rules on judicial assistance, money

laundering and asset recovery. DAC5 and beneficial ownership rules are of significance here, as well

as their interaction with the 4th AML Directive (AMLD4). Due to the secrecy surrounding the types

of arrangements used in tax abuse, establishing beneficial ownership, as well as recovering assets,

can be a significant challenge. Anti-money laundering legislation in the EU currently provides for

beneficial ownership information to be kept accurately and up to date by relevant entities, and this

should be made available to competent authorities (CAs) and obliged entities. In conjunction with

this AML legislation, DAC5 allows TAs, amongst other things, to access beneficial ownership

information.

EPRS | European Parliamentary Research Service

2. Effectiveness of DAC

Assessing the effectiveness of any Directive requires an examination of how it performed in

comparison to the objectives it set out to achieve and its intended outcomes. As a whole, DAC’s

objective is to enhance cooperation between Member States in order to tackle tax avoidance and

evasion more effectively and improve the level of transparency, given the increasingly global and

cross-border nature of these activities. In particular, it equips Competent Authorities (CAs) across

the EU, usually the Tax Administration (TA) or Ministry of Finance (MoF) of the relevant Member

State, with the requisite processes and structure to cooperate towards a direct taxation framework

that ensures everyone contributes fairly, and mandates the exchange of tax relevant information

between Member States to increase transparency and reduce tax avoidance and evasion.

The analysis presented below examines how DAC1 to DAC4 performed vis-a-vis their respective

objectives and contribution to the wider overall objective and is based on three pillars presented in

the previous chapter (see section 1.4.1).

Assessing the effectiveness of DAC

When looking to assess the effectiveness of the Directive and its amendments, there exists a set of

overarching features that can contribute to its effectiveness, as outlined below. These features will

be discussed throughout this Research Paper and will form the basis of the final assessment of

whether DAC1 to DAC4 function effectively and achieve their intended objectives.

Comprehensiveness or scope: to what extent do the provisions in the Directives capture all

relevant categories of income and capital in order to effectively address tax fraud and evasion in

an evolving globalised context?

Relevance or appropriateness: are the requirements and definitions under the provisions in

relation to information exchanged, the roles and duties of reporting entities and CAs, amongst

other requirements relevant and appropriate for the purposes of allowing Member States to

assess tax correctly?

Timeliness: are the time limits set out in the provisions for the production and exchange of

information appropriate in order for Member States to be able to assess tax correctly and in a

timely manner, and deter potential tax fraud and evasion?

2.1. Pillar 1: Understanding market reactions to DAC

This section focuses on evolution of the definition of income and any discrepancies between the

definitions captured within DAC, and those which evolved in the global economic environment and

the exploitation of circumventions or loopholes which act against the spirit of the Directives.

2.1.1. Evolution of the different features of DAC

While the previous chapter provided a brief historical overview of the DAC1 and its subsequent

amendments, the present section highlights a number of features of the DAC1 and its amendments

which are important for understanding and assessing the market reactions to these various DAC

Directives

DAC1 - Council Directive 2011/16/EU

As noted in the previous chapter, DAC1 mandates the AEOI for EI, DF, PEN, LIP and IP. The

information relating to these categories must be readily available and CAs must report on at least

three of the five income categories listed. The Directive also allows for spontaneous exchange of

information (SEOI) where one Member State has information to suggest that there may be a loss of

tax in another Member State, and exchange of information on request (EOIR) where one Member

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

State can request from another Member State information that it has in its possession or that it

obtains from administrative enquiries.

DAC1 is a set of minimum rules and Member States can choose to cooperate more closely with one

another. Indeed, the Directive goes beyond EOI and additionally allows for simultaneous controls,

and the presence in administrative offices and participation in administrative enquiries of one

Member State’s officials in another Member State.

DAC2 - Council Directive 2014/107/EU

Following the recognition of the importance of AEOI at an international level through FATCA and

the OECD’s CRS (see Section 4.1.3), and the European Commission’s Action Plan on tax fraud and

evasion

, DAC2 was adopted to implement the OECD CRS in EU law. It mandates the AEOI of

financial account information on an annual basis, including information such as interest, dividends,

or other income generated by the financial account, gross proceeds from a sale or redemption, and

account balances. Whereas under DAC1 there was a condition of availability for information

exchanged automatically, under DAC2 this condition does not apply.

Certain FIs (‘Reporting FIs’), including custodial institutions, depository institutions, investment

entities and specified insurance companies, as well as specified non-FIs, are required to report

relevant accounts (‘Reportable Accounts’) and information in respect of those accounts to the

Member State’s CA. Reporting FIs are required to have conducted the required review and due

diligence procedures to obtain and review the information, including ascertaining the tax residence

of the account holder(s) to establish whether they are a Reportable Person, and establishing who

the controlling persons of an entity are.

Member States are required to have rules and procedures in place to ensure that the required due

diligence procedures are carried out, and that the Reporting FIs comply with the reporting

requirements, including enforcement provisions to address non-compliance.

DAC3 - Council Directive (EU) 2015/2376

Under DAC1 SEOI was already possible and allowed a Member State to exchange information

relating to advance cross-border rulings (ACBRs) and advance pricing arrangements (APAs), where

there was a reason to suppose that there may be a loss of tax in another Member State. However,

this was a discretionary provision which, amongst other reasons mentioned in Section 2.2.2

(practical obstacles), meant that this information was not exchanged in an effective manner. ACBRs

and APAs are common practice and enhance business certainty and consistent application of the

law, however they can also be used to create artificial distortions in the level of taxation across

Member States. In order to counter this, DAC3 was introduced and mandates the AEOI in relation to

ACBRs and APAs.

DAC4 - Council Directive (EU) 2016/881

In line with global developments and, specifically, Action 16 in the BEPS Action Plan, DAC4

introduced the requirement for mandatory AEOI of CbCRs of MNE groups. Firstly, the Reporting

Entity within the MNE group is responsible for providing the CA of the Member State a CbCR for the

group. This is then exchanged on a yearly basis under AEOI by the competent authority (CA) of the

European Commission, Communication from the Commission to the European Parliament and the Council, An Action

Plan to strengthen the fight against tax fraud and tax evasion, {SWD(2012) 403 final}, {SWD(2012) 404 final},

COM(2012) 722 final 6 December 2012

https://ec.europa.eu/taxation_customs/sites/taxation/files/com_2012_722_en.pdf

EPRS | European Parliamentary Research Service

relevant Member State with any other Member State in which any of the constituent entities of the

group are resident for tax purposes, or where they are subject to tax on the basis of having a

permanent establishment in the Member State.

The requirement to report excludes MNE groups with total consolidated group revenue of less than

€750 000 000 during the fiscal year immediately preceding the reporting fiscal year, and the

incentive to comply and enforcement of non-compliance through penalties is at Member States’

discretion, who should provide for ‘effective, proportionate and dissuasive’ penalties.

The requirement to report excludes MNE groups with total consolidated group revenue of less than

€750 000 000 during the fiscal year immediately preceding the reporting fiscal year, and the

incentive to comply and enforcement of non-compliance through penalties is at Member States’

discretion, who should provide for ‘effective, proportionate and dissuasive’ penalties.

2.1.2. Evolution of income

DAC sets out to achieve its intended objectives in an increasingly globalised context with

dynamically transforming value chains and evolving types of economic activity. These conditions

magnify the dimensions of the information set required to tackle tax fraud and tax evasion and

necessitate efficient and effective use of technological tools and technical expertise. Indeed, as can

be seen from the above summaries, each amendment to DAC is intended to widen the scope of the

provisions to capture these evolutions relevant to cross-border taxation. An important first step in

examining market reactions is to understand which different categories of income currently exist

and how these might have evolved also due to, for example, different corporate structures or

technological advances.

The table below summarises different categories of income and outlines their potential sources.

Even though a number of these categories can be perceived as more ‘traditional’ than others, e.g.

income from employment versus income from cryptocurrencies, they have also been impacted by

technological progress and globalised value chains.

Table 1: Income categories and their potential sources

Income category Source

Income from employment

Working for a company

Working for one’s own company (declared

income)

Self-employment (e.g. sole trader)

Directors’ fees

Working for a company

Working for your own company

Life insurance products

State/Government

Private/Personal

Pensions

State/Government

Private/Personal

Ownership of and income from immovable

property

Rental property and holiday homes

Online Platforms

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Income category Source

Capital gains

Sale of assets

Dividends

Private share ownership

Interest

Savings/Current Accounts

Debt securities held directly (i.e. not through

custodial account)

Loans (made directly or intermediated

through peer-to-peer lending platforms or

crowdfunding platforms)

Payments in kind

Gross proceeds from the sale or redemption

of financial assets

Online platforms

Private investments

Royalties

Ownership of assets e.g. intellectual

property, franchises and natural resources

Cryptocurrency assets and related income

Online platforms

Private investments

Income from online platforms

Sale of goods

Sale of services

Rental of goods and services

Increasingly connected and digitalised economies have enabled novel asset classes such as

cryptocurrencies, which can be mined and traded, generating income, and can serve as an

alternative to standard currency. They have also enabled alternative uses for assets such as property

and vehicles, or skills, through the introduction of online platforms that facilitate the sharing

economy. While there is a large number of regulatory and policy questions surrounding these

modern economic activities, they are income-generating and often substitute some of the more

‘traditional’ means of earning income.

Innovative business models have enlarged a variety of sectors while posing a challenge to

incumbents. For example, the financial sector has seen the landscape change which has allowed a

number of ‘fintech’ challengers to compete with incumbent FIs through offering exposure to a

number of financial products and services to retail investors and businesses. Fintech platforms can

unlock new avenues to earn income from equity and fixed income investments, short-term trading

or pension and insurance products.

As income categories have evolved so too has DAC, with its subsequent amendments including an

increasingly wider range of income types. The table below summarises the types of income included

across DAC1 to DAC4 and is the basis for a comparison between available income types, and those

covered under the Directive.

EPRS | European Parliamentary Research Service

Table 2: Income categories and their potential sources

DAC Income type

DAC1

Income from employment

Directors fees

Life insurance products

Pensions

Ownership of and income from immovable

property

DAC2

Dividends

Interest

Gross proceeds from the sale or redemption

of financial assets

Cash-value insurance contracts

DAC3

Cross border tax rulings can cover various

types of taxes and income, and therefore all

income types are in scope of the provisions

APAs relate to transfer pricing agreements

for companies

DAC4

Company revenue/income

The effectiveness of DAC1 is, therefore, by definition, limited to the five categories of income that it

covers. Even for those categories however, Member States are only required to report what is readily

available and for a minimum of three out of five categories, further limiting effectiveness. DAC2 built

upon DAC1’s requirements by adding a suite of financial income under its scope as shown in the

table above, and by strengthening the data collection and reporting requirements. Lastly, DAC3 and

DAC4 did not expand the definitions of income included but introduced additional reporting

requirements for ACBRs, APAs and MNEs, respectively.

As illustrated above, the majority of income categories are, at least in principle, covered by DAC. The

following table lists the categories of income that are either not covered by DAC1 to DAC4 or that

involve complexities in their assessment.

Table 3: Income categories not covered by DAC provisions

Income category Issue with DAC coverage

Royalties

Royalties have not been covered by DAC1-4,

but they are covered under the DAC7

proposal

Capital gains

Capital gains related to immovable property

are not covered, though DAC1 covers data

related to ownership and income from real

estate.

Only proceeds from the sale of financial

assets are covered under DAC2. No

information on capital gains.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Income category Issue with DAC coverage

Capital gains from other types of assets not

explicitly covered above are not, however,

covered by DAC1-4.

Directors’ fees

Literature suggests directors’ fees can be

difficult to distinguish from general

employment income in practice.

Cryptocurrency assets and related income

Member States interviews highlighted that

cryptocurrency assets and related income

are not explicitly captured under DAC1-4.

DAC2 leaves the classification of

cryptocurrencies as a ‘financial account’ up

to Member States discretion

The DAC7

proposal, however, introduces detailed

reporting requirements for platform

operators while DAC8 covers cryptocurrency

trading platforms and e-money.

Income from online platforms

As noted above, this is an income category

enabled by technological advancements,

digital connectivity and innovative business

models whose relevance has increased over

time.

This has been noted extensively in Member

States interviews as an area that is receiving

attention.

Income from the sale of goods or services, or

from crowdfunding investments and loans

through online platforms is covered under

the DAC7 proposal and has not been

covered by DAC1-4.

Interest on loans

The analysis conducted suggests that

information on interest on loans made

directly or intermediated through peer-to-

peer lending platforms or crowdfunding

platforms is not held and/or not received by

the Reporting Entities, therefore falling

outside the scope of DAC2.

DAC7 proposal, however, introduces

detailed reporting requirements for platform

operators that can include income from such

platforms.

Payments in kind (PIK)

PIK in lieu of cash should be captured under

employment income under DAC1 but only if

information is available.

Economisti Associati 2019.

See Andres Knobel, Reporting taxation: Analysing loopholes in the EU’s automatic exchange of information and how to

close them, A Report Commissioned by the Greens/EFA Group in the European Parliament, October 2018

EPRS | European Parliamentary Research Service

Income category Issue with DAC coverage

PIK related to financial instruments where an

investor receives, e.g. equity instead of cash,

would not be covered under DAC2 despite

relating to a financial account.

It is clear from the analysis above that the definition of income has evolved with the adoption of

DAC amendments, and captures a large majority of the categories of income. However, the

definition of income in the global and digitalised economy has also evolved, with new forms of

earning income emerging, such as through cryptocurrencies. Notwithstanding this, there is overall

agreement among Member States interviewed that the scope of income covered by the DAC

provisions is satisfactory, with the notable exceptions of royalties and income related to the digital

economy, which are covered by proposals for DAC7.

2.1.3. Circumventions and loopholes

Whilst the section above highlights areas where market actors may have circumvented the

Directives as a result of not falling within the income scope, there are also examples of market

conduct by individuals and companies that have circumvented the spirit of the Directive despite the

underlying income categories falling within scope. This section highlights a selection of these

instances where DAC has been ineffective, and categorises these by the amendment to the Directive

(DAC2 to DAC4) that they are most closely associated with. DAC1 is not explicitly considered below

as Section 2.1.2 has already explored circumventions in relation to the evolution of income. It should

be noted that interviews with Member States and the stakeholder survey overall did not report any

particular loopholes in relation to DAC1 to DAC4.

DAC2 circumventions and loopholes

A number of circumventions and loopholes can be identified with regards to DAC2, primarily

through the literature review, and these have been set out below, categorised by themes. The list

below highlights key areas where the design of the Directive has allowed behaviours to take place

which go against the spirit of it. In addition, whilst DAC2 is the implementation of CRS in the EU, for

the purposes of this Research Paper the section below focuses purely on DAC, with any differences

related to CRS discussed in Section 4.1.4.

Reporting

1 Under Subparagraph A(3) of Section VIII of Annex I of DAC2, the definition of an FI primarily

refers to custodial or depository institutions, investment entities, and specified insurance

companies. Although at first sight it may seem to cover a broad range of institutions, within

this scope there are exceptions, given that, for example, investment entities managed by an

individual rather than a reporting FI are not included. In addition, under subparagraph B of

Section VIII of Annex I of the Directive, some FIs are specifically excluded, such as certain

types of pension funds, qualified credit card issuers, or FIs that present a low risk of being

used to avoid tax. Additionally, some electronic money institutions may not be captured

within the scope of the Directive, given that they would need to meet the definition of a

depository institution and accept deposits in the ordinary course of banking or similar

business, which is uncommon for these types of institutions.

2 Where an FI does not have any reportable persons, it does not have to report anything at all,

and no reasons are given as to why no information has been reported

See Andres Knobel, October 2018 and Andres Knobel, ‘How to assess the effectiveness of automatic exchange of banking

information’, article, Tax Justice Network, December 2018.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

3 Although the Directive initially sets out that ‘thresholds not be generally included in this

Directive as they could be easily circumvented by splitting accounts into different Financial

Institutions’, Subparagraph A of Section V of Annex I excludes pre-existing entity accounts

with a value of less than $250 000 from being reviewed, identified or reported. This is an

aggregate value, and any new account in the same FI can be treated also as a pre-existing

account as per Subparagraph C9(b) of Section VIII of Annex I, if the account holder already

holds another account, no new information is required to open the new account, and the

aggregate value of the accounts is no larger than the $250 000 threshold.

4 Subparagraph C17 of Section VIII of Annex I excludes certain types of accounts from

reporting. These ‘excluded accounts’, such as escrow accounts, accounts related to certain

types of pension or retirement funds, or accounts related to court orders or judgements, do

not need to be reported by the FI. In addition, similar to reporting FIs, accounts which present

a low risk of being used to avoid tax are also excluded.

Tax residence

5 For pre-existing individual accounts, residence is primarily determined by the address held

by the FI (lower value accounts) or other additional indicators (high value accounts), and, for

new individual accounts, a self-certification must be obtained from the account holder and

checked against AML/KYC information held by the FI. The reliance on the address for

evidence of residence can prove problematic, as the terms ‘address’ and ‘residence’ do not

naturally equate, and an account holder may be able to provide a substantiated address in a

jurisdiction where it is not resident

. For pre-existing entity accounts, the FI should rely on

AML/KYC information, including place of incorporation or, similar to individual accounts, an

address in a Member State. For new entity accounts, a self-certification must be obtained

from the account holder, which allows the FI to determine their residence, and checked

against AML/KYC information held.

6 The reporting FI is not required to check the information is still relevant at regular intervals,

or carry out the procedures set out above, unless there is a change of circumstances which

suggests that a review is required. Therefore, an account holder may hold a residence

incorrectly and avoid being considered a reportable person for a substantial amount of time,

if there is no change in circumstances.

7 As per Subparagraphs B5 and C5(c) of Section III of Annex I, if the reporting FI is unable to

establish the residence of the account holder for a pre-existing individual account, the

account is treated as undocumented and must be reported as such by the reporting FI, but

no other action is required to be taken.

8 An account holder can purport to be resident for tax purposes in another jurisdiction outside

of EU Member States in order to not be considered a reportable person under DAC and

therefore escape exchange of information requirements under the Directive’s provisions.

Misreporting residence can also help avoid tax even if the reported (rather than actual)

residence is based in another EU Member State. This effect can be seen as a consequence of

different schemes which have been developed, including golden visa and other

residency/citizenship schemes. In these schemes, residency is granted in exchange for a

substantial investment, rather than through living in the particular jurisdiction. A more in-

depth analysis of these types of schemes is outlined in Figure 1.

See Antoine Dayez, ‘DAC2/CRS: multireporting and data protection’, AEFI meeting, November 2016.

EPRS | European Parliamentary Research Service

Figure 1: Effectiveness of the DAC and Golden Passports / Residency Visas

While Investor Citizenship and Residence Schemes, i.e. the Golden Passports and Golden Visas, offered by many

EU Member States raise a number of issues and risks

, especially regarding EU citizenship, security, money

laundering, tax evasion and corruption, the present discussion focuses on the potential implications of such

schemes for the effectiveness of DAC, especially DAC2.

As of early October 2020, three Member States (Bulgaria, Cyprus and Malta), offered the possibility of acquiring

their citizenship in return for investments into the country. On 13 October 2020 Cyprus announced that it was

suspending, as of November 2020, its scheme following revelations of potentially inappropriate political

interventions in providing access to the scheme

. Subsequently, on 20 October 2020 the EC launched

infringement procedures against Cyprus and Malta by issuing letters of formal notice regarding their investor

citizenship schemes also referred to as golden passport schemes. According to the Commission ‘….the granting

by these Member States of their nationality – and thereby EU citizenship – in exchange for a pre-determined

payment or investment and without a genuine link with the Member State concerned, is not compatible with

the principle of sincere cooperation enshrined in Article 4(3) of the Treaty on European Union. This also

undermines the integrity of the status of EU citizenship provided for in Article 20 of the Treaty on the

Functioning of the European Union.’

Although only three Member States operated golden passport schemes in October 2020, 19 Member States

offered in 2019 residency rights in exchange for investments into these Member States under the so-called

residence by investment schemes

. While the specifics of the different schemes vary somewhat, a common

feature is that they impose only minimal or very light residency requirements.

These light residency requirements are at the source of the tax avoidance issues. Under DAC2 and the OECD

CRS, Reporting FIs are required to report to their TAs the value of financial accounts owned by non-residents

and the income associated with these accounts, and TAs send this information to the TAs of the home countries

of the non-residents.

As pointed out by the OECD, golden passport and visa schemes do not in themselves offer a way to escape

reporting under the CRS, which requires taxpayers to self-certify in all their jurisdictions of residence for tax

purposes. Residence status granted by these schemes indeed does not necessarily grant tax residence status.

However, in part due to the light residency requirements, it is possible for a beneficiary of a golden passport or

visa to declare misleadingly to the financial institution where the beneficiary holds an account that she/he is a

resident of the country where the financial institution is located while being effectively a resident of another

country.

However, the TAs of the country of effective residence (from a taxation perspective) will not receive

any account and income information from the TAs of the country having granted the golden passport or golden

visa because that information is not reportable by the Reporting Financial Institution due to the misleading

information provided by the account holder. DAC2 and the CRS cannot detect such tax avoidance.

See Report from the Commission to the European Parliament, the Council the European Economic and Social Committee

and the Committee of the Regions, Investor Citizenship and Residence Schemes in the European Union,

COM(2019)

12 final, European Commission, January 2019.

BBC News, Cyprus to suspend 'EU golden passports' scheme, 13 October 2020, https://www.bbc.co.uk/news/worl d-

europe-54522299.

See European Commission, January 2019.

Bulgaria, Czechia, Estonia, Ireland, Greece, Spain, France, Croatia, Italy, Cyprus, Latvia, Lithuania, Luxembourg, Malta,

Netherlands, Poland, Portugal, Romania and Slovakia. The United Kingdom also operated such a scheme.

See European Commission, January 2019.

See OECD, Preventing abuse of residence by investment schemes to circumvent the CRS, Consultation document, 2018

and European Parliament, EPRS, Citizenship by Investment (CBI) and Residency by Investment (RBI) schemes in the

EU, State of play, issues and impacts, Study

, October 2018.

For examples of how taxpayers can use golden passport and visa schemes to engage in tax avoidance, see OECD, 2018

and Knobel, October 2018.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

The incentive to use a golden passport or visa to avoid paying taxes is particularly strong when the tax rate in

the country granting the golden passport or visa is much lower than in the country of effective residence of the

taxpayer. It may well be that the recipient of the golden passport or visa fully declares to the TAs of the country

granting the golden passport or golden visa the income received from the financial account, but the fact is that

such taxes should have been paid in the country of effective residence.

Beneficial ownership

9 With regards to ownership of accounts and entities, DAC2 refers primarily to an ‘account

holder’ or ‘controlling persons’, rather than beneficial owners. In defining the term

‘controlling persons’ Subparagraph A(6) of Section VIII of Annex I indicates that it must be

interpreted alongside the Financial Action Task Force (FATF) recommendations, but this is at

the discretion of the Member State. FATF recommends setting a minimum threshold to

determine beneficial ownership, and indicates 25 % ownership interest as an example of this.

Across the EU Member States, the 25 % example has been taken as the standard threshold

to apply

and this means that anyone holding 25 % or less of the ownership interest will not

be identified as a beneficial owner or controlling person.

Under DAC2, beneficial ownership is determined for both individual accounts, where the

individual directly holds the interest in the account, or where the individual holds the

account through an entity, and this is a passive non-financial entity (NFE). Tax avoidance and

evasion arrangements are usually highly complex structures so that real, and not simply

legal, ownership of the underlying assets or income can be hidden. Therefore, it is necessary

to look through the arrangements to ascertain who the correct taxable person is. Under

DAC2, a passive NFE will be looked through to identify who the controlling persons are and

the residence of both the entity and the controlling persons reported under DAC2. However,

the same does not apply to active entities, where only the residence of the entity needs to

be reported, and not that of controlling persons / beneficial owners. As is explored in Section

4.3.1 of this Research Paper, only if the entity is resident in the EU and is subject to rules under

the Anti-Money Laundering Directive, will the beneficial ownership be identified.

The above suggests that, for passive NFEs, as long as no party has over 25 % beneficial

ownership of an entity, there will be no reportable persons in relation to that entity. Indeed,

with regards to issues surrounding beneficial ownership, one issue that was mentioned by

some tax officials in interviews is that financial accounts owned by several owners through a

trust or similar financial structure will not be reportable if none of the owners of the trust

owns more than 25 % and therefore none will have to be declared by the trust to the

reporting FI as ultimate beneficial owner. It is possible as a result of this, that accounts in

relation to that entity may be considered undocumented accounts. In addition, where the

entity is active, the beneficial owners will remain hidden irrespective of the 25 % threshold

10 Under Subparagraph D(5) of Section VIII of Annex I, with reference to trusts, the term

‘controlling persons’ includes the settlor(s), the trustee(s), the protector(s) (if any), the

beneficiary(ies) or class(es) of beneficiaries, and any other natural person(s) exercising

ultimate effective control over the trust. These must be identified for the purposes of DAC2,

unless the trustee is a reporting FI and reports all reportable accounts, where the trust itself

is then considered to be a non-reporting FI, allowing for the identity of beneficial owners to

remain hidden, and any economic benefit accruing to them potentially avoiding taxation.

Additionally, the categories of person included in the definition of ‘controlling persons’ for

trusts can be difficult to assess. For example, dummy settlors are a commonly used tool for

the avoidance of tax in trust structures, whereby the settlor of the trust is legally person A,

however the economic value of the settlement of the trust was in actuality provided by

See KPMG, Global Legal Services, ‘UBO disclosure requirements within the EU’, April 2019.

https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:02011L0016-20180101.

Based on cross-analysis of the AML and DAC legislation alongside SME knowledge.

EPRS | European Parliamentary Research Service

person B. In this instance person B would be the real settlor, but their identity would remain

hidden and person A would instead be identified as a ‘controlling person’ in relation to the

trust. Furthermore, with reference to beneficiaries of the trust, in some cases, in particular

those relating to classes of beneficiaries, it is not possible to establish who the beneficiaries

are until a distribution takes place, making it difficult for FIs to identify all relevant ‘controlling

persons’.

11 A key factor in the success in identifying the real beneficial owner is being able to distinguish

between the legal and beneficial owner, which can be difficult to achieve. For example,

settlors are considered to be beneficial owners in relation to trusts, however, this does not

take account of the use of dummy settlors. In addition, whilst documentation may suggest

that a particular individual is the legal owner, the economic benefit may be obtained by a

different individual, the real beneficial owner. As is discussed below, the appropriate

functioning of the Directive in relation to beneficial ownership relies on well-functioning due

diligence measures.

Due diligence procedures and correctness of information

12 Under DAC2 most due diligence requirements are satisfied by checking that the self-

certification provided by the account holder seems reasonable or by relying on AML/KYC

documentation. This approach poses several challenges.

Firstly, for lower value individual accounts, there are no enhanced due diligence procedures

required, and the reporting FI must only look at the address on file (for pre-existing accounts)

or be satisfied of the reasonableness of the self-certification provided by the account holder.

Evidently, this can provide an opportunity for misrepresentation by account holders so that

they are not captured within the scope of DAC2, and the FI would not be under any

obligation to ensure that the information is correct beyond the process outlined above.

In addition, whilst in principle relying on AML/KYC information may seem appropriate, this

assumes that there is a high level of compliance with these procedures by Member States

and that the information obtained through these processes can be considered to be

accurate. In the most recent assessment (updated 4 November 2020) of countries’ AML/CTF

measures carried out by FATF, the 18 Member States included in the assessment

, did not

perform well across key indicators

. Indeed, when being ranked against whether FIs

adequately apply AML/CTF measures (Immediate Outcome 4), most were rated as displaying

a ‘moderate’ or ‘low’ level of effectiveness, with only Spain being rated as having a

‘substantial’ level of effectiveness, and no Member State attaining a ‘high’ level of

effectiveness. In addition, when being ranked on compliance with recommendations on

customer due diligence (Recommendation 10), only 3 Member States (Austria, Belgium and

Greece) were assessed to be ‘compliant’, with the other Member States being assessed as

either ‘largely compliant’ or ‘partially compliant’. Therefore, the correctness of the

information cannot be guaranteed and may mean that persons are still able to avoid or

evade tax.

13 There are no prescribed sanctions under DAC2 for FIs who do not report, or report

information falsely or incorrectly. Indeed, Section IX of Annex I simply states that Member

States must have rules and procedures in place to ensure that due diligence procedures are

carried out, and that FIs comply with reporting, including effective enforcement provisions

to address non-compliance. Measures vary across Member States, with some imposing

financial penalties, and others publishing details of the non-compliant FIs. Regarding

financial penalties, there is some disparity between Member States: for example, in Estonia

https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:02011L0016-20180101.

Austria, Belgium, Cyprus, Czechia, Denmark, Finland, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Malta, Portugal,

Slovakia, Slovenia, Spain and Sweden.

Financial Action Task Force, 4th Round Ratings, November 2020.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

the maximum penalty is €3 300

while in Austria the fine ranges from €10 000 to €200 000

depending on the offence. Poland, in contrast, can impose a fine of up to €1 000 000 and also

publishes the details of those who are non-compliant with DAC2 on the TA’s website. Whilst

Member States interviewed reported that FIs are generally compliant, there is no evidence

to suggest the impacts that can be attributed to the deterrence effect of the potential

sanctions.

Other circumventions and loopholes

In addition to the circumventions and loopholes discussed above, there are also a number of others

worth highlighting which, although it could be argued that they have been possible as a result of

the design of the DAC2 provisions, present a more systematic exploitation through the use of

avoidance schemes, or holding of assets outside the scope of the Directive.

Dividend arbitrage

14 Since its introduction in 2011, the effectiveness of DAC has been a topic of intense debate

across the EU. Observers (e.g. the media) have pointed to the number of tax evasion or

avoidance schemes and scandals exposed by investigative journalists and whistle-blowers

alike, as proof of the legislation not working as well as intended

. In particular, the cum-ex

schemes reported in 2018 by German media

and the parallel cum-cum schemes have

generated significant media and political attention. Figure 2 below, sets out a short case

study briefly outlining the schemes, their interaction with DAC, and whether an improved

framework for exchange of information would have avoided them.

Figure 2: Case study on cum-ex/cum-cum schemes

The cum-ex scheme was an aggressive type of dividend arbitrage understood to have started around 2001 and

ended, to a large extent, in 2012. Various EU Member States were affected, in particular Germany. Institutional

investors in Germany took advantage of a loophole in the German tax code that allowed multiple investors to

claim back dividend withholding tax (WHT) for the same set of shares. The parallel cum-cum schemes involved

a foreign investor loaning their shares in a domestic company to a domestic investor so they would receive the

dividend and claim the refund of the WHT for which they were not entitled, returning the shares to the foreign

investor and splitting the dividend and tax refund. These schemes were made possible due to minor loopholes

resulting from differences in double taxation treaties between countries.

Based on independent analysis by the EP, the cost to EU taxpayers is estimated to have been roughly €55 billion

from both the scandal and schemes combined

. Given the large numbers involved it is unsurprising to see that

a number of EU Member States were impacted, in particular Germany, but also other countries such as Austria,

Belgium, Denmark, Norway and Switzerland. The widespread nature of the schemes has led to questions as to

how regulators and officials were not alerted to the issue earlier, and over the effectiveness of DAC provisions

and exchange of information more broadly.

The schemes in the context of DAC and exchange of information

DAC1 came into force in 2011, with exchanges commencing in 2013 for non-AEOI and 2015 for AEOI on taxable

periods from 1 January 2014. Therefore, any information pertaining to cum-ex schemes would not have fallen

under these provisions and exchanged by Member States, as changes in the German banking and legal systems

Estonia, Tax Information Exchange Act, 2019.

Austria, Beneficial Owners Register Act, 2017.

See group of various media enterprises led by !Corrective, Cumex Files, A cross-border investigation, How Europe's

taxpayers have been swindled of €55 billion, 18 October 2018, https://cumex-files.com/en/

Ibid.

See European Parliament, The Cum-ex files, Information document, November 2018.

EPRS | European Parliamentary Research Service

were made in 2012 that resulted in these schemes no longer being possible. However, by the time cum-cum

schemes were closed in 2017, AEOI provisions under both DAC1 and DAC2 had been in place for two years, with

exchanges under DAC2 taking place since 2017 (for periods since 2016). Under DAC2 in particular, AEOI with

regards to dividend income could have been used, in conjunction with other information held by the relevant

Member State, to identify both cum-ex (if provisions had existed at the time) and cum-cum schemes. The

relevant tax authority could cross-reference this information against WHT refund claim

s to identify

discrepancies.

An analysis of the literature on how the schemes operated, combined with an analysis of DAC provisions at the

time has identified a number of reasons as to why they may not have been sufficient to identify these schemes.

In particular, four key areas are mentioned and are summarised below:

Timeliness of exchanges

DAC2 requires information to be exchanged on an annual basis, covering the previous calendar year reporting

period. Given the pace at which the trades were taking place

, even if the information would have been

captured under the Directive, all of the transactions from the reporting period would have already taken place

and the economic benefit derived from them. Any trades currently taking place would not be identified until at

least the following year, with EOI required within 9 months of the end of the relevant reporting period.

Accuracy of information exchanged

DAC2 requires FIs to exchange information on relevant persons, who are defined as having control and are

beneficial owners. However, there is no requirement for them to check the accuracy of this information beyond

that which was required under AML provisions. In addition, the limited incentives for FIs to correctly report

information under DAC2 could have aggravated any inaccuracies. From the perspective of the TAs receiving the

information, it would appear that there were two owners when in actuality there was only one.

Scope of the provisions

The WHT claimed as part of the scheme was effectively the withheld capital gains tax deducted from the

dividend payments.

The DAC provisions did not, and still do not, include any exchange of information

requirements for capital gains, however information exchanged under the provisions can be used to assess

various types of taxes, including capital gains tax. In addition, at the time there were no requirements for

intermediaries to report schemes which displayed certain characteristics of tax avoidance.

Use of the provisions

DAC1 allows for spontaneous exchange of information and group requests, and sets minimum standards for

cooperation across Member States, however, Member States can cooperate more closely. The 2019 EC

evaluation highlighted that Member States do not often go beyond the minimum

, and this is reflected in the

case of these schemes. Member States did not sufficiently cooperate through appropriate mechanisms such as

spontaneous exchange in order to alert other relevant Member States of the schemes. Information sharing was

limited, and this resulted in the schemes being able to continue for a substantial amount of time and Member

States losing out on tax revenues.

The response, the future and an improved framework

In response to the schemes, the view of Commissioner Moscovici was that the provisions under DAC were

sufficient; it was the lack of enforcement and use which prevented the schemes being identified in other

countries sooner. Several Members of the European Parliament (MEPs) went further, seeking a variety of actions

Duncan Wigan, ‘Case study: The Cum-Cum and Cum-Ex schemes’, COFFERS D4.5, September 2019 http://coffers.eu/wp-

content/uploads/2020/03/D4.5-Case-Study.pdf.

Rahman Ravelli, ‘Cum-Ex Trading Schemes Explained & FAQs’, downloaded on 20 December 2020 from

https://www.rahmanravelli.co.uk/articles/cum-ex/

See European Commission, September 2019.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

in response including opening investigations and an expansion of the coverage of DAC. On this latter point,

there was agreement between MEPs that an extension of the framework was needed to cover other elements

of the schemes and, as can be seen above, there are other concerns mentioned in the literature

Partly as a result of these and other schemes, the scope of DAC has been extended under DAC6 to cover

intermediaries. Multiple intermediaries were involved in the facilitation of the cum-ex and cum-cum schemes

and this amendment now requires cross-border transactions concerning at least one EU Member State to be

reported, if they meet one of a number of hallmarks. Although dividend arbitrage is not specifically mentioned

within DAC6, some of the hallmarks would capture elements of the schemes and would therefore ensure

automatic information exchange on these transactions. It is likely that had this amendment been in place at the

time of these schemes, they would have been identified.

DAC6 reflects changes to improve the framework for EOI in relation to these schemes, however it does not fully

address the remaining concerns around timeliness of exchanges, accuracy of the information reported and

scope of the Directive. The concerns around timeliness could have been addressed to some extent through

more frequent reporting. However, given the speed at which transactions took place it is unlikely that this would

have had a significant effect. With regards to ensuring the exchange of accurate information, it is possible that

stronger due diligence procedures and accurate reporting may have ensured that only one person was

identified as owner, therefore, potentially avoiding paying out an incorrect refund. As discussed in Section 4.3.1,

the introduction of DAC5 has to a large extent addressed concerns around due diligence and beneficial

ownership. Finally, although information on capital gains tax is not exchanged under the DAC provisions, DAC2

requires the reporting of information on gross proceeds from the sale of financial assets in relation to custodial

accounts, which would have identified amounts relating to the transactions within the cum-cum and cum-ex

schemes.

As can be seen above, a strengthened framework for EOI could have identified these schemes, and is likely to

do so to a large extent in future. The factors discussed above are further considered throughout this Research

Paper in more detail and further suggestions for improvements are provided in relation to them where relevant

in Section 5.2. Nevertheless, an essential factor in the non-d

etection of these schemes was the lack of

cooperation across Member States beyond the minimum requirements. Although amendments to the initial

DAC1 have strengthened cooperation and enforced mandatory EOI requirements, these require Member State

collaboration and investment. Reinforcing this would improve the framework for EOI, and could have alerted

authorities across EU Member States about the schemes early on once identified in one Member State, even if

the information exchanged under the provisions was not sufficient.

As set out above, an improved framework for EOI is likely to identify these schemes and help Member States

tackle the related tax avoidance. It could also have had a deterrence effect to some extent, avoiding the

emergence of these schemes to begin with. However, within the framework at the time of the original scheme

there were already some measures in place that would have allowed Member States to cooperate sufficiently

to tackle the schemes effectively, although these measures were underused. Therefore, it will be critical to the

success of the DAC provisions for Member States to cooperate more closely and use the provisions to the largest

extent possible to identify these and other similar types of avoidance schemes.

The bilateral Austria - Cyprus case

15 A special bilateral case has been made possible through the double taxation agreement

between Austria and Cyprus. In this instance, company profits distributed by private

companies in Cyprus are classified as distributed benefits which Cyprus has the right to tax,

but in practice does not do so for company owners who are not residents in Cyprus. A

complicated three-company structure is required to set up this scheme which, ultimately,

European Parliament, 2018.

EPRS | European Parliamentary Research Service

taxes corporate profits in Cyprus but allows profit distribution to company owners to remain

untaxed both in Cyprus and Austria.

Other non-financial assets

16 As mentioned above, some types of FIs and financial accounts are not covered within the

scope of DAC2. This is also the case in relation to certain types of non-financial assets such as

cash, art, gold or other valuables held at free ports or safe deposit boxes, ownership of real

estate, yachts, horses, etc.

. Information on these assets will not be exchanged and, given

these assets are not considered as ‘financial’ assets, any investment entities that directly hold

investment in e.g. real estate will not be considered to be a reporting FI. Therefore, there may

be an incentive to hold hard, non-financial assets over financial assets that may be reported

under the DAC2 provisions. Nevertheless, ownership and income from immovable property

is covered under DAC1 as long as the information is readily available.

As can be seen from the above, there are a number of circumventions and loopholes which can be

identified in relation to DAC2. A number of legal and practical obstacles are also present, and these

are discussed at Section 2.2.

DAC3 circumventions and loopholes

Focusing on DAC3, the following circumventions have been identified in the literature and through

interviews with Member States and all share the common characteristic of primarily being a

circumvention or loophole by design of the legislation.

1 The provisions under DAC3 cover rulings issued, amended or renewed from 1 January 2017

and those from 1 January 2012 to 31 December 2016 if still applicable at 1 January 2014. This

means that any rulings issued, amended or renewed before 2012 are not covered, even if

they still apply. Given that information on these rulings is not publicly available or

exchanged, it is not possible to ascertain the extent of the loophole and loss of revenue from

not including pre-2012 rulings. This may raise questions over the fairness of the provisions,

given then corporates with older (pre-2012) CBTRs or APAs that still apply will benefit from

them not being exchanged compared to those with more recent ones. However, this needs

to be balanced with the practicality of needing to set a cut-off date, in order to avoid an

excessive administrative burden of complying with DAC3.

2 Although tax rulings and APAs are most commonly associated with corporates, they are also

used by high net worth individuals (HNWIs) who have the ability to enter into complex tax

arrangements. Under Paragraph 4 of Article 8a, the provisions of the Directive do not apply

to tax rulings which concern exclusively natural persons (individuals). The implications of this

are that HNWIs are able to agree advantageous arrangements in one Member State without

other relevant Member States being aware, which may result in a loss of tax for the Member

States.

3 The provisions only concern agreements or transactions which are considered to be cross-

border in nature, and do not cover other arrangements of a non-cross border nature which

benefit the corporate. For example, where an advantageous domestic agreement is obtained

by a corporate, although not cross-border in nature, it may have taxation implications for

another Member State. Under the current DAC1 provisions, the Member State where the

domestic agreement is made could spontaneously exchange information of the agreement

on the basis of foreseeable relevance. However, this is subjective and the provisions are not

mandatory or commonly used

, so there remains a possibility that the details of domestic

See Tax Free Today, ‘Living tax free in Austria through corporate structure in Cyprus’, 23 October 2019.

See Knobel, October 2018.

See European Commission, September 2019.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

agreements, which could be of interest to authorities in another Member State, are not

shared.

4 Tax rulings and pricing agreements falling under the scope of the Directive must be

‘advance’ rulings or agreements. That is, they must be made before the transaction takes

place, activities that create a permanent establishment take place, or ahead of the filing of

returns. In practice, many agreements are made post-transaction and after filing returns,

where, for example, the corporate concerned is subject to a tax audit. These agreements

would not be covered by DAC3 and could create a perverse incentive to not obtain

agreements pre-transaction in order to avoid exchange of information under DAC3.

5 Paragraph 2 of Article 8a allows for non-financial or investment corporate entitie s w ith

group-wide annual net turnover of less than €40 million to be excluded from the DAC3

provisions. Although this is a loophole, it is difficult to assess the scale of it due to potential

differences in tax-driven behaviour between larger and smaller firms with varying access to

capital and structures that would allow the behaviours to take place. Although not all firms,

large or small, will engage in cross-border transactions, in an increasingly globalised context

it is becoming a much more important source of business. However, the majority of

businesses in the EU fall under the DAC3 exemption threshold (almost 99 % of EU

businesses

are classed as small or micro). Nevertheless, not all of those that are below the

threshold will actually engage in cross-border activity and, in addition, a small number will

enter into ACTRs or APAs. Therefore, the extent to which this loophole affects the

effectiveness of the DAC provisions may be limited.

6 Under Paragraph 3 of Article 8a, bilateral and multilateral APAs are excluded from the

provisions under DAC3 where the related international tax agreement does not allow its

disclosure. Instead, the Member State concerned in the APA will be able to solely exchange

information in relation to the issuance of the APA, rather than anything contained within it.

This means that Member States may not be aware of practices taking place that may affect

them or lead to a tax loss in their jurisdiction.

The above represents a comprehensive understanding of the circumventions and loopholes for

DAC3. The desk-based research combined with the primary data collection (survey and interviews)

identified a number of practical and legal obstacles for DAC3 which are explored in greater detail

below in Pillar 2.

DAC4 circumventions and loopholes

In a similar manner to the circumventions and loopholes observed for DAC3, those identified for

DAC4 have primarily arisen as a result of the way the provisions have been designed. Despite this

commonality, the actual number of circumventions and loopholes identified through the literature

and the stakeholder consultation is limited:

1 Excluded MNE groups are not subject to reporting requirements under DAC4. Paragraph 4

of Section I of Annex III of the Directive defines excluded MNE groups as being those with

total consolidated group revenue of less than €750 million. This exclusion reflects the aim of

the provisions to not place extra administrative burden on small firms through additional

reporting requirements and recognises the relatively higher tax burden of SMEs. Regardless,

this threshold immediately takes the majority of small scale MNE groups out of scope,

although given their size the fiscal impact of the loophole is likely to be small.

See European Commission, September 2019.

See European Commission, Annual Report on European SMEs 2018/2019 - Research & Development and Innovation by

SMEs, SME Performance Review 2018/2019, Report prepared by LE Europe in collaboration with DIW Econ, PwC

Luxembourg and CARSA, November 2019.

EPRS | European Parliamentary Research Service

2 Article 25a of the Directive provides for national discretion in regard to compliance by MNE

groups with the provisions, suggesting that ‘effective, proportionate and dissuasive’

penalties should be imposed. Similar to challenges surrounding enforcement and sanctions

under DAC2, there is a lack of consistency across the levels of penalties across Member States,

which may affect levels of compliance, in particular when the MNE group has to report in

various Member States. The incentive to comply may be impacted in Member States where

the sanctions are limited in comparison.

3 Interviews with Member States noted that certain business structures are not covered within

the scope of DAC4 such as, for example, single enterprises with one or more foreign

permanent establishments, and enterprises that are under the common control of one or

more individuals.

Research and primary data tools were not able to identify any further issues beyond those described

above. This is in line with expectations given the small body of literature covering the provisions

under DAC4. This is likely a reflection of reporting requirements having only been in place for two

years. However, a range of practical obstacles were identified through the stakeholder survey and

interviews with Member States, which are explored in further detail below in Pillar 2.

2.2. Pillar 2: Obstacles in the exchange of information

Even in a hypothetical environment where market opportunities to circumvent DAC were not

present, there is an array of obstacles that could potentially limit its effectiveness. In the paragraphs

below, a distinction has been made between legal and practical obstacles. On the one hand, legal

obstacles could consist in conflicting obligations arising from other directives or legislation that

supersede the DAC or, indeed, from limitations expressed within the DAC itself. On the other hand,

practical limitations, such as the lack of available information or the timeliness of exchanges might

be limiting effectiveness due to impediments to information exchange itself or the ultimate use of

the information.

2.2.1. Legal obstacles

The literature review and interviews with Member States highlighted various types of legal obstacles

that can impede DAC’s effectiveness: the purpose of the exchanges, limitations regarding business

secrets, public disclosure, requirement of legislative changes, legal interpretation, scope for legal

challenges, and other challenges.

Purpose of the exchanges

The purpose of DAC includes improving and enhancing transparency in order to support the aim of

tackling tax fraud and evasion. However, there are broader areas linked to these aims, such as illicit

financing and activities, where the information collected under DAC1 to DAC4 could assist and

would be of wider benefit. Under Paragraph 1 of Article 16 of the Directive, the information

exchanged may only be used for tax purposes, however Paragraph 2 allows for the use of the

information for other purposes if the Member State sending the information provides authorisation

for this, and where the information could be used for similar purposes in the authorising Member

State. This poses two challenges in relation to the legal provisions: not only does this rely on the

sending Member State to provide authorisation voluntarily but, as noted in interviews with Member

States, as there is no time limit for providing authorisation or responding to the request, the

information may not be accessible in a timely manner that supports the aims of the provision.

Business secrets

The provisions under DAC1 to DAC4 do not require Member States to exchange information when

business secrets are involved. This is reflected in Article 17(4) of the Directive and specifically covers

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

cases where exchange of information ‘would lead to the disclosure of a commercial, industrial or

professional secret or of a commercial process, or of information whose disclosure would be

contrary to public policy’.

It should be noted that this only encompasses business secrets and not

banking secrets. Indeed, it could be argued that safeguarding the former upholds freedom to

conduct business whereas protecting the latter facilitates tax evasion.

Public disclosure

As noted earlier, one of the key objectives of DAC is to improve transparency in relation to tax

matters, with one of the primary methods of achieving this being the public disclosure of

information in relation to taxation matters. However, the legal provisions under DAC do not make

public disclosure of certain types of information mandatory. Rather, they focus on information being

accessible between Member States’ CAs only.

Focusing on DAC3, there is no public access to ACBRs or APAs or a summary of their content, thereby

limiting the transparency surrounding these agreements. The issue of public disclosure is largely

replicated in DAC4 also, where there are no public disclosure requirements at present. This lack of

transparency is viewed as a legal obstacle. In response, there are proposals being discussed by the

European Parliament and European Commission which would enable public disclosure, primarily

driven by public interest in the issue. However, these proposals have been under discussion for over

3 years, with progress impeded by a lack of agreement between Member States.

Legislative changes

Certain Member States identified the need to engage in extensive processes to amend national

legislation causing them to miss reporting deadlines. This process took a number years in some

cases but was necessary to ensure the availability of data required for exchanges under DAC1.

This

issue continued with the DAC2-4 amendments. Indeed, the recent EC evaluation of DAC highlighted

as an example that Belgium adopted six measures between 2015 and 2017 in order to transpose

DAC2, while Bulgaria achieved the same with one measure in 2015.

Legal interpretation

Interviews with Member States highlighted that there are challenges surrounding the interpretation

of DAC provisions, and definitional issues. With regards to the former, one Member State indicated,

in particular, that they faced a conflict in how they legally interpreted DAC3 provisions in

comparison to national legislation. They gave two examples. The first, under national measures, the

interpretation of when the issuing of an APA took place could be both the moment of mutual

signing, as well as the entry into force of the agreement. The second example highlighted the

difference in the definition of an APA between the national legislation and DAC3.

Article 17(4) of Council directive 2011/16/EU. The business secrets limitation is referred to in Marcel Schaper, ‘Data

Protection Rights and Tax Information Exchange in the European Union: An Uneasy Combination‘, Maastricht Journal

of European and Comparative Law, June 2016 and Roman Seer and Sascha Kargitta, Exchange of information and

cooperation in direct taxation, Research Handbook on European Union Taxation Law, 2020, p. 504.

Eurodad, ‘EU Country by country reporting: Overview of the political process and existing country by country reporting,

Financial Transparency Coalition’, Briefing paper, December 2019.

See 18th Meeting of the Sub-Group on the Automatic Exchange of Information, Minutes of the Working Group (and its

sub-groups) of Administrative Cooperation in the field of Direct Taxation, Register of Commission Expert Groups,

2020.

See Economisti Associati, April 2019.

EPRS | European Parliamentary Research Service

In terms of the latter, a small number of Member States interviewed suggested that the

interpretation of definitions such as ‘turnover’ and ‘tax’ could prove challenging, as these can differ

between Member States. However, this does not seem to be a significant issue overall.

Legal challenges

The literature has highlighted the scope for legal challenges against the DAC provisions. The

potential to mount legal challenges can negatively impact effectiveness, as it can prevent certain

types of exchanges either through a direct injunction or through entangling the concerned parties

in lengthy legal proceedings.

The grounds for legal challenge primarily revolve around the proportionality principle which

requires the processing of personal data to not go beyond what is appropriate and necessary to fulfil

its legitimate purpose. Where DAC fails to present objective, clear and precise criteria that warrant

restriction on subject’s data rights, then violation of the proportionality principle can offer scope for

legal challenges.

This risk has been flagged by the EC’s Expert Group on Automatic Exchange of

Financial Information (AEFI) who refer to examples such as reporting sales proceeds independent of

whether the receiving country already taxes the underlying income. According to the Expert Group,

beyond the potential for unnecessary data exchange, DAC2, as it stood in 2014, did not offer

appropriate rights to data subjects. There was neither a permanent access right to their information,

nor a provision to receive notifications on potential breaches. Lastly, the group noted that the EOI

did not require any indications of non-compliant taxpayer behaviour that would grant sufficient

cause to the requests and the exchange.

Other challenges

There are also additional challenges, although less adversarial in nature, that may restrict the

Directive’s effectiveness. As far as SEOI is concerned, DAC1 lists five circumstances in which the

competent authority of a Member State must communicate the information.

Establishing whether

these circumstances are met is, in itself, a significant undertaking for a tax official of a different

jurisdiction that is not working on a specific request.

Similarly, Member States are not required to provide information related to requests failing the

standard of ‘foreseeable relevance’, under EOIR. As noted from interviews with Member States,

meeting this standard to obtain information can be difficult as, to a large extent, this can be

subjective. In addition, SEOI is only specified when the information is relevant to a possible loss of

or increase in tax liabilities in the potential recipient country, and voluntary in every other case,

thereby naturally limiting the effectiveness of DAC. In both circumstances, the required standards

are challenging and subjective to apply and can drive a wedge between intended and achieved

outcomes.

See Schaper, 2016.

See First Report of the Commission AEFI expert group on the implementation of Directive 2014/107/EU for automatic

exchange of financial account information, European Commission, 2015 and Schaper, 2016.

These five circumstances are: ‘a. the competent authority of one Member State has grounds for supposing that there

may be a loss of tax in the other Member State; b. a person liable to tax obtains a reduction in, or an exemption from,

tax in one Member State which would give rise to an increase in tax or to liability to tax in the other Member State; c.

business dealings between a person liable to tax in one Member State and a person liable to tax in the other Member

State are conducted through one or more countries in such a way that a saving in tax may result in one or the other

Member State or in both; d. the competent authority of a Member State has grounds for supposing that a saving of

tax may result from artificial transfers of profits within groups of enterprises; and, e. information forwarded to one

Member State by the competent authority of the other Member State has enabled information to be obtained which

may be relevant in assessing liability to tax in the latter Member State.’ (article 9 of DAC1).

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

2.2.2. Practical obstacles

In addition to legal obstacles, there exists a number of practical obstacles to DAC’s effectiveness

which can impact either the exchange or the processing of information. These have been identified

in the literature and through interviews with Member States. They have been categorised below as

either general practical obstacles to the Directive as whole or the DAC amendment to which it

relates.

DAC1 - Council Directive 2011/16/EU

The practical concerns around DAC1 primarily revolve around the completeness of the data

available across the five income categories. The EC 2019 evaluation report highlighted that there

were only eight Member States with complete information across all five DAC1 income and capital

categories, nine Member States with information on all categories except life insurance products

and ten Member States that only possessed information on three categories. Cyprus had the most

limited information availability covering only the pension category.

The wide disparity between

the availability of information reduces the effectiveness of DAC1 and contributes to information

gaps between Member States.

DAC2 - Council Directive 2014/107/EU

As DAC2 expanded the scope of the Directive to cover FIs, practical issues emerged around the

owners of the accounts for which information would be exchanged. This was especially the case

when looking at joint accounts. The Member State interviewees highlighted that joint accounts

pose additional complexity in this regard as it is difficult to allocate a specific individual taxpayer to

a joint account and ascertain the true income of the joint account holders. The tax administration

(TA) unfortunately does not have accurate information and needs to conduct additional analysis in

order to determine an account holders tax position.

DAC3 - Council Directive (EU) 2015/2376

The research and data collection highlighted two particular areas of concern with DAC3, the first

being around the quality of the summary information provided on the central directory. The

information was considered to be too limited by Member States and required a follow-up

clarification request to be able to fully understand the implications of various rulings. This is likely

due to the fact that there are no guidelines on what TAs should provide as a summary leading them

to create their own, coupled with having to write these summaries in another language, which can

prove challenging for some TAs. As a result, TAs are still having to follow up with a request for more

information in order to get a better understanding of the information on the directory.

In addition, the literature points to an additional potential obstacle to reporting under DAC3. TAs

need to share information on rulings and APA which they have determined to be cross-border in

nature

. In some instances, the sharing of this information could put the TA in a compromising

position if the ruling or APA is particularly favourable to the firm. This could create the perverse

incentive for the TA to not exchange information on these rulings. This could be achieved by the TA

by making use of the discretion the provisions allowed, for example deciding whether or not a ruling

or APA was cross-border in nature.

DAC4 - Council Directive (EU) 2016/881

See European Commission, September 2019.

EPRS | European Parliamentary Research Service

Due to reporting under DAC4 coming into force only recently there have no t been many pr actical

issues identified to date. However, the stakeholder survey highlighted issues with filing reports

under DAC4 and meeting the demand for reporting in multiple Member States. MNEs suggested

that this is due to there being no common format/template for them to follow/use for filing the CbCr

reports. The implications are that given a lack of processing capacity the quality of the reports being

filed has diminished, reducing their usefulness to TAs.

General and cross-amendment practical obstacles - TIN

The availability, or lack, of the required information can be a major obstacle for administrative

cooperation for DAC1 and DAC2. The information exchanged under DAC1 covers several different

types of income, being collected by several different Member States operating in different capacities

and with different processes. There are, therefore, several areas where the type of information

required by a receiving authority is not available in the expected form.

Tax identification numbers (TINs) are important in information exchange as they enable authorities

across different jurisdictions to match income streams with individual records. This can facilitate

consolidation with their own records but also with foreign administrations’ records over time. The

EC evaluation identified the lack of a TIN associated with receiving country records for the vast

majority of taxpayers for which information was exchanged.

100

) It appears that only Lithuania and

Ireland included a TIN, as recognised by the receiving country, in the information they shared

101

Interviews with Member States have confirmed that a lack of TIN is impeding efficient information

exchange and processing.

Focusing on DAC2 in particular, a TIN is not required to be reported unless this information is

collected in the ordinary course of business. The EC reported however, that a large percentage of

accounts held by natural persons or legal persons were associated with a TIN (70 % and 73 %

respectively).

102

Although these numbers seem significant, this is only applicable to financial

account information, and does not guarantee that the TINs recorded are the correct ones, given that

they are not verified. This has implications for the use of data by Member States as at least 27 % of

the information exchanged requires additional work in order to be made usable. To address this gap

all the tax administrations covered by the interviews use the name, date of birth and address to

match information received under DAC1-2 with their own records. Due to the fact that names may

not always be recorded identically, some form of fuzzy matching is used, and match rates are

reported to be generally in the 90 %+ range.

Availability of information can be further explored by examining the justification provided by

Member States receiving a request for information when they are unable to provide the requested

information. The most common justifications related to the taxpayer being unknown, unattainable

or no longer existent - in all three cases a mutually acceptable TIN could contribute to clarity and

transparency.

103

General and cross-amendment practical obstacles - quality of reporting

100

See European Commission, September 2019.

101

See European Commission, Report from the Commission to the European Parliament and the Council on overview and

assessment of the statistics and information on the automatic exchanges in the field of direct taxation, COM(2018)

844 final, December 2018.

102

See European Commission, December 2018

103

See European Commission, Report from the Commission to the European Parliament and the Council on the application

of Council Directive (EU) 2011/16/EU on administrative cooperation in the field of direct taxation, COM(2017) 781 final,

December 2017 and Economisti Associati, April 2019.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

High quality reporting standards, across all types of exchanges, are essential to ensuring that

receiving Member States can effectively process the data in line with their intended use. Low quality

reporting can increase the resources required to process data and can lead to significant process

delays.

There is limited evidence in the European Commission’s evaluation of DAC that significant issues

with the quality of reporting exist. The 2014 and 2015 waves of the Questionnaires on the

Functioning of the Directive (QFD) suggested that feedback was broadly positive, with limited SEOI

exceptions where non-standard formats were received. Limited concerns were also raised, by one

respondent in the yearly assessment, in relation to the level of detail received by some Member

States sharing information on tax rulings, as outlined under DAC3. The majority of interviewed

Member States, however, identified some concerns over the quality of EOIR, indicating that the

receiving Member State requested additional background information to supplement the original

request.

General and cross-amendment practical obstacles - Timeliness

As highlighted previously in the case study on the cum-ex and cum-cum schemes, the reporting lag

for AEOI under DAC1-4 means identifying similar schemes and emerging issues / threats in a timely

manner is not practically possible. As a proactive approach is not possible, TAs are always playing

catch up and reacting to problems. This issue is further compounded by the fact that, pre-Covid, a

significant proportion of Member States were reporting after the deadline set in the provisions in

DAC1 to DAC4. For example, 45 % of Member States reported late in 2017.

104

Concerns over the timeliness of responses to information requests have been raised in the EC

reports.

105

Delays have been observed both in receiving information requested but also in providing

mandatory feedback linked to AEOI.

106

The reporting lag is not exclusive to AEOI, as it is also the case for EOIR under DAC, albeit to a lesser

extent. While AEOI has a year lag between reporting periods, Member States must respond within

two months if they have the information available or six months if the information needs to be

retrieved. The scale of this practical issue could be small as only one tax official interviewed reported

having encountered delays, but not on the initial response but any follow-up they have requested

as there are no deadlines for these per the Directive.

General and cross-amendment practical obstacles - IT / Technical

Volumes of data exchanged under DAC1-4 have been increasing both in quantity and in complexity

as additional types of income and information are being added. The EC’s 2017 report highlighted

that, in contrast to the growth in data, however, the tax administrations’ capacity to process and

analyse the data received has not followed suit.

107

This can be a particular issue for smaller Member

104

See European Commission. Evaluation of the Council Directive 2011/16/EU on administrative cooperation in the field

of taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final, September 2019.

105

See European Commission, Staff Working Document on the application of Council Directive (EU) no 2011/16/EU on

administrative cooperation in the field of direct taxation, SWD(2017) 462 final, December 2017 and European

Commission, December 2017.

106

See Minutes of the 18th meeting of the Sub-Group on the Automatic Exchange of Information Register, Working Group

(and its sub-groups) of Administrative Cooperation in the field of Direct Taxation, Register of Commission Expert

Groups, 2020.

107

European Commission, December 2017.

EPRS | European Parliamentary Research Service

States that have more limited resources at their disposal as they may struggle to make full use of the

information exchanged.

Member States have also reported a range of technical issues with the uploading and downloading

of data both insofar as exchanges under DAC2 and 3 and international exchanges are concerned

108

Two Member States also highlighted that they have experienced technical issues with the interface

between national and EU systems which can slow down or, in extreme cases, halt the process of

data matching, albeit temporarily.

General and cross-amendment practical obstacles - Compliance costs

Taxpayers, FIs, and TAs face high compliance costs as a result of DAC1-4. DAC2 in particular carries

a high compliance burden, and this is most heavily felt by FIs. As highlighted in the 2019 evaluation,

FIs from five Member States were estimated to have had initial compliance costs of approximately

€340 000 000 as a result of the procedures required for DAC2. Compliance costs are also borne by

TAs, although to a lesser extent. In comparison to FI costs for DAC2, all Member States spent an

estimated €118 000 000 for DAC1-3. Although significantly lower, this still remains a substantial

sum. Compliance costs can therefore be considered a practical obstacle, in particular if this affects

the ability or incentive to comply with the provisions.

109

As DAC has evolved and expanded over time it is unsurprising to see that practical issues have

emerged that have impacted the effectiveness of the Directive. Many were unforeseen, while others

such as IT/technical issues can be expected given the scale of infrastructure needed at both the EU

and Member State level. Despite these challenges the EP, EC and Member States continue to make

progress towards tackling issues through further amendments to the Directive. Section 5.2 sets out

suggestions based on the research conducted and data gathered to contribute towards this process

of improving the effectiveness of DAC going forward.

2.3. Pillar 3: Information exchange, usage and fiscal impacts

The market reactions to the implementation of DAC and its amendments assessed under Pillar 1, as

well as the legal and practical obstacles identified under Pillar 2, are likely to have had an impact on

the exchange and use of information under the Directive. The EC's 2017 report on the DAC’s

application highlighted categories of information that were not used to a substantial extent. These

included employment income, directors’ fees, life insurance products, pensions and income related

to immovable property. In line with the analysis of Pillar 2, the lack of capacity and format

inconsistency were flagged as key reasons behind limited usage. Moreover, the report suggested

that the levels of administrative cooperation between Member States did not appear to be in line

with the underlying level of economic activity across Member States.

110

Section 2.3.1. explores the

plausibility of exchanges based on macroeconomic data, in particular in relation to DAC2 financial

account information.

With regards to how CAs use the information received under the DAC provisions, officials from all

Member States interviewed reported using the information to identify tax risks and address these

risks through targeted information campaigns to ensure increased ex-ante compliance, and to

check and audit tax filings from their residents and increase ex-post compliance by requiring

corrections to tax filings in case of non-declared income and imposing penalties and fines.

108

See Minutes of the 18th meeting of the Sub-Group on the Automatic Exchange of Information Register, 2020.

109

See European Commission, September 2019.

110

See European Commission, December 2017b and European Commission, December 2017.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Almost all Member States reported that knowledge among taxpayers of EOI between tax

administrations has some deterrence effect and increased ex-ante compliance. However, none were

able to quantify the additional tax revenues arising from either increased ex-ante and /or ex-post

compliance. Indeed, attributing amounts of additional revenues to information exchanged under

the Directive can be challenging given that EOI will be one of many tools used in order to increase

revenues and tackle non-compliance.

However, Spain noted that between 2017 and 2019, information for CRS, DAC2 and FATCA yielded

more than €435 million in additional tax revenues from 1 200 taxpayers (Agencia Tributaria, 2020).

111

Moreover, the information from DAC1 yielded more than €55 million from more than 26 000

taxpayers in 2019. In the case of Belgium, according to the Cour des Comptes

112

, the Belgian TAs

estimate that the first use of DAC1 information on employment income and immovable property

yielded more than €300 million in additional tax revenues. In the case of Finland, Tokola

113

reports

that the use of the CRS information to May 2019 yielded €6 million in additional tax revenues, mainly

on unreported capital gains.

2.3.1. Using macroeconomic data to monitor the plausibility of DAC2

information exchanges

DAC2 foresees the automatic exchange of information on the value of a number of financial assets,

the income streams associated with such assets during the reporting period and the value of the

proceeds of the sale of any such assets.

The assets include:

Custodial accounts at Reporting FIs

Depository accounts at Reporting FIs

Other types of accounts at Reporting FIs

Any Cash Value Insurance Contract and any Annuity Contract issued or maintained by a Financial

Institution (other than a non-investment-linked, non-transferable immediate life annuity that is

issued to an individual and monetises a pension or disability benefit provided under an account

that is an Excluded Account).

Reporting FIs include custodial Institutions, depository institutions, investment entities or specified

insurance companies.

Exceptions/exclusions are foreseen by DAC2 in terms of both some types of accounts for which

information has to be reported by Reporting FIs and the set of Reporting FIs. These exceptions and

exclusions are not discussed in the present section as the granularity of the macro-economic

information that is reviewed below is much lower than the range of financial assets, FIs and account

holders defined by the Directive.

Four types of macro-economic data sets were identified which a priori could be useful for deriving

high level estimates of cross-border financial income flows or financial assets ownership between

residents of Member States and between residents of Member States and other countries which

follow the OECD Standard for Automatic Exchange of Financial Account Information.

111

See Agencia Tributaria, Principales Resultados 2019, Resultados económicos y tributarios en el Impuesto sobre el Valor

Añadido, 2020. The latest figures show €293 million in 2019 from 699 taxpayers.

112

See Cour des Comptes, Échange automatique de données fiscales au niveau international. Premier rapport d’évaluation

à la demande de la commission Panama, December 2019.

113

See Antti Tokola, ‘Tax evasion in international investment – Messages from a Finnish perspective’, Tax Administration,

Grey Finance Clearing House, September 2019.

EPRS | European Parliamentary Research Service

These four data sets include:

1 the BIS locational banking statistics

2 the IMF Coordinated Portfolio Investment Survey (CPIS) data

3 the ECB data on the domestic and cross-border positions of euro area MFIs (excluding the

Eurosystem) and EU-27 non-euro area MFIs (excluding national central banks)

4 the Eurostat balance of payments statistics of individual Member States

In order to assess the usefulness of these datasets for deriving high-level quantitative estimates of

cross-border ownership of financial assets and related income flows, a number of criteria were used,

namely:

1 Are the data covering cross-border ownership of financial assets and related cross-border

income flows?

2 Are the cross-border asset ownership data available for pairs of Member States?

3 Do the data cover a wide range of financial assets owned cross-border?

4 Do the data cover cross-border flows of income associated with the financial assets?

5 Are the cross-border income data available for pairs of Member States?

6 Are the cross-border data published at a frequency that would make them potentially useful

for high-level validation of the DAC2 exchanges?

7 Are the cross-border data available on a timely basis, i.e., not too long after the end of the

period to which the data refer to?

A summary overview of the assessment is shown in the table below and detailed comments are

provided thereafter.

Table 4: Summary of the assessment of the four potential databases

Test Databases

Criterion

BIS locational

banking

IMF CPIS

ECB MFI cross-

border positions

Eurostat –

balance of

payment of

Member States

Are the data covering cross-

border ownership financial

assets and related cross-

border income flows?

Are the cross-border asset

ownership data available for

pairs of Member States?

Yes

Partially

Not for within

EU cross-

ownership

Do the data cover a wide

range of financial assets

owned cross-border?

Not for within

EU cross-

ownership

Do the data cover cross-

border flows of income

associated with the financial

assets

To some extent

Are the cross-border income

data available for pairs of

Member States?

To some extent

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Test Databases

Are the cross-border data

published at a frequency that

would make them potentially

useful for high-level

validation of the DAC2

exchanges?

Yes

Are the cross-border data

available on a timely basis,

i.e., not too long after the end

of the period to which the

data refer to?

Yes

Source: BIS, IMF, ECB and Eurostat

Both the BIS and ECB data sets cover only the cross-border liabilities of banks (BIS) or cross-border

deposits at monetary FIs (ECB).

Moreover, in the case of the BIS data, this information is provided only for the aggregate exposure

to the world of the banks of a reporting country.

In the case of the ECB data, only the cross-border exposure to the euro-zone residents is reported

(separately for MFIs, households and non-financial corporations). For some Member States, this

exposure is further broken down by Member State (for both euro-zone and non-eurozone Member

States).

The IMF data provide information on within and extra-EU cross-border investment in equity and

investment funds by households and non-financial corporations of each Member State.

The Eurostat Balance of Payments (BoP) data provides information on economy-wide cross-border

ownership of a wider range of financial assets, but only for cross-ownership between residents of a

Member State and the non-EU area. No data are available for within EU cross-border ownership of

various types of financial assets held by the economy as a whole or individuals or non-financial

corporations.

With regards to income accruing to owners of cross-border financial assets, bilateral income flow

data is available from the Eurostat BoP data for 18 EU-27 Member States (Belgium, Croatia, Cyprus,

Czechia, Estonia, Finland, Germany, Ireland, Italy, Latvia, Lithuania, Luxembourg, Netherlands,

Poland, Romania, Slovakia, Slovenia and Sweden) at the level of the total economy for total portfolio

investment income, investment income on equity and fund shares and interest on debt securities.

Of the three items, only the investment income on equity and fund shares may be useful for

assessing the plausibility of part of the information exchanged under DAC2.

In short, a partial plausibility assessment could be tested using the ECB MFI, IMF and Eurostat

information. The first two sources could be used for a plausibility assessment of the value of cross-

border deposit and asset ownership while the last source could be used for a plausibility assessment

of reported incomes associated with the financial assets owned cross-border.

EPRS | European Parliamentary Research Service

However, in light of the broad sectoral coverage of the data and the lack of complete information

on both cross-border financial asset ownership and bilateral asset ownership and income, such a

validation exercise may not yield robust information.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

3. Quantitative analysis of the DAC information

This chapter reviews the available information on the evolution of the volume and value of

information exchanged between the tax administrations under DAC. Next, it explores whether the

data on cross-border financial asset holdings show any change in cross-border holdings of financial

assets the time when DAC2 was adopted or the DAC2 information exchanges became effective.

3.1. Evolution of the information exchanges under DAC

It is not possible to provide a comprehensive overview up to 2019 of the evolution of the number

and value of the various types of DAC information exchanges between the Member States’ TAs as

Member States did not give permission to the project team to access the relevant data held by

European Commission (DG TAXUD).

However, data to 2017 or 2018 from the Economisti Associati study prepared for the European

Commission

114

are presented below together with some partial, more up-to-date, information

which could be gathered from responses by governments to questions on DAC raised by members

of national parliaments and other documents. Such information was kindly provided by the national

parliaments of Austria, Belgium, Finland, Germany, Luxembourg and Portugal.

3.1.1. Key findings from the Economisti Associati study

DAC1 AEOI

The data reported in the Economisti Associati study

115

show that the number of taxpayers for which

DAC1 information (i.e. information on income from employment, director’s fees, life insurance

products not covered by other Union legal instruments on exchange of information and other

similar measures, pensions and ownership of and income from immovable property) was

exchanged automatically between Member States rose from about 3.5 million in 2015 to about 7.5

million in 2016 and 5 million in the first half of 2017. The period refers to the calendar year in which

the messages were sent and not the tax year to which the information refers to. The value of the

items covered by the information exchanged automatically rose from about €27 billion in 2015 to

slightly more than €50 billion in 2016 and was marginally above €40 billion in 2017. Overall, 11 000

messages concerning about 16 million taxpayers and incomes/assets worth about €120 billion had

been exchanged from the beginning of DAC1 exchanges in 2015 to mid-2017. Among the five DAC1

items for which information is exchanged automatically, pension income was by far the most

important in terms of the number of taxpayers for which information was exchanged automatically

while employment income was the most important in terms of the value of the information

exchanged.

DAC1 specifies that information should be exchanged automatically between the TAs of Member

States for five types of incomes and assets if the information is available. By 2017, only eight Member

States were sending information regarding all five DAC1 items and another ten were sending

information for all DAC1 items expect for life insurance products.

114

See Economisti Associati, 2019.

115

See Economisti Associati, 2019, p.9.

EPRS | European Parliamentary Research Service

Figure 3: Number of Member States with information available in 2017 by DAC1 category

Source: European Commission, Directorate-General Taxation and Customs Union, Evaluation of

Administrative Cooperation in Direct Taxation Final Report prepared by Economisti Associati in collaboration

with ECOPA and Oxford Consulting, April 2019

With regards to the sending messages, France and Germany were in 2016 by far the two Member

States whose messages concerned the largest number of taxpayers (mainly pension recipients) and

Luxembourg and Germany were the two Member States with the highest value of information

exchanged (almost all related to employment income in the case of Luxembourg reflecting the large

share of cross-border commuters in its labour force and about equally distributed between

employment and pension income in the case of Germany).

The Member States having received messages concerning the largest number of taxpayers were

France, Germany and Spain in 2016 and in terms of the value of the information received France and

Germany were the largest recipients of messages.

Regarding the number of taxpayers, the largest bilateral flows in 2016 were Belgium->France,

France->Spain and France->Portugal, Germany->Italy, Germany->Spain, France->Spain and France-

>Portugal. The picture of the bilateral flows in terms of the value of information exchanged is very

different with the most important bilateral flows in 2016 being Luxembourg->Belgium,

Luxembourg->Germany, Luxembourg->France, Denmark->Sweden and France->Spain.

DAC2 AEOI

The Economisti Associati study provides only very limited information on DAC2 AEOI covering the

period September 2017 to March 2018 for most sending Member States. About 4 000 messages

were exchanged during this period covering some 8.3 million accounts. In terms of value, the value

of the end-of-year account balance information was by far the largest in these exchanges.

116

totalled €2 865 billion. The second largest item in value terms was information on redemption or

sales of financial assets (€850 billion). The value of the exchanged information on other unspecified

payments stood at €59.4 billion while that of interest income and dividend income was much lower,

€18.2 billion and €14.8 billion respectively.

During the period September 2017 – March 2018, Luxembourg was the Member State with the

largest number of accounts covered by the messages sent (somewhat less than 1.5 million) and was

followed closely by Germany (slightly less than 1.4 million accounts). In terms of the value of the

information exchanged, Luxembourg’s overall value in the messages sent stood at about

€2.4 billion, more than the value in the information sent by all the other Member States. Overall,

116

As noted by Economisti Associati, April 2019, the deadline for the due diligence of low value (i.e. with a value equivalent

to less than 1 billion US$) was 31 December 2017, the volume and value of information exchanged was likely to

increase in subsequent years. The due diligence deadline for high value accounts was 31 December 2016.

Employment

income

Pension Income Immovable

property

Directors' fees Life insurance

products

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Luxembourg accounted for 18 % of the accounts and 85 % of the amounts reported in messages

sent under the DAC2 AEOI.

During this period, Belgium, Sweden and the UK are by far the EU-28 Member States having received

DAC2 AEOI messages with the greatest value while Germany, Italy, France and the UK are the EU-28

Member States having received information covering the largest numbers of accounts.

The largest five bilateral DAC2 AEOI flows in terms of value originated in Luxembourg: Luxembourg-

> Belgium (€607 billion), Luxembourg-> Sweden (€538 billion), Luxembourg-> United Kingdom

(€400 billion), Luxembourg-> France (€215 billion EUR) and Luxembourg-> Denmark (€105 billion).

These five Member States accounted for about ¾ of the total value of the information sent by

Luxembourg during the period September 2017-March 2018. Moreover, during this period.

Luxembourg was also the source of 9 out of the 10 largest bilateral DAC2 AEOI flows.

In terms of bilateral DAC2 AEOI flows in the number of accounts, Luxembourg was the source of

three of the top five information flows (Luxembourg->Germany, 647 316 accounts, Luxembourg-

>Belgium, 304 729 accounts, Luxembourg->France, 231 440 accounts) while Germany and Ireland

accounted each for one of the top five bilateral flows (Ireland->Italy, 617 466 accounts and Germany

-> Austria, 287 354 accounts).

Other important bilateral flows which are highlighted in the Economisti Associati study include the

Ireland-> United Kingdom and the Czechia-> Slovakia flows.

DAC3 AEOI

The exchange of information became mandatory in June 2017. As of 1 February 2018, according to

the Economisti Associati study, ‘Member States had sent information on 17 652 ATR/APA, of which

11 547 related to tax rulings in force and 6 195 on tax rulings discontinued before the start of the

mandatory exchange of information’. The Netherlands and Luxembourg provided the largest

number of rulings (nearly 9 000 in the case of the Netherlands and 5 000 in the case of Luxembourg).

The UK was a distant third with slightly more than 1 200 rulings and all the other Member States

contributing together only 15 % of the total number of rulings.

The mandatory exchange of ATR/APA information resulted in a very large increase in reported

ATR/APA in 2017. Before the entry into force of the mandatory exchange, information was shared

between Member States on a spontaneous basis. Overall, the number shot up to 17 652 ATR/APA

reported in 2017 from only 2 529 in 2016, 113 in 2015 and 11 in 2014.

Exchange of information request and spontaneous exchange of information

under DAC

The number of requests for information sent out by Member States stood at 9 239 in 2017, and, on

average over the period 2013-2017, this figure was 57 % higher than, on average, over the two years

preceding the introduction of DAC. France, Germany and Poland were the three Member States

having sent the most requests over this period. Together they accounted for about half of all

requests sent by Member States over this period.

In terms of the bilateral flows of such requests, four of the top five related over the period 2013-2017

to requests between three Member States, namely Germany, Netherlands and Poland (Poland->

Germany, 3 061 requests, Germany-> Poland, 1 929 requests, Netherlands-> Poland, 1 408

EPRS | European Parliamentary Research Service

requests

117

and Netherlands-> Germany, 1 357 requests). The fifth most important bilateral flow was

from France to Luxembourg (1 536 requests). More generally, the bulk of the bilateral flows occurs

between neighbouring countries.

Member States can also send spontaneously tax information to the TAs if they are of the opinion

that the information may be of interested to the receiving authorities. The number of such

spontaneous exchange of information under DAC fluctuated between 10 860 in 2014 and 79 013 in

2017.

118

In contrast, over the five years preceding DAC (i.e. 2008 to 2012), the number of

spontaneous exchanges of information had grown more or less steadily from 19 349 in 2008 to

29 234 in 2012.

In terms of bilateral flows of spontaneous exchanges of information, the information sent from the

Netherlands to Germany is the most voluminous, accounting annually from 2013 to 2017 for

between one fourth and over four-fifth of the total number of such exchanges between Member

States.

Information which was sent spontaneously typically concerned tax rulings (prior to DAC3 the

exchange of such information was not automatic), employment and business transactions.

3.1.2. Findings from selected Member States

Publicly available data on DAC exchanges of information at the Member State level is scarce and

very patchy when available. In most cases, the publicly available information is the number of

records exchanged. However, it is important to note that one cannot infer the number of taxpayers

whose information was exchanged under DAC from the number of DAC records sent or received as,

in some cases, information concerning a particular taxpayer may be provided in a number of

different records and, in other cases, a particular record may concern more than one taxpayer.

Austria

Over the period 2014-2016, the Austrian TAs received 255 590 records under the DAC1 AEOI, of

which 77 % originated from Germany. Italy, Sweden and the Netherlands, with respectively 6 %, 3

% and 3 % were distant second, third and fourth in terms of DAC1 records sent to Austria

119

Belgium

A 2020 report published by the Belgian Cour des Comptes

120

shows that the volume of information

received by the Belgian TAs regarding income from employment, director’s fees, life insurance

products not covered by other Union legal instruments on exchange of information and other

similar measures, pensions and ownership of and income from immovable property increased

rapidly from 2015 to 2017 as the systems implemented by Member States to gather and exchange

117

According to Economisti Associati, April 2019, the large number of requests sent by the Netherlands to Poland reflects

mainly requests sent in 2013 as part of actions taken by Dutch TAs related to the so-called Polenfraude scandal

whereby some Polish workers unduly claimed Dutch unemployment benefits.

118

12 470 in 2013, 40 730 in 2014, 10 860 in 2015, 14 860 in 2016 and 79 013 in 2017.

119

Rechnungshof Österreich, Internationaler Informationsaustausch in Steuerangelegenheiten, Bericht des

Rechnungshofes, Reihe BUND 2019/33, August 2019.

120

See Cour des Comptes, Échange automatique de données fiscales au niveau international. Second rapport d’évaluation

à la demande de la commission Panama, November 2020.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

the information bedded down. By 2019, Belgium received 599 297 records regarding DAC1

reportable items.

121

The number of records related to financial accounts and received under DAC or CRS AEOI has grown

even more rapidly. By 2019 the overall volume of such information items received by the Belgian

TAs was almost 2.7 times as large as that of information items not related to financial accounts and

the value of the accounts in the financial records stood at €161.8 billion in 2019.

Table 5: Belgium: Number of records received under AEOI

2015 2016 2017 2018 2019

Records not related to

financial accounts

321 083 453 067 640 344 644 201 599 297

Records related to financial

accounts

- - 521 879 1 404 238 1 599 670

Value of accounts in records

of financial accounts

- - €90.8 billion €173.8 billion €161.8 billion

Source: Cour des Comptes, Échange automatique de données fiscales au niveau international. Premier rapport

d’évaluation à la demande de la commission Panama, November 2020

Finland

No information on DAC specific exchanges of information is available but a recent study notes that

a first phase of information received under the CRS concerned less than 50 000 individual account

holders who were Finnish residents, of which approximately 2 000 were companies. The income

paid to these accounts amounted to approximately €883 million and the reported value of these

accounts amounted to approximately €9.6 billion. A second phase concerned less than 150 000

individual account holders, of which less than 4 500 were companies. The number of accounts was

about 260 000, the financial asset income totalled about €2.5 billion and the value of accounts

amounted to approximately €16.6 billion.

122

Germany

DAC1

Germany sent about 1.8 to 1.9 million DAC1 related records each year from 2014 to 2018 and

received 1.1 to 1.2 million records annually from 2014 to 2016 (Figure 4).

123

121

As noted by the Belgian Cour des Comptes, 2019, the coverage of a single record item varies across sending Member

States. For example, in the case of a person receiving several pensions in a Member State other than the Member

State of tax residence, some Member States will send the information on the different pensions as different records

while others may report the different pensions in a single record.

122

See Tokola, September 2019.

123

See Deutscher Bubdestag, Drucksache 19/13797, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion Die Linke. –

Drucksache 19/12953 – Andauernde Umsetzung des Informationsaustauschs in Steuersachen, Deutscher Bundestag,

8 October 2019.

Data on received records were incomplete for 2017 and 2018 when the data were compiled by the

German Government.

EPRS | European Parliamentary Research Service

Figure 4: Germany: number of DAC1 records sent to other EU Member States and received

from other EU Member States

Note: Data on received records were incomplete for 2017 and 2018 when the data were compiled by the

German Government.

Source: Deutscher Bundestag, Drucksache 19/13797, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion Die Linke. –

Drucksache 19/12953 – Andauernde Umsetzung des Informationsaustauschs in Steuersachen, 8 October 2019

Two thirds to three quarters of records sent by Germany related to pensions and about one quarter

to one third to employment income from 2014 to 2016. In contrast, 42 % to 46 % of the records

received by Germany related to immovable property, 36 % to 38 % to pensions and only 18 % to

employment income (Table 6). Data on received records were incomplete for 2017 and 2018 when

the data were compiled by the German Government.

124

Table 6: Number of DAC1 records received and sent by Germany under AEOI in per cent of

total number of records received / sent

Employment

income

Directors' fees

Life-insurance

products

Pensions

Immovable

property

Messages received

2014 18.0 %

0.3 %

37.4 %

44.1 %

2015

18.1 %

0.3 %

38.9 %

42.4 %

2016

17.6 %

0.3 %

36.1 %

45.6 %

2017

21.4 %

0.4 %

0.2 %

45.1 %

32.9 %

2018

45.8 %

0.9 %

0.6 %

16.4 %

36.3 %

Messages sent

2014

22.5 %

0.0 %

74.4 %

3.0 %

124

Of note is the fact that Germany did not sent any records on directors’ fees and life insurance products during this

period.

1.051.650

1.052.261

1.165.530

846.283

285.361

1.771.931

1.826.369

1.846.740

1.938.273

1.946.708

2014 2015 2016 2017 2018

Records received Records sent

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Employment

income

Directors' fees

Life-insurance

products

Pensions

Immovable

property

2015

25.8 %

0.0 %

71.1 %

3.1 %

2016

26.2 %

0.0 %

71.2 %

2.7 %

2017

30.0 %

0.0 %

68.0 %

2.0 %

2018

32.6 %

0.0 %

67.3 %

0.1 %

Note: Data on received records were incomplete for 2017 and 2018 when the data were compiled by the

German Government.

Source: Deutscher Bundestag, Drucksache 19/13797, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion Die Linke. –

Drucksache 19/12953 – Andauernde Umsetzung des Informationsaustauschs in Steuersachen, 8 October 2019

DAC2

The number of DAC2 records sent and received by Germany dwarfs the number of DAC1 records,

with the German TAs receiving 4.5 million DAC2 records in 20187 and sending 3.3 million such

records in 2018

125

Figure 5: Germany: number of DAC2 records sent to and received from other EU Member

States

Source: Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen - Nachfrage zur

Antwort der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020

Luxembourg and Austria were the most important source countries of the DAC2 records received

by Germany in 2018 with Luxembourg alone accounting for almost 30 % of such records (Table 7).

125

See Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen-Nachfrage zur Antwort

der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020.

1,306,805

3,843,418

4,514,037

1,509,466

2,645,780

3,271,921

2016 2017 2018

Records received Records sent

EPRS | European Parliamentary Research Service

Austria was also one of the most important destination country of records sent by Germany and,

together with France, accounted for almost 40 % of all records sent in 2018 (Table 7).

Table 7: Share (in %) of Member States which account for 5 % or more of DAC2 messages

received and sent by Germany in 2018

2016 2017 2018

Share of DAC2 records received

Luxembourg

52.5 %

35.4 %

29.8 %

Austria

1.3 %

6.4 %

13.4 %

France

5.8 %

7.0 %

7.7 %

Netherlands

7.8 %

6.4 %

7.0 %

Poland

1.8 %

2.9 %

5.9 %

Spain

2.8 %

6.4 %

5.6 %

Share of DAC2 records sent

France

18.7 %

16.8 %

20.6 %

Austria

19.8 %

22.6 %

18.9 %

Italy

6.2 %

9.8 %

Spain

7.9 %

8.3 %

Netherlands

10.3 %

9.2 %

8.3 %

Source: Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen - Nachfrage zur

Antwort der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020

Luxembourg accounted for more than more than half (56.2 %) of the total value of the financial

accounts covered by the records received by Germany from Member States in 2018 (Figure 6).

Austria (13.6 %) and France (10.7 %) were a distant second and third, and only 6 other Member States

(Belgium, Ireland, Netherlands, Poland, Spain and Sweden) accounted each for 1 % or more of the

total value.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

Figure 6: Germany: Share of the total value of the financial accounts records received by

Germany from EU Member States under DAC2 AEOI in 2018- Member States which account

for 1 % of the total value

Source: Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen - Nachfrage zur

Antwort der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020

Luxembourg also sent in 2018 to Germany financial accounts records with by far the highest average

value of € 93 504, 325 % higher than the average value of the records sent by the other EU-27

Member States (Figure 7). France and Ireland were second and third (in terms of the average value

of the financial accounts records) with €68 713 and €67 040 respectively. Fourth and fifth were

Austria and Slovakia with €50 305 and €49 379 respectively. The average value was much smaller in

the other EU-27 Member States and stood at less than €20 000 in 16 Member States (Bulgaria,

Cyprus, Estonia, Finland, Greece, Hungary, Hungary, Italy, Latvia, Lithuania, Malta, Poland, Portugal,

Romania, Slovenia and Spain).

56.2%

13.6%

10.7%

5.0%

3.2%

2.2%

1.5%

1.2%

1.0%

EPRS | European Parliamentary Research Service

100

Figure 7: Germany: Average value (in €) of financial accounts records received by Germany

from EU Member States under DAC2 AEOI in 2018

Source: Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen - Nachfrage zur

Antwort der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020

In terms of the value of the financial records sent, Luxembourg and Austria were also the most

important countries to which Germany sent DAC2 records in 2018, although to a much lesser extent

than in the case of records received by Germany. Together, these two countries accounted in 2018

for 35 % of all DAC2 records sent by Germany and a further 14 Member States (Belgium, Cyprus,

Czechia, Denmark, France, Greece, Hungary, Ireland, Italy, Netherlands, Poland, Portugal, Spain and

Sweden) accounted for at least of 1 % of these records (Figure 8).

Figure 8: Germany: Share of the total value of the financial accounts records sent by

Germany to EU Member States under DAC2 AEOI in 2018- Member States which account for

1 % or more of the total value

93,504

68,713

67,040

50,305

49,379

35,130

34,215

27,175

26,435

20,329

19,700

18,237

14,204

14,154

13,523

11,866

11,807

11,399

10,452

10,293

8,375

7,685

6,493

6,318

5,162

2,206

Luxembourg

France

Ireland

Austria

Slovakia

Netherlands

Belgium

Denmark

Sweden

Czechia

Spain

Estonia

Romania

Slovenia

Bulgaria

Latvia

Portugal

Italy

Finland

Hungary

Poland

Greece

Hungary

Malta

Lithuania

Cyprus

€

21.5%

13.6%

9.8%

9.5%

9.0%

6.9%

5.6%

4.0%

3.8%

2.5%

1.9%

1.7%

1.5%

1.2%

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

101

Source: Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen - Nachfrage zur

Antwort der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020

The average value per DAC2 record sent by Germany in 2018 exceeded €100 000 in the case of three

Member States (Luxembourg, €152 095, Cyprus €115 144 and Malta €105 597). It stood at less than

€39 000 in all other EU-27 Member States except Ireland where the average value was € 57 028

(Figure 9).

Figure 9: Germany: Average value (in €) of financial accounts records sent by Germany to EU

Member States under DAC2 AEOI in 2018

Note: 2017 data for Bulgaria

Source: Deutscher Bundestag, Drucksache 19/19985, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE,

Drucksache 19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen - Nachfrage zur

Antwort der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797, 16 June 2020

152,095

115,144

105,597

57,028

38,632

37,759

30,757

29,781

29,545

28,969

24,218

23,939

21,521

19,902

18,236

18,119

17,551

17,416

15,039

14,939

12,829

12,514

11,778

11,361

10,020

9,045

Luxembourg

Cyprus

Malta

Ireland

Slovakia

Greece

Netherlands

Spain

Austria

Lithuania

Belgium

Portugal

Denmark

Bulgaria

Sweden

Hungary

Finland

Czechia

Estonia

Italy

Croatia

Poland

Latvia

France

Romania

Slovenia

EPRS | European Parliamentary Research Service

102

DAC3

The number of APR and ATR rulings sent by Germany to the central platform run by EC DG TAXUD

stood at 119 in 2017, 20 in 2018 and 23 in 2019.

126

DAC4

The number of CbCRs prepared by German MNE grew from 359 in 2016 to 454 in 2017 and stood at

420 in 2018.

127

Luxembourg

Partial data on the evolution in the number of AEOIs is available for Luxembourg.

128

Regarding DAC1, the number of reports received by Luxembourg rose from 75 946 in tax year 2014

to 90 475 in tax year 2016 (the data for 2017 are incomplete). Immovable property accounted for

the bulk of these reports (62 % in 2015 and 2016, 64 % in 2014) and pensions were a distant second,

accounting for about 30 % of the reports received.

Table 6: Number of DAC1 reports received by Luxembourg under AEOI by tax year

Employment

income

Directors'

fees

Life-

insurance

products

Pensions

Immovable

property

Total

2014

5 399

291

370

20 992

48 894

75 946

2015

5 604

304

417

25 053

51 009

82 387

2016

6 190

342

1 317

26 496

56 130

90 475

2017

4 433

266

875

11 485

34 972

52 031

Notes: Data for tax year 2017 are incomplete

Source : Réponse de Monsieur le Ministre des Finances Pierre Gramegna à la question parlementaire n. 163 du

8 janvier 2019 de l’honorable Député Laurent Mosar concernant l’échange automatique d’informations en

matières fiscales

The total number of records received and sent by Luxembourg under DAC, FATCA and the CRS is

available for the years 2014 to 2017 on a country-by-country basis. It should be noted that the

information combines calendar and tax years. The number of records sent by Luxembourg to other

EU Member States has grown markedly following the coming into force of the DAC2 AEOI, rising

from 0.35 million in 2014 to 3.35 million in 2017. Luxembourg sends out many more records than it

126

See Deutscher Bundestag, Drucksache 19/13797, Antwort der Bundesregierung auf die Kleine Anfrage der

Abgeordneten Fabio De Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE –

Drucksache 19/12953 –Andauernde Umsetzung des Informationsaustauschs in Steuersachen, 8 October 2019.

127

See Deutscher Bundestag, Drucksache 19/20953, Schriftliche Fragen mit den in der Woche vom 6. Juli 2020

eingegangenen Antworten der Bundesregierung, 6 July 2020.

128

See Réponse de Monsieur le Ministre des Finances Pierre Gramegna à la question parlementaire n. 163 du 8 janvier 2019

de l’honorable Député Laurent Mosar concernant l’échange automatique d’informations en matières fiscales and

Réponse de M. Bob Kieffer à la question parlementaire n.1076 du 19 août 2019 de Monsieur le Député Sven Clement

concernant l'échange automatique d'informations.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

103

receives – by 2017, the number of records received stood at 0.51 million or 15 % of the number of

records sent.

Figure 5: Luxembourg - number of DAC messages sent to other EU-27 Member States and

received from other EU-27 Member States

Note: the yearly data combine calendar year and tax year data

Source: Réponse de M. Bob Kieffer à la question parlementaire n.1076 du 19 août 2019 de Monsieur le Député

Sven Clement concernant l'échange automatique d'informations

The EU-27 accounted for more than 90 % of the records sent and three quarter of records received.

The US was also a relatively important source of records received.

Figure 5: Luxembourg - records sent to EU-27 Member States and the US, and received from

EU Member States and the US as a percentage of total number of records sent and received

Note: the yearly data combine calendar year and tax year data

Source: Réponse de M. Bob Kieffer à la question parlementaire n.1076 du 19 août 2019 de Monsieur le Député

Sven Clement concernant l'échange automatique d'informations

In 2017, 45 % of all records sent by Luxembourg to EU-27 Member States went to Germany. The

other three major EU-27 countries to which Luxembourg sent DAC2 records were France, Belgium

and Italy which respectively accounted for 17 %, 14 % and 7 % of all records sent by Luxembourg in

2017.

In contrast, France was the major source of DAC2 records received by Luxembourg from EU EU-27

Member States, accounting for 28 % of all inbound DAC2 records received from EU-27 Member

States in 2017. The other important source EU-27 Member States were Belgium, Germany, Portugal

353,477

407,157

2,003,290

3,245,418

78,228

80,851

268,954

473,361

2014 2015 2016 2017

Records sent Records received

97%

83%

15%

95%

83%

12%

92%

68%

24%

89%

81%

to EU-2 7 MS

to USA from EU-27 MS from USA

Records sent Records received

2014 2015 2016 2017

EPRS | European Parliamentary Research Service

104

and Italy which respectively accounted for 19 %, 17 %, 15 %, and 7 % of all records received by

Luxembourg in 2017 from EU Member States.

Portugal

In 2019, the Portuguese TAs received 1 137 889 records and sent 2 698 799 records under the DAC,

CRS and FATCA AEOI with other TAs.

129

TAs from the EU accounted for 87 % of the records received

and 70 % of the records sent. The main Member States from which records were received under the

DAC AEOI include France (45 % of all records received), United Kingdom (17 %), Germany (9 %) and

Spain (8 %). France, Germany and the United Kingdom were also the Member States to which most

of the records were sent by the Portuguese TAs under the DAC AEOI. These three Member States

accounted for respectively 59.0 %, 13.6 % and 10.5 % of all records sent to TAs of EU Member States.

3.2. Impact of exchange of tax related information

The limited literature on the impact on cross-border financial asset holdings of the various measures

adopted by countries to reduce and tax evasion and fraud finds that following the implementation

of FATCA and the CRS off-shore holdings of financial assets by taxpayers has decreased. For

example:

Ahrens and Botner

130

conclude that ‘household assets in tax havens that are not hidden behind

corporate identities are estimated to be 67 per cent lower than they would have been without

automatic exchange of information. Furthermore, this reduction is not offset by an increase in

treaty circumvention using identity concealment or asset shifting to non-compliant jurisdictions.

FATCA and CRS thus implement the first effective international cooperation against tax evasion.’

Beer, Coelho and Leduc

131

find that ‘recent automatic exchange of information frameworks

reduced foreign-owned deposits in offshore jurisdictions by an average of 25 percent. This effect

is statistically significant and, as expected, much larger than the effect of information exchange

upon request, which is not significant’.

Casia, Spengel and Stage

132

note that the ‘the CRS induced a reduction of 11.5 % in cross-border

deposits parked in tax havens’.

De Simone, Lester and Markle

133

find ‘an average $7.8 billion to $15.3 billion decrease in e quity

foreign portfolio investment to the United States from tax‐haven countries after FATCA

implementation, consistent with a decrease in round‐tripping investments attributable to U.S.

investors’ offshore tax evasion’. Moreover, ‘when testing total worldwide investment out of

financial accounts in tax havens post‐FATCA, we find an average decline of $56.6 billion to

$78.0 billion’.

O’Reilly, Parra-Ramirez and Stemmer

134

find that, following the commencement of CRS and

FATCA AEOI, bank deposits in International Financial Centres fell by between 20 % and 25 %,

over and above the reduction associated with EOIR.

129

See República Portuguesa, Relatório sobre o Combate à Fraude a Evasão Fiscais e Aduaneiras 2019, Cabinete do

Secretário de Estado Adjunto e dos Assuntos Fiscais, July 2020.

130

See Leo Ahrens and Fabio Bothner, ‘The Big Bang: Tax Evasion After Automatic Exchange of Information Under FATCA

and CRS‘, New Political Economy, 25:6, 2020, pp. 849-864.

131

See Sebastian Beer, Maria Coelho, and Sebastien Leduc, Hidden Treasures: The Impact of Automatic Exchange of

Information on Cross-Border Tax Evasion, IMF Working Paper Fiscal Affairs Department, WP/19/286, 2019.

132

See Elisa Casi, Christoph Spengel, and Barbara M.B. Stage, ‘Cross-border tax evasion after the common reporting

standard: Game over?‘, Journal of Public Economics, Volume 190, October 2020.

133

See Lisa De Simone, Rebecca Lester, and Kevin Markle, ‘Transparency and Tax Evasion: Evidence from the Foreign

Account Tax Compliance Act (FATCA)‘, Journal of Accounting Research, Volume 58, Issue1, March 2020, pp. 105-153.

134

See Pierce O’Reilly, Kevin Parra Ramirez and Michael A. Stemmer, ‘Exchange of Information and Bank Deposits in

International Financial Centres’, OECD Taxation Working Papers, No. 46, OECD Publishing, 2019.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

105

The next section examines whether the adoption of DAC2 has resulted in a reduction of cross-border

holdings of financial assets in the EU.

3.2.1. The IMF Coordinated Portfolio Investment Survey (CPIS) and DAC2

The IMF Coordinated Portfolio Investment Survey (CPIS)

135

provides data on the reported portfolio

investment assets of participating economies in total and by sector of the holder and economy of

the non-resident issuer. Economies participating in the CPIS provide data on their holdings of

portfolio investment securities (separately for equity and investment fund shares; long-term debt

instruments; and short-term debt instruments). It should be noted that data is not available for every

country and every sector at the sectoral level.

136

Three sectors are considered in the analysis:

Households

Non-financial corporations

FIs other than the Central Bank. The foreign portfolio holding of this sector is the sum of holdings

of Depository-taking Corporations except the Central Bank and Other Financial Corporations

An analysis focusing on the evolution of foreign portfolio holdings of all EU-27 Member States in i)

the other EU-27 Member States (in aggregate); ii) in the rest of the world and iii) the world in total

reveals no notable change or significant structural break in portfolio holdings at the time of the

adoption of DAC2 in December 2014, or at its effectiveness date in January 2016. A review of foreign

portfolio holdings by Member States yields a similar conclusion.

Figure 6: Foreign portfolio holdings of EU-27 FIs (other than central banks), non-financial

corporations and households (as % of EU-27 GDP)

135

The CPIS data are available at https://data.imf.org/?sk=B981B4E3-4E58-467E-9B90-9DE0C3367363

136

In particular, no data was available at the sectoral level for the foreign portfolio holdings of Croatia, Luxembourg and

Malta. These three Member States therefore could not be included in the analysis. However, it is important to note

that data was available on the holdings of other EU Member States in these three Member States.

20%

40%

60%

80%

100%

120%

31/12/2001

31/12/2002

31/12/2003

31/12/2004

31/12/2005

31/12/2006

31/12/2007

31/12/2008

31/12/2009

31/12/2010

31/12/2011

31/12/2012

30/06/2013

31/12/2013

30/06/2014

31/12/2014

30/06/2015

31/12/2015

30/06/2016

31/12/2016

30/06/2017

31/12/2017

30/06/2018

31/12/2018

30/06/2019

31/12/2019

Portfolio holdings of Financial Institutions other

than the Central Bank (% of EU-27 GDP)

EU-27 investment in EU-27 (Financial Institutions other than the Central Bank)

EU-27 investment in non-EU (Financial Institutions other than the Central Bank)

EU-27 investment in World (Financial Institutions other than the Central Bank)

EPRS | European Parliamentary Research Service

106

Note: DAC2 was adopted in December 2014 and became effective in January 2016

Source: IMF Coordinated Portfolio Survey

3.2.2. The deposits of monetary financial institutions in the euro-zone and

DAC2

A detailed review by geographical origin of the evolution of the deposits (other than those by

monetary FIs (MFIs)) which are on the books of the monetary FIs in the euro-zone also shows no

impact of DAC2.

Within the euro-zone, at 20.4 %, Luxembourg has in 2019 the highest share of non-MFI deposits

which originate from other EU Members. If DAC2 had had any impact within the EU on cross-border

deposits by non-MFIs, one would expect such an impact to be reflected in particular in the volume

of foreign non-MFI deposits on the books of Luxembourg MFIs. However, neither the volume of non-

31/12/2001

31/12/2002

31/12/2003

31/12/2004

31/12/2005

31/12/2006

31/12/2007

31/12/2008

31/12/2009

31/12/2010

31/12/2011

31/12/2012

30/06/2013

31/12/2013

30/06/2014

31/12/2014

30/06/2015

31/12/2015

30/06/2016

31/12/2016

30/06/2017

31/12/2017

30/06/2018

31/12/2018

30/06/2019

31/12/2019

Portfolio holdings of Non-financial Corporations

(% of EU-27 GDP)

EU-27 investment in EU-27 (Non-financial Corporations)

EU-27 investment in non-EU (Non-financial Corporations)

EU-27 investment in World (Non-financial Corporations)

10%

12%

31/12/2001

31/12/2002

31/12/2003

31/12/2004

31/12/2005

31/12/2006

31/12/2007

31/12/2008

31/12/2009

31/12/2010

31/12/2011

31/12/2012

30/06/2013

31/12/2013

30/06/2014

31/12/2014

30/06/2015

31/12/2015

30/06/2016

31/12/2016

30/06/2017

31/12/2017

30/06/2018

31/12/2018

30/06/2019

31/12/2019

Portfolio holdings of Households

(% of EU-27 GDP)

EU-27 investment in EU-27 (Households)

EU-27 investment in non-EU (Households)

EU-27 investment in World (Households)

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

107

domestic deposits by non-MFIs originating from the eurozone nor the volume of such deposits

originating from non-eurozone EU Member States shows any marked impact of DAC2 (see Figure

7).

Figure 7: Deposits on the books of MFIs in Luxembourg (million EUR)

Source: ECB balance sheets of MFIs

3.2.3. The BIS locational banking statistics and DAC2

Data from the BIS locational banking statistics are used next to delve in greater detail into the

evolution of the Luxembourg banking system as a destination for cross-border financial asset

investment from other EU Member States. These BIS statistics provide, among other, information on

the cross-border bank loan and deposit liabilities to non-bank counterparties from other EU Member

States. As in the case of the ECB data, the BIS data do not show any impact of DAC2 on the exposure

of Luxembourg banks to non-bank counterparties in the EU (Figure 8).

Figure 8: Loan and deposit liabilities of Luxembourg banks to non-bank EU counterparties

(million EUR)

Source: BIS Locational Banking Statistics

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

100,000

Monthly, Other Euro area member states (all countries except the reference area), Non-MFIs Quarterly, EU member states not belonging to the Euro area, Non-MFIs

Quarterly, All areas other than EU and ref./home area, Non-resident non-banks

20.000

40.000

60.000

80.000

100.000

120.000

y1999-Q1

y1999-Q4

y2000-Q3

y2001-Q2

y2002-Q1

y2002-Q4

y2003-Q3

y2004-Q2

y2005-Q1

y2005-Q4

y2006-Q3

y2007-Q2

y2008-Q1

y2008-Q4

y2009-Q3

y2010-Q2

y2011-Q1

y2011-Q4

y2012-Q3

y2013-Q2

y2014-Q1

y2014-Q4

y2015-Q3

y2016-Q2

y2017-Q1

y2017-Q4

y2018-Q3

y2019-Q2

EPRS | European Parliamentary Research Service

108

4. Coherence with other provisions

The 2019 evaluation of DAC by the EC looked at the internal and external coherence of the Directive.

From an international perspective, it focused on the coherence with OECD provisions, with the final

assessment suggesting that although improvements are possible, overall, the coherence is high

between the provisions. However, the evaluation did not consider coherence in a wider context.

DAC is an EU specific mechanism for exchange of information between Member States but sits

within broader efforts globally to improve transparency and cooperation between jurisdictions to

tackle tax avoidance and evasion.

This section focuses on two of these efforts, namely FACTA and the OECD’s CRS. It also provides an

outline of the reporting obligations of jurisdictions previously covered by the European Savings Tax

Directive (repealed with the introduction of DAC2). DAC sits within the broader EU legislative

framework, other parts of which are also relevant and interact with DAC and the fight against tax

evasion and avoidance. Therefore, this Section also outlines these provisions in relation to key areas

in particular: anti-money laundering (AML) and judicial assistance, including recovery of assets.

4.1. FATCA and CRS

The present section provides first an overview of FATCA which was passed as part of the HIRE Act

137

and the CRS developed by the OECD and the Global Forum on Transparency and Exchange of

Information for Tax Purposes in cooperation with a large number of countries.

While both FATCA and CRS focus on AEOI on financial assets held by taxpayers outside their home

country and foresee the provision of such information to the TAs of the home country or country of

residence of a taxpayer, they differ in a number of aspects.

4.1.1. FATCA

FATCA was introduced in 2010 and requires that foreign (non-US) FIs and other certain foreign non-

FIs report on the foreign assets held by US account holders.

138

US FIs operating abroad with US

account holders are outside the scope of FATCA. Under FATCA, reporting FIs which do not comply

with the requirements are subject to 30 % withholding tax on income sourced from the US.

139

The

introduction of FATCA made several amendments to the Internal Revenue Code (IRC), allowing the

Internal Revenue Service (IRS) to receive reports from foreign FIs on US holder accounts.

140

B o th FIs

and TAs can exchange FATCA data with the US.

141

The US is one of the few jurisdictions in the world that taxes all income earned worldwide by US

citizens, rather than taxing income where it is earned, irrespective of whether the US citizens are

resident or non-resident in the US. Therefore, the challenge of gathering information relevant for

tax purposes on US citizens by the US could be considered to be a far greater one than for those

jurisdictions looking to tax income earned either within the jurisdiction or in relation to a relatively

137

The Hiring Incentives to Restore Employment Act.

138

See the Foreign Account Tax Compliance Act (FATCA), 2010.

139

See James F. Kelly, ‘International Tax Regulation by United States Fiat: How FATCA represents unsound international

tax policy’, Wisconsin International Law Journal, Vol. 34 Issue 4, May 2017.

140

ibid.

141

See the Foreign Account Tax Compliance Act (FATCA), 2010.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

109

small number of citizens resident abroad. This is one of the reasons why the US has looked to

implement a mechanism which enhances global exchange of information in relation to US citizens.

FATCA does not impose a legally enforceable requirement on the foreign FIs, as they are not subject

to the US legal system. Therefore, intergovernmental agreements (IGAs) are required between the

US and other countries in order for the US to be able to enforce FATCA requirements.

142

There are

two model types of IGA: under Model 1 the other jurisdiction collects the information from the

reporting foreign FIs in their jurisdiction and transfers all the information to the US. Under Model 2,

this information is not collected by the jurisdiction, but instead is directly transferred from the

foreign FI to the US.

143

Moreover, under Model 2, the US does not automatically send information

on financial accounts held in the US by taxpayers of the other jurisdiction party to the Model 2 IGA

to that TA of that jurisdiction. All EU-27 Member States but Austria have signed a Model 1 IGA with

the US. Austria has signed a Model 2 IGA

144

. Therefore, among EU-27 Member States, only Austria

does not receive information automatically from the US on its taxpayers with accounts in the US.

4.1.2. CRS

Following the introduction of FATCA and the global financial crisis, significant efforts continued to

be made globally to progress the international exchange of information to improve transparency.

In 2013 a formal request was made by the G20 to the OECD to develop a common reporting (CRS)

standard. The OECD’s CRS was approved by the OECD Council on 15 July 2014. It sets out a standard

for AEOI and requires jurisdictions to obtain and automatically exchange information from their FIs

on an annual basis. Its introduction did not replace or amend other exchange of information

agreements, and instead serves to act as a minimum required standard for the exchange of

information. Currently, over 100 jurisdictions have signed up to CRS.

145

CRS, in contrast to FATCA, is a multilateral instrument, although it draws heavily on the

intergovernmental dynamics used in the implementation of FATCA. The CRS uses the Multilateral

Convention on Mutual Administrative Assistance in Tax Matters (amended in 2011) as its legal basis

and the CRS Multilateral Competent Authority Agreement (CRS MCAA) as the agreement on the

exchanges, specifying the details of what information is exchanged and when. However,

jurisdictions may also exchange through bilateral agreements, such as double tax treaties.

146

4.1.3. Comparison of the provisions of FATCA and the CRS

Although FATCA and the CRS are similar in many aspects they are not identical. Appendix 2 provides

an extensive comparison of FATCA and the CRS, and a summary overview of the key differences is

provided below.

Persons covered by FATCA and the CRS

FATCA focuses on foreign asset holdings by US persons (natural or legal) and US residents holding

financial assets outside the US. In contrast, the focus of the CRS is all residents of a jurisdiction (from

a tax perspective) holding financial assets in other jurisdictions which have signed a CRS agreement

with the home jurisdiction of the residents.

142

See Kelly, 2017.

143

See Kelly, 2017.

144

The IRS website lists all the countries having signed a Model 1 or Model 2 agreement with the US.

145

See OECD, International Framework for the CRS, OECD website.

146

ibid.

EPRS | European Parliamentary Research Service

110

Another key difference is that in the case of the CRS, controlling persons of a NFE are reportable

irrespective of whether they are resident in the same jurisdiction as the NFE while, in the case of

FATCA, only US controlling persons a passive NFE are reportable.

Moreover, under FATCA, an entity is a related entity if direct or indirect ownership is 50 % of either

the vote or the value of the entity. In contrast, under the CRS, the ownership is more demanding

requiring that both tests of 50 % of the vote and 50 % of the value are met.

Financial assets to be reported

One major difference is that non-debt direct investments in real property are formally excluded from

the CRS. Such investments, however, are covered by DAC1.

There are also significant differences in reporting thresholds. Under the FATCA any depository

accounts with a balance of US$ 50 000 or less are not reportable and a Pre-existing Entity account

with a balance or value that does not exceed US$ 250,000 does not have to be reported until it

exceeds US$ 1 million. The CRS does not have such thresholds. Moreover, FATCA also excludes Cash

Value Insurance Contracts with a value of US$ 50 000 or less. The CRS has no reporting threshold for

such contracts.

4.1.4. DAC2 and the CRS

Within the EU, CRS was implemented through DAC2, although some of the structure of the

mechanism already formed part of DAC1

147

, and CRS information is exchanged on the basis of this

Directive rather than the CRS MCAA. Whereas CRS provides for high level generic provisions, DAC2

concretely defines these for EU Member States and covers a slightly broader range of obligations.

As noted in the recital of the DAC2, the objective of DAC2 is that ‘Member States should require their

Financial Institutions to implement reporting and due diligence rules which are fully consistent with

those set out in the Common Reporting Standard developed by the OECD. Moreover, the scope of

Article 8 of Directive 2011/16/EU should be extended to include the same information covered by

the OECD Model Competent Authority Agreement and Common Reporting Standard. It is expected

that each Member State would have only one single list of domestically-defined Non-Reporting

Financial Institutions and Excluded Accounts that it would use both when implementing this

Directive and for the application of other agreements implementing the global standard.’

148

Overall, DAC2 incorporates the ‘full extent of the CRS and some important elements of the

Commentaries’.

149

The CA of each Member State was required to start exchanging automatically

with the CA of any other Member State information regarding taxable periods as from

1 January 2017 concerning residents in that other Member State, on all categories of income and

capital listed in paragraph 1.

150

The key features of DAC2 and the CRS are provided at Appendix 3.

147

See European Commission, Working Document of the Expert Group on Automatic Exchange of Financial Account

Information, October 2014.

148

See Recital 9 of DAC2.

149

See Working Document on the Implementation of Directive 2014/107/EU, WPIV Meeting on 19 June 2015, European

Commission (DG TAXUD.)

150

See Article 1.c of DAC2.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

111

4.2. EU Savings Tax Directive

Council Directive 2003/48/EC

151

commonly referred to as the STD was a form of AEOI introduced in

2003 and repealed in 2015.

152

Under the STD, the TAs of a Member State were required to provide

other Member States’ TAs with details relating to payments of interest or similar income which was

paid by an entity or person within its jurisdiction to an individual resident in another Member

State.

153

The Directive was adopted to encourage proper operation of the internal market and to

address tax evasion issues.

154

DAC2 differs in many key aspects from the STD:

1 Under DAC2, both income on financial assets and the value of financial assets are reportable

and subject to the automatic information exchange between TAs while under the STD only

income (paid or accrued) was.

155

2 Under DAC2, all types of income (interest, dividends, etc.) accruing to financial assets are

reportable while under the STD only interest income was.

3 Under DAC2, income and asset value of accounts held by natural and legal persons are

reportable whereas under the STD only interest paid to individuals (i.e. natural persons) was

subject to the automatic exchange of information.

156

151

See Council Directive 2003/48/EC of 3 June 2003 on the taxation of savings income in the form of interest payments.

152

See Repeal of the Savings Directive in line with international and EU developments, European Commission website.

153

See ICMA, ‘Savings Tax Directive’, ICMA website.

154

See Tax Justice Network, ‘EU Savings Tax Directive’, Tax Justice Briefing, 2008.

155

According to Article 6 of the STD, interest payments that were reportable under the STD include:

‘(a) interest paid or credited to an account, relating to debt claims of every kind, whether or not secured by mortgage and

whether or not carrying a right to participate in the debtor's profits, and, in particular, income from government

securities and income from bonds or debentures, including premiums and prizes attaching to such securities, bonds

or debentures;

(however), penalty charges for late payments shall not be regarded as interest payments;

(b) interest accrued or capitalised at the sale, refund or redemption of the debt claims referred to in (a);

(c) income deriving from interest payments either directly or through an entity referred to in Article 4(2), distributed by: (i)

an UCITS authorised in accordance with Directive 85/611/EEC, (ii) entities which qualify for the option under

Article 4(3), (iii) undertakings for collective investment established outside the territory referred to in Article 7;

(d) income realised upon the sale, refund or redemption of shares or units in the following undertakings and entities, if

they invest directly or indirectly, via other undertakings for collective investment or entities referred to below, more

than 40 % of their assets in debt claims as referred to in (a): (i) an UCITS authorised in accordance with Directive 85/

611/EEC, (ii) entities which qualify for the option under Article 4(3), (iii) undertakings for collective investment

established outside the territory referred to in Article 7.

However, Member States shall have the option of including income mentioned under (d) in the definition of interest only

to the extent that such income corresponds to gains directly or indirectly deriving from interest payments within the

meaning of (a) and (b)’.

156

Article 2 of the STD specifies that: ’For the purposes of this Directive, ‘beneficial owner’ means any individual who

receives an interest payment or any individual for whom an interest payment is secured, unless he provides evidence

that it was not received or secured for his own benefit, that is to say that: (a) he acts as a paying agent within the

meaning of Article 4(1); or (b) he acts on behalf of a legal person, an entity which is taxed on its profits under the

general arrangements for business taxation, an UCITS authorised in accordance with Directive 85/611/EEC or an entity

referred to in Article 4(2) of this Directive and, in the last mentioned case, discloses the name and address of that

entity to the economic operator making the interest payment and the latter communicates such information to the

competent authority of its Member State of establishment, or (c) he acts on behalf of another individual who is the

beneficial owner and discloses to the paying agent the identity of that beneficial owner in accordance with

Article 3(2).

EPRS | European Parliamentary Research Service

112

4 Under DAC2, reporting FIs have to provide information on any accounts held by a Passive

Non-Financial Entity (NFE) with one or more Controlling Persons which are Reportable

Persons. In such cases, information on the Controlling Person(s) must also be reported. The

STD did not include such an obligation.

5 Under DAC2, the reporting obligation is narrower than under the STD. In DAC2, the

obligation falls on the Reporting Financial Institution while under the STD the obligation fell

on the paying agent.

157

The concept of paying agent is broader than that of Reporting

Financial Institution as the former is ‘any economic operator who pays interest to or secures

the payment of interest for the immediate benefit of the beneficial owner’. Thus, entities

other than Reporting FIs having sold debt securities directly to retail investors have no

reporting obligations under DAC2 but had under the STD.

The STD was applicable to all EU Member States, as well as any dependent or associated territories

of EU Member States, being Anguilla, Aruba, British Virgin Islands, Cayman Islands, Guernsey, Jersey,

Isle of Man, Montserrat, Netherlands Antilles, and the Turks and Caicos Islands.

158

Bilateral

agreements were signed with each of these territories by the EU Member States for the STD to be

implemented. Equivalent measures were also applied in the following participating ‘third countries’:

Andorra, Liechtenstein, Monaco, San Marino and Switzerland.

159

However, in the case of these

countries, a taxpayer could choose not to disclose interest payments to the EU Member State where

they were resident, and instead a withholding tax would be levied.

DAC2, implementing the CRS on automatic exchange of financial account information across the

EU, resulted in the repeal of STD in order to avoid dual reporting obligations and to save costs both

for TAs and economic operators. DAC2 has a broader scope which covers interest income, other

types of capital income and annual balances of accounts producing these incomes. Therefore, the

STD was repealed in order to prevent cases of unnecessary overlap.

160

The OECD’s CRS has been signed up to by over 100 jurisdictions, including those no longer covered

by the STD and which are not EU Member States.

161

These jurisdictions, namely the dependent

territories and third countries listed above were already committed to applying CRS reporting at the

time the STD was repealed.

The European Council adopted Amending Protocols with the 5 ‘third countries’ in order to align

their reporting requirements with the obligations on Member States under DAC2. Table 7 below

sets out the Amending Protocols.

2. Where a paying agent has information suggesting that the individual who receives an interest payment or for whom an

interest payment is secured may not be the beneficial owner, and where neither paragraph 1(a) nor 1(b) applies to

that individual, it shall take reasonable steps to establish the identity of the beneficial owner in accordance with

Article 3(2). If the paying agent is unable to identify the beneficial owner, it shall treat the individual in question as the

beneficial owner.’

157

Article 4 of the STD specifies that a paying agent means ‘any economic operator who pays interest to or secures the

payment of interest for the immediate benefit of the beneficial owner, whether the operator is the debtor of the debt

claim which produces the interest or the operator charged by the debtor or the beneficial owner with paying interest

or securing the payment of interest’.

158

See Tax Justice Network, 2008.

159

See European Commission, EU savings taxation rules and savings agreements with third countries: frequently asked

questions, Memo

, 2014.

160

See European Commission, Repeal of the Savings Directive in line with international and EU developments, European

Commission website.

161

See OECD, International Framework for the CRS, OECD CRS website.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

113

Table 7: Amending Protocols for third countries

Jurisdiction

Employment

income

Directors' fees

Principality of Andorra

12/02/2016

Amending Protocol attached to Council Decision (EU) 2016/1751

(OJ L268/2016) on conclusion.

Principality of Liechtenstein

28/10/2015

Amending Protocol attached to Council Decision (EU) 2015/2453

(OJ L339/2015) on conclusion.

Monaco

12/07/2016

Amending Protocol attached to Council Decision (EU) 2016/1392

on signature and provisional application (OJ L225/2016).

Republic of San Marino

08/12/2015

Council Decision 2016/1830 on conclusion adopted on

11/10/2016 (OJ L280/2016).

Swiss Confederation

27/05/2015

Amending Protocol attached to Council Decision (EU) 2015/2469

on signature and provisional application (OJ L346/2015).

Bilateral agreements with dependent territories were replaced by CRS based (multilateral or bilateral

agreements) and exchange of information is now provided for by the OECD mechanism.

4.3. Other EU provisions

DAC works alongside AML and judicial assistance provisions in the fight against tax evasion and

avoidance. While sharing a common objective, these different provisions have different overarching

priorities, creating a risk of lack of coherence between them. The issue of coherence between these

areas of legislation and DAC is explored in further detail in the sub-sections below.

4.3.1. AML provisions

AML provisions in the EU have recently been updated to reflect the increasing importance of

combating money laundering activities. AML and exchange of information provisions have a

substantial number of interactions and dependencies, due to the type of information required to

both tackle money laundering and tax avoidance and/or evasion. BO in particular is a key aspect for

both and is discussed in more detail below.

In 2015, Directive (EU) 2015/849, commonly referred to as the AMLD4

162

was adopted. This Directive

required legal entities to hold complete, accurate and up-to-date information on their beneficial

ownership, and for that information to be made easily accessible to competent authorities. In

addition, obliged entities are required to evidence that they have taken the necessary and

appropriate steps to identify BO, which depend on the level of risk associated with the customer. In

relation to the DAC provisions, this is particularly important in relation to DAC2. Under DAC2, where

the account holder is a passive NFE (an intermediary structure), reporting FIs must, under the due

diligence requirements, look through to assess and identify who the ‘controlling persons’ are. This

aspect of the Directive, as noted in Section 4.3.1, relies on AML information obtained by the FI under

AMLD4. TAs must ensure that FIs are complying with the due diligence procedures in order for DAC

to be effective, and therefore access to AML information is an essential part of this. AMLD4 therefore

162

See Directive (EU) 2015/849 of the European Parliament and of the Council of 20 May 2015 on the prevention of the use

of the financial system for the purposes of money laundering or terrorist financing, amending Regulation (EU) No

648/2012 of the European Parliament and of the Council, and repealing Directive 2005/60/EC of the European

Parliament and of the Council and Commission Directive 2006/70/EC.

EPRS | European Parliamentary Research Service

114

allows Member States to have access to the AML information and check the correctness of that

reported under DAC2.

Directive (EU) 2018/843, the AMLD5

163

, amending AMLD4, was published in the Official Journal of

the European Union in 2018. Amongst other amendments and additions, it:

sets up publicly available registers for companies, trusts and legal arrangements;

provides access to data on the beneficial ownership of trusts without restrictions to competent

authorities (including tax administrations) and other relevant parties;

interconnects beneficial ownership registers at the EU level, facilitating cooperation and

exchange of beneficial ownership information across Member States;

requires Member States to have mechanisms in place to verify the information collected by the

registers; and

enhances the powers of Financial Intelligence Units to cooperate more easily between

themselves and other competent authorities (including tax administrations).

Previous to the introduction of Directive (EU) 2016/2258 (DAC5) (out of scope for the purposes of

this Research Paper), there was limited interaction between DAC provisions and AML provisions, as

highlighted above. Although AMLD4 required information to be made accessible to competent

authorities, this was only required at a national level, limiting the opportunity for cooperation

between Member States.

DAC5 enhanced the interaction between the provisions by allowing for access by TAs to BO

information collected under the AML provisions, through non-AEOI provisions. It also includes

dynamic references to the AML provisions, ensuring that any changes (such as the AMLD5

amendments) automatically feed into DAC5

164

. This means that under the provisions of DAC5 and

their interaction with AMLD5, Member States will be able to access BO information across Member

States, though the interconnected BO registers. Nevertheless, given that exchanges under DAC5

have only taken place recently

165

, and that AMLD5 was required to be transposed into national law

by 10 January 2020, there is limited evidence as to effectiveness of these provisions.

Challenges surround the interaction between DAC and AML provisions

Although significant steps have been made to enhance cooperation and integration between both

types of provisions, some challenges still remain. In particular, challenges surrounding the definition

of beneficial ownership have been raised and both national and international fora. Currently, the

definition of beneficial ownership is within the discretion of Member States, although there is a

significant alignment with the recommendations set out by the Financial Action Task Force

166

, in

particular within the AML provisions. In addition, there is also a challenge surrounding the

supervisory element of the Directive, which ensures that the information held is accurate and

complete so that any exchange of information is relevant, useful, and adds value to the Member

State receiving the information, in addition to other challenges. These challenges are outlined in

more detail below.

Definition of tax crime

163

See Directive (EU) 2018/843 of the European Parliament and of the Council of 30 May 2018 amending Directive (EU)

2015/849 on the prevention of the use of the financial system for the purposes of money laundering or terrorist

financing, and amending Directives 2009/138/EC and 2013/36/EU.

164

See Council Directive (EU) 2016/2258 of 6 December 2016 amending Directive 2011/16/EU as regards access to anti-

money-laundering information by tax authorities.

165

Source: interview with DG TAXUD.

166

See FATF Guidance: Transparency and Beneficial Ownership, Financial Action Task Force, 2014.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

115

The purpose of the AML provisions is to ensure that money laundering is prohibited to the greatest

extent, and that any proceeds derived from criminal activity are identified for this purpose. In

relation to tax, when defining ‘criminal activity’, paragraph 4 of Article 3 of AMLD4 includes tax

crimes. The definition of tax crime is not detailed in the Directive, given that this is within Member

States’ competence. There is no unified definition of tax crime across the EU, and indeed, it has

proven difficult to achieve consensus around this.

167

However, although the preamble to AMLD4

does not call for harmonisation of the term, it does call for Member States to allow EOI in this area

to the greatest extent possible under national law as to enhance the effectiveness of the provisions.

Nevertheless, obliged entities may not be required to keep BO ownership if transactions do not fall

within the national definition of tax crime where they are subject to the AML provisions, meaning

that the relevant TA will miss out on this information.

Beneficial ownership

There is no common or consistent definition of BO across Member States, or between the provisions

of DAC (DAC2 in particular) and the AML provisions. DAC2 specifies that the definition of beneficial

ownership should be in line with FATF recommendations regarding beneficial ownership, as

discussed in Section 2.1.3. In addition, the AML provisions also take their steer from the FATF

guidance. This guidance focuses on the concept of control, and defines a beneficial owner as being

‘the natural person(s) who ultimately owns or controls a customer and/or the natural person on

whose behalf a transaction is being conducted. It also includes those persons who exercise ultimate

effective control over a legal person or arrangement.

168

A defining feature of BO is being able to exercise significant influence and, whilst to some extent

this features within the concept of control, it is not the main focus. Rather, control tends to be based

around rights and power, most commonly demonstrated through legal ownership. When looking

to define BO, this usually entails searching for the natural person holding BO. For corporate entities

or structures, most ownership is obtained through shareholding. Whilst this can indicate legal

ownership, this does not always equate to BO, in particular when looking at complex structures

where the main purpose is to hide the real ownership for tax fraud purposes. For example, bearer

shares are commonly used for these purposes, and shadow directors, whilst controlling a

corporate’s actions and having BO, will not usually be declared as a legal owner. A similar issue exists

with trusts, for example where discretionary trusts are settled by a ‘dummy settlor’. This example is

discussed further at section 2.1.3. This challenge of identifying real beneficial ownership is linked

also to that of ensuring that the information is accurate, discussed below.

In addition, there is also a challenge surrounding the slight variances in the definitions of BO

between DAC and AML provisions, and between Member States themselves. Indeed, one of the

main challenges discussed in section 2.1.3 is that which relates to the percentage of ownership

interest, which constitutes BO (usually 25 %). These inconsistencies can limit the effectiveness of the

interaction between the provisions and therefore the EOI for tax purposes.

Access to and successful exchange of information

In addition, whilst ensuring that the correct information is held regarding BO is crucial to the

effectiveness of the AML provisions, a key aspect for the effectiveness of DAC is the access and

167

Marcos Alvarez Suso, ‘The boundaries of abusive practices: the grey zone‘, ECJ – Recent Developments in Value Added

Tax: The Evolution of European VAT Jurisprudence and Its Role in the EU Common VAT System, Linde Verlag,

Schriftenreihe, 2014.

168

See FATF, FAFT Guidance, Transparency and Beneficial Ownership, 2014.

EPRS | European Parliamentary Research Service

116

successful exchange of this information. This relies on the effective implementation of the

provisions, which has been a challenge across Directives. In relation to AML provisions, at the time

of writing, transposition for AMLD4 stands at 89 %

169

across EU Member States, with the EC

launching infringement procedures against 8 Member States

170

. It should be noted that the

transposition deadline for these provisions was 16 June 2017. As for AMLD5, transposition stands at

70 %

171

, with a transposition deadline of 10 January 2020, and infringement procedures launched

against 16 Member States.

172

173

Furthermore, as discussed in Section 2.1.3 the FATF assessments indicate that Member States are

generally underperforming across the various indicators and recommendations, which not only

impacts the effectiveness of the mechanisms, but also the accuracy of the information to be

exchanged. The latter is discussed further below.

In addition, DAC5 relies on AML information being requested through EOIR, primarily to avoid

‘fishing’. However, to a large extent, this requires Member States to know what and who they need

information on in order to make the request. In addition, national resources must be exhausted, and

the standard of ‘foreseeable relevance’ must be met. This means that there is a purpose limitation

in that the information must be for tax purposes. Therefore, it may be possible that the effectiveness

of DAC may be limited to the extent that there needs to be prior suspicion in order to make the

request for information, and that this information may only be used for the stated purposes.

Accuracy of information

In order for the AML and DAC mechanisms to function effectively, the BO information must be

accurate. Indeed, under AML provisions, obliged entities must maintain accurate and up-to-date

information. However, the extent to which the accuracy of the information can be guaranteed or

any inaccuracies minimised can be debated. Neither AMLD4 or AMLD5 set minimum standards or a

common approach for the supervision or regulation of AML activities. This has led to some

divergence in Member States’ approach to supervision and has resulted in AML supervision

approaches not being always effective.

174

This not only impacts the effectiveness of the

identification of BO AML provisions and indirectly those of DAC, but also creates imbalances in the

169

As of 25 November 2020. See European Commission website Anti-money laundering Directive IV (AMLD IV) –

transposition status at https://ec.europa.eu/info/publications/anti-money-laundering-directive-4-transposit i on-

status_en.

170

Information as of 22 December 2020: Czechia, Denmark, Estonia, Ireland, Italy, Luxembourg, Romania and Slovakia (see

European Commission website:

https://ec.europa.eu/atwork/applying-eu-law/infringements-

proceedings/infringement_decisions/index.cfm?lang_code=EN&typeOfSearch=false&active_only=1&noncom=0&r

_dossier=&decision_date_from=&decision_date_to=&title=Directive+2015%2F849&submit=Search.

171

As of 25 November 2020. See European Commission website Anti-money laundering Directive V (AMLD V) –

transposition status at https://ec.europa.eu/info/publications/anti-money-laundering-directive-5-transposit i on-

status_en.

172

Information as of 22 December 2020: Austria, Belgium, Cyprus, Czechia, Estonia, Greece, Hungary, Ireland, Italy,

Luxembourg, Netherlands, Portugal, Romania, Slovakia, Slovenia and Spain (see European Commission website:

https://ec.europa.eu/atwork/applying-eu-law/infringements-

proceedings/infringement_decisions/index.cfm?lang_code=EN&typeOfSearch=false&active_only=1&noncom=0&r

_dossier=&decision_date_from=&decision_date_to=&title=Directive+2015%2F849&submit=Search.

173

Please note that there is discrepancy in the number of MS reported on the EC AML page as facing infringement

procedures compared to the infringement decisions database. The infringement figures shown above are those per

the database as these are the most up to date and accurate.

174

See European Banking Authority, Report on Competent Authorities’ Approaches to the Anti-Money Laundering and

Countering the Financing of Terrorism Supervision of Banks, June 2020.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

117

EOI system, whereby information from some Member States may be distinctly more accurate than

that from others where supervision of activities is limited. In recognition of the challenges posed by

differing AML supervision across the EU, the EC released an Action Plan in May 2020 which supports

the creation of an EU-level AML supervisor.

175

4.3.2. Judicial assistance provisions

This section reviews how the DAC provisions interact with the arrangements implemented in the EU

to facilitate mutual assistance in criminal matters and the recovery of financial assets in criminal

cases.

‘Contrary to fears that it might hinder mutual assistance in criminal matters there is evidence of

improvements in interaction between administrative cooperation and criminal matters. For

example, information gathered from exchanges may be used for other purposes such as in criminal

investigations of tax crimes.’

176

DAC and the Protocol established by the Council in accordance with Article 34 of

the Treaty on European Union to the Convention on Mutual Assistance in Criminal

Matters between the Member States of the European Union

Under the Protocol, a Member State can send a request to another Member State to determine

whether a natural or legal person that is the subject of a criminal investigation holds or controls one

or more accounts, of whatever nature, in any bank located in the territory of the other Member State

and, if so, provide all the details of the identified accounts. This can also include accounts for which

the person that is the subject of the proceedings has powers of attorney, provided such information

can be gathered within a reasonable amount of time.

Under DAC2, information on holders of cross-border accounts is automatically exchanged between

TAs of Member States and, if this information could be used for purposes other than tax matters, it

would speed up considerably the process of finding out whether a person under criminal

investigation has bank accounts in one or several Member States. This would increase the efficiency

of the criminal investigation.

However, at the present time, the use for non-tax matters of the information having been

exchanged between TAs requires the authorisation of the Member State(s) having sent the

information. Such authorisation is not always granted, often because the national laws of the

sending Member State(s) do not permit the use of tax related information in non-tax matters.

DAC and the Regulation (EU) 2018/1805 of the European Parliament and of the

Council of 14 November 2018 on the mutual recognition of freezing orders and

confiscation orders

The Regulation aims to facilitate cross-border asset recovery and make the freezing and confiscation

of criminal assets across the EU quicker and simpler. The Regulation became applicable from 19

December 2020 and applies to all freezing and confiscation orders issued within the framework of

proceedings in criminal matters. The interaction between DAC and the Regulation is only indirect in

the sense that information exchanged under DAC could result, in some situations, in fraud and tax

175

See Communication from the Commission on an Action Plan for a comprehensive Union policy on preventing money

laundering and terrorist financing, C(2020) 2800 final

, May 2020.

176

See European Commission, Staff Working Document on the application of Council Directive (EU) no 2011/16/EU on

administrative cooperation in the field of direct taxation, SWD(2017) 462 final, December 2017.

EPRS | European Parliamentary Research Service

118

evasion cases being pursued, and freezing and confiscation orders of financial assets being issued

against financial assets held abroad which had been identified through the AEOI under DAC.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

119

5. Conclusions

This section outlines the conclusions stemming from the analysis and summarises the suggestions

presented in the literature and in interviews with Member States for improving the effectiveness

and implementation of DAC1 to 4.

5.1. General conclusions on the implementation of DAC

In order to assess the implementation of DAC1 to DAC4, this Research Paper focused on three key

areas of analysis: effectiveness of DAC, evolution of information exchanges, and coherence with

other EU and international provisions. The conclusions for each area are summarised below. Overall,

the implementation of DAC has been effective between DAC1-4, with each successive amendment

helping to further achieve its core objectives to increase transparency and enhance cooperation

between Member States in order to help tackle tax fraud and evasion across the EU.

Due to a lack of data it is not possible to estimate in quantitative terms what the overall impact of

the Directive has been. However, in the absence of this data Member State interviews did highlight

their satisfaction with the effectiveness of the Directive. In contrast, further analysis of the literature

in sections 2.1 and 2.2 has highlighted that the Directive is not perfect in its implementation, as there

are a number of circumventions and obstacles which have reduced the overall effectiveness of

DAC1-4.

With these considerations in mind, and taking into account the views of Member States,

stakeholders and findings from a review of relevant literature, section 5.2 sets out a number of

suggestions for potential improvements in the functioning of DAC, highlighting the areas that can

be more impactful in improving the effectiveness and coherence of the Directive.

5.1.1. Effectiveness of DAC

Section 2 outlined the conceptual framework for this evaluation of the effectiveness of the

provisions, highlighting three key assessment areas: comprehensiveness of scope, relevance or

appropriateness of the provisions, and timeliness.

Comprehensiveness of scope

Evidence from the literature and interviews with Member States both indicate that the coverage of

the Directive has become increasingly comprehensive, keeping up with the evolution of the

definition of income to a large extent as outlined in section 2.1.2., capturing income earned through

new and developed business models, as well as traditional forms of earning. However, some income

categories remain outside the scope of the Directive. The main areas of concern appear to be

covered in the proposals considered for DAC7, for the income generated by online platforms, and

DAC8, which covers crypto assets and e-money. These potential amendments are still a work-in-

progress and there remain some categories of income which will continue to fall out of the scope of

DAC, such as royalties, capital gains, interest of loans, and payments in kind.

In particular, a broader coverage to include, for example, capital gains may have been beneficial in

helping TAs identify cum-ex and cum-cum schemes sooner. In relation to DAC1 in particular, the

effectiveness of DAC is also limited to the extent that information on only three out of five income

categories must be exchanged automatically, and, even then, only if this is available. Furthermore,

as identified in section 2.1.3., a number of circumventions and loopholes have been identified in the

literature and consultation with Member States and other stakeholders, which have impacted the

effectiveness of the Directive, either as a result of the design of DAC and its amendments, or through

EPRS | European Parliamentary Research Service

120

the development of mechanisms in the market which have enabled a circumvention of

requirements under the provisions.

As a result, these issues can reduce the effectiveness of DAC by either preventing all or some

information from being exchanged by TAs, or reducing the quality of the information exchanged to

the extent it impacts the utility of the information for the purposes of assessing tax liabilities

correctly. Section 5.2. provides some indication as to how these issues may be overcome to improve

the effectiveness of the Directive.

Despite the challenges noted above, DAC’s scope has continued to evolve in a globalised context

to help TAs effectively tackle tax fraud and evasion. This is likely to continue, given ongoing

discussion on proposals for DAC7 and DAC8.

Relevance and appropriateness

The Research Paper has described a number of areas which, either as a result of the design of the

provisions or through subsequent market reaction, have limited the effectiveness of DAC. Included

in these areas are the circumventions and loopholes noted above. As can be seen from section 2.1.3,

a large number of the circumventions relate to the reporting requirements designed in the

provisions. The literature indicates that the current provisions of DAC could, to some extent, be more

appropriate and relevant so that reporting requirements and exchanges of information could be

fulfilled effectively. In particular, the design of the requirements under DAC2 poses several

challenges, principally around definitional issues, exclusions, and due diligence requirements.

In addition, an overarching consideration in this regard is the extent to which compliance is

observed and enforced. A relevant provision can be rendered ineffective if incentives to comply are

limited and enforcement in the case of non-compliance is either not applied or not identified in the

first place. Moreover, even in a hypothetical setting where all provisions are relevant and

appropriate, practical obstacles can obstruct their effective implementation.

The Research Paper identified and categorised several practical and legal obstacles both at the

Directive and at a cross-cutting level in section 2.2. The cross-cutting issues centred around t he TIN

and the quality of reporting, with many of the Directive-specific obstacles relating to these and

timeliness (detailed below). The lack of the TIN in some cases and the accuracy of reporting can

cause delays in the process and be onerous for TAs to address.

In respect of legal obstacles, there were five key issues identified:

1 Purpose of exchanges – exchanges are for tax purposes only and rely on the sending Member

State providing authorisation voluntarily.

2 Business secrets –there is an exemption for the exchange of information that would be

considered a “business secret”.

3 Public disclosure – there are no public disclosure requirements under DAC3 which impact

the objective of transparency.

4 Legislative changes – there is wide disparity in how and when Member States transpose

amendments to the Directive into national legislation.

5 Legal interpretation – qualitative evidence indicates there can be a conflict between how the

Directive and national legislation define the same concepts, for example an APA for DAC3.

Timeliness

The 2019 EC evaluation had already highlighted timeliness as a key concern affecting the

effectiveness of the Directive and the literature and interviews with Member States considered in

this Research Paper support this. Timeliness remains a notable issue for DAC, and is illustrated in the

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

121

case study, where the reporting lags made it difficult for TAs to identify cum-ex and cum-cum

schemes in a timely manner.

The concerns around timeliness for reporting mainly concern the AEOI provisions but also extend

to EOIR, albeit to a lesser extent. In respect of EOIR, TAs must respond within two months (or six

months if the data needs to be retrieved from elsewhere). However, there are no time limits for any

follow-up exchanges which creates the potential for delays. Nevertheless, the extent of this issue

cannot be assessed on the qualitative evidence alone.

5.1.2. Evolution of information exchanges

Although it was not possible to undertake a detailed analysis of the evolution of information

exchanges (AEOI, SEOI and EOIR) between TAs as access to such data was not granted by Member

States, the results of the Economisti Associati study

177

and selected data from a number of Member

States show that the volume information being exchanged has grown considerably, especially

following the coming into force of DAC2. Exchanges of DAC2 information dwarf by now the DAC1

exchanges.

The evolution of within EU-27 cross-border holdings of financial assets was examined to determine

whether changes in such cross-border holdings could be observed at the time that DAC2 was

adopted or became effective. However, none of the data from the BIS, ECB, IMF show any significant

changes in cross-border holdings.

5.1.3. Coherence

DAC, OECD CRS and FACTA

The DAC2 transposes into EU law the various disposition of the OECD CRS and is entirely consistent

with the latter.

The scope of DAC2 (and the OECD CRS) is broader than that of FATCA as FATCA focuses only on

foreign asset holdings by US persons (natural or legal) and US residents holding financial assets

outside the US. In contrast, the focus of the CRS is all residents of a jurisdiction (from a tax

perspective) holding financial assets in other jurisdictions which have signed a CRS agreement with

the home jurisdiction of the residents.

Another important difference is that non-debt direct investments in real property which are within

the scope of FATCA are formally excluded from the CRS. However, such cross-border investments

within the EU are covered by DAC1.

AML provisions

The 2019 evaluation reported evident synergies between DAC and the EU AML provisions, and these

are reflected in this Research Paper. Overall, the provisions are coherent and to a large extent

enhance the functioning of DAC, in particular that of DAC2. Nevertheless, several challenges remain

with respect to the interaction between DAC and AML provisions. The challenges can be grouped

into two categories: definitional issues, and access to and accuracy of information exchanged. With

reference to definitional issues, there is currently no common definition of ‘tax crime' across Member

States, which can affect whether information is held in the first place, and defining “beneficial

ownership” remains a substantial challenge internationally. Variances also remain both between

177

Economisti Associati, April 2019.

EPRS | European Parliamentary Research Service

122

DAC and AML provisions, as well as between Member States, in defining beneficial ownership, and

any inconsistencies can impact the effectiveness of the interaction between the provisions and EOI.

With respect to access to information, this has been impacted by the sub-optimal transposition of

AML Directives, in addition to the underperformance of Member States in relation to AML indicators

and recommendations, based on assessments carried out by FATF. Coupled with this, the accuracy

of the information held can also severely affect the interaction between DAC and AML provisions,

and the effectiveness of exchange of tax information. There are currently no minimum standards for

ensuring the accuracy of information under AML provisions, or a common approach to supervising

AML activities across Member States.

EU legal provisions of judicial assistance.

The interactions between DAC and a) the Protocol established by the Council in accordance with

Article 34 of the Treaty on European Union to the Convention on Mutual Assistance in Criminal

Matters between the Member States of the European Union and b) the Regulation (EU) 2018/1805

of the European Parliament and of the Council of 14 November 2018 on the mutual recognition of

freezing orders and confiscation orders were examined and no inconsistencies between DAC and

the judicial assistance were identified.

However, it was noted that the process of finding out whether a person under criminal investigation

has bank accounts in one or several Member States is facilitated if financial account information

received by TAs can be used in criminal investigations. At the present time, the use for non-tax

matters of the information having been exchanged between TAs requires the authorisation of the

Member State(s) having sent the information. Such authorisation is not always granted, often

because the national laws of the sending Member State(s) do not permit the use of tax related

information in non-tax matters.

5.2. Suggestions

Table 8 below sets out a series of potential areas for improvement flowing from the analysis of DAC1

to 4’s implementation and effectiveness. The table summarises suggestions and emerging findings

identified through an analysis of the available literature, interviews with Member States and through

consultation with other stakeholders, and should not be taken to represent the opinion of the

authors. For each area identified below, several potential adjustments are highlighted, as well as the

underlying basis or source.

5.2.1. Costs of compliance and other considerations

It should be noted that whilst the areas noted below are deemed to potentially have benefits for the

functioning of the Directive based on the research, other aspects outside the scope of this Research

Paper should be considered when looking at amending the Directive, some of which were captured

in the 2019 EC evaluation. In particular, DAC1 and 2 have created high compliance costs for obliged

entities and taxpayers, while DAC3 and 4 have increased compliance costs for companies and MNEs.

Indeed, the compliance burden has also been borne by the TAs as new mechanisms have been

implemented and recurring costs have emerged. Concerns over the costs of complying with the

requirements of the provisions were noted by stakeholder in the online consultation.

An increase in compliance costs may also have unintended consequences and adverse impacts,

such as increasing costs of capital and investment and consideration should be given to the

coherence and interaction with other Union priorities outside the scope of DAC. The potential

increase in the costs of compliance may however be compensated by other factors, such as an

increase in tax revenues or improved internal processes. Therefore, the additional costs of

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

123

compliance should be weighed against any potential benefits derived from the amendments. Given

the potential for an increased burden on various stakeholder groups, although out of scope for this

paper, a full cost-benefit analysis of any changes considered would be essential to thoroughly assess

the efficiency and appropriateness of the measures. Indeed, as with proposals for DAC 7 and 8, any

further proposals to amend the Directive should be accompanied by an impact assessment to fully

understand the potential consequences of the amendment.

Table 8: Areas where DAC effectiveness can be improved and associated suggestions

(please see footnotes for original source of suggestion)

Area Suggestion Basis

Availability of TIN

data

1.1: Introduce mandatory TIN requirements to improve

data matching and identification across Member

States. This is currently considered in DAC7 proposals

for exchanges related to DAC1 and DAC2

requirements. Ongoing work for TIN verification, such

as 'TIN on Europa' or 'TIN on the Web', should be

leveraged, to the extent possible.

178

Based on literature

review, further

details available in

section 2.2.2.

Due diligence

related to DAC2

requirement

2.1: Enhanced due diligence requirements for tax

residence and beneficial ownership for all accounts,

including those where account holders state residence

in countries with golden visa schemes.

179

2.2: Introduce regular checks to ensure information

held is accurate instead of relying on checks only when

there is a change in circumstances.

180

2.3: Where a Member State is not compliant with FATF

recommendations, FIs should not be able to solely rely

on AML information and should instead be required to

conduct their own independent checks

181.

Based on the

literature review,

further details

available in section

2.1.3.

Reporting

requirements

3.1: Under DAC1: Remove the availability criterion and

introduction of reporting requirement for 5 out of 5

categories.

182

3.2: Under DAC2: Where there is no information to

report by a reporting FI it should then file nil returns.

183

3.3: Under DAC2: Remove the thresholds on the value

of accounts that are required to report information.

184

3.4: Under DAC2: Include a marker to signal joint

ownership of different account holder to avoid

Reporting

Requirement

suggestion 3.1 is

based on the

literature review,

further details

available in section

2.1.2.

Reporting

Requirement

178

See Economisti Associati, 2019.

179

See Andres Knobel, 2018.

180

As above.

181

As above.

182

See Economisti Associati, 2019.

183

See Andres Knobel, 2018.

184

As above.

EPRS | European Parliamentary Research Service

124

Area Suggestion Basis

duplicate reporting and to facilitate accurate

identification of account balances.

185

3.5: Under DAC2: In addition to having a flag for joint

accounts, entities could record the ownership share of

each account holder and flag when an account is held

by owners from different jurisdictions.

186

3.5: Under DAC3: The structure and presentation of

information provided by Member States should be

standardised.

187

3.6: Under DAC4: Extend the scope of information

provided by MNE that own several entities within the

same jurisdiction beyond providing only aggregate-

level information.

188

3.7: DAC can improve transparency beyond the

information that is made available to TA, by publishing

more information on statistics, non-compliance,

rulings and corporate disclosures.

189

3.8: Requirement to establish ownership for active

entities as well as passive NFEs (e.g. by applying the

look through principle) at both the entity level and the

Beneficial Owner level.

190

3.9: Lower the 25 % ownership threshold for classifying

Beneficial Owners.

191

suggestions 3.2, 3.3,

3.7 and 3.9 are based

on the literature

review and cross-

analysis of DAC and

AML legislation,

further details

available in section

2.1.3.

Reporting

Requirement

suggestions 3.4 to

3.6 are based on the

Member State

interviews, further

details available in

section 2.1.3.

Coverage

4.1: Under DAC2: When assets are sold, the sales and

purchase price should be reported, to the extent

possible. At the present time it is not possible to

determine whether the seller made any capital gains

(which would be taxable in some jurisdictions).

192

4.2: Under DAC2: Include FIs that are currently

excluded within scope, e.g. electronic money

institutions.

193

4.3: Under DAC2: Include cryptocurrencies and e-

accounts under scope - currently in the DAC 8

proposals.

194

All Coverage

suggestions are

based on the

literature review and

analysis of DAC

legislation, further

details available in

section 2.1.3.

185

Based on responses during the Member State interviews.

186

As above.

187

As above.

188

As above.

189

See Andres Knobel, 2018.

190

As above.

191

As above.

192

As above.

193

As above.

194

As above.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

125

Area Suggestion Basis

4.4: Under DAC2: Commodities including precious

metals should be included as financial assets if held

through an account at a broker or trading platform.

195

4.5: Under DAC2: Accounts at larger peer-to-peer

lending platforms, crowdfunding platforms, and

similar, etc. should be covered by DAC2.

196

4.6: Under DAC3: Include non-cross-border rulings and

APAs that are likely to be of interest to other Member

States.

197

4.7: Under DAC3: Include natural persons within

scope.

198

4.8: Under DAC3: Include ruling and APAs that are not

advanced.

199

4.9: Under DAC3: Include rulings or APAs issued,

amended or renewed before 2012 but still valid.

200

Monitoring and

evaluation

5.1: Develop a common or recommended approach for

assessing/evaluating the impact of the different

DACs.

201

Based on multiple

responses to the

Member State

interviews.

Simplicity and

certainty

6.1: Standardise the approach to the foreseeable

relevance standard by introducing guidelines that can

facilitate Member States to exchange information.202

6.2: Simplify the exchange of information for other

purposes, like AML, by setting timelines around

providing authorization to share this type of

information.

203

6.3 Simplify the compliance requirements for

individual taxpayers and companies/MNEs, in order to

increase compliance and manage the compliance

burden.

204

All Simplicity and

certainty

suggestions are

based on the

Member State

interviews and the

stakeholders survey,

as well as cross-

analysis of the DAC

legislation, see

section 2.2.1 for

further details.

195

As above.

196

See Andres Knobel, 2018.

197

As above.

198

As above.

199

As above.

200

As above.

201

Based on responses during the Member State interviews.

202

As above.

203

As above.

204

Based on responses to the stakeholder survey.

EPRS | European Parliamentary Research Service

126

6. References

Agencia Tributaria, Principales Resultados 2019, Resultados económicos y tributarios en el Impuesto

sobre el Valor Añadido, 2020

Ahrens, L. and Bothner, F., ‘

The Big Bang: Tax Evasion After Automatic Exchange of Information Under

FATCA and CRS’, New Political Economy, 25:6, 2020, pp. 849-864

Austrian Beneficial Owners Register Act, Federal Act on the establishment of a register of beneficial

owners of companies, other legal entities and trusts (Beneficial Owners Register Act – WiEReG;

Wirtschaftliche Eigentümer Registergesetz), 2017

Beer, S., Coelho, M. and Leduc, S., Hidden Treasures: The Impact of Automatic Exchange of Information

on Cross-Border Tax Evasion, IMF Working Paper Fiscal Affairs Department, WP/19/286, 2019

Casi, E., Spengel, C. and Stage, B., ‘

Cross-border tax evasion after the common reporting standard: Game

over?‘, Journal of Public Economics, Volume 190, October 2020

Council Directive 2003/48/EC on the taxation of savings income in the form of interest payments, of 3

June 2003

Council Directive 2014/107/EU amending Directive 2011/16/EU as regards mandatory automatic

exchange of information in the field of taxation, 9 December 2014

Cour des Comptes, Échange automatique

de données fiscales au niveau international. Deuxième rapport

d’évaluation à la demande de la commission Panama, November 2020

Dayez, A., DAC2/CRS: multireporting and data protection, AEFI meeting, November 2016

De Simone, L., Lester, R. and Markle. K., ‘Transparency and Tax Evasion: Evidence from the Foreign

Account Tax Compliance Act (FATCA)‘, Journal of Accounting Research, Volume 58, Issue1, March 2020,

pp. 105-153

Deutscher Bundestag, Drucksache 19/20953, Schriftliche Fragen mit den in der Woche vom 6. Juli 2020

eingegangenen Antworten der Bundesregierung, 19. Wahlperiode, 6 July 2020

Deutscher Bundestag, Antwort der Bundesregierung auf die Kleine Anfrage der Abgeordneten Fabio De

Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion DIE LINKE, Drucksache

19/17231, Andauernde Umsetzung des Informationsaustauschs in Steuersachen (Nachfrage zur Antwort

der Bundesregierung auf die Kleine Anfrage auf Bundestagsdrucksache 19/13797) 19. Wahlperiode,

16 June 2020

Deutscher Bundestag, Antwort

der Bundesregierung auf die Kleine Anfrage der Abgeordneten Fabio De

Masi, Jörg Cezanne, Klaus Ernst, weiterer Abgeordneter und der Fraktion Die Linke. – Drucksache

19/12953 – Andauernde Umsetzung des Informationsaustauschs in Steuersachen, 19.

Wahlperiode,

3 September 2019

Duncan Wigan, Case study: The Cum-Cum and Cum-Ex schemes, COFFERS D4.5, September 2019

http://coffers.eu/wp-content/uploads/2020/03/D4.5-Case-Study.pdf

Estonia, Tax Information Exchange Act, translation published April 2019

Eurodad, EU Country by country reporting: Overview of the political process and existing country by

country reporting, Financial Transparency Coalition, Briefing paper, December 2019

European Banking Authority, Report on Competent Authorities’ Approaches to the Anti-Money

Laundering and Countering the Financing of Terrorism Supervision of Banks, June 2020

European Commission, Anti-money laundering Directive IV (AMLD IV) – transposition status

, European

Commission website

European Commission, Anti-money laundering Directive V (AMLD V) – transposition status, European

Commission website

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

127

European Commission, Repeal of the Savings Directive in line with international and EU developments,

European Commission website

European Commission, Communication from the Commission to the European Parliament and the

Council, An Action Plan to strengthen the fight against tax fraud and tax evasion, {SWD(2012) 403 final},

{SWD(2012) 404 final}, COM(2012) 722 final 6 December 2012

European Commission, EU savings taxation rules and savings agreements with third countries: frequently

asked questions, Memo, Memo

, 2014

European Commission, Working Document of the Expert Group on Automatic Exchange of Financial

Account Information, 2014

European Commission, First Report of the Commission AEFI expert group on the implementation of

Directive 2014/107/EU for automatic exchange of financial account information, 2015

European Commission (DG TAXUD), Working Document on the Implementation of Directive

2014/107/EU, WPIV Meeting on 19 June 2015

European Commission, Report from the Commission to the European Parliament and the Council on the

application of Council Directive (EU) 2011/16/EU on administrative cooperation in the field of direct

taxation, COM(2017) 781 final

, December 2017

European Commission, Staff Working Document on the application of Council Directive (EU) no

2011/16/EU on administrative cooperation in the field of direct taxation, SWD(2017) 462 final

December 2017b

European Commission (DG TAXUD), Note on Council Directive 2018/822/EU of 25 May 2018

European Commission, Report from the Commission to the European Parliament and the Council on

overview and assessment of the statistics and information on the automatic exchanges in the field of

direct taxation, COM(2018) 844 final

, December 2018

European Commission, Report from the Commission to the European Parliament, the Council the

European Economic and Social Committee and the Committee of the Regions, Investor Citizenship and

Residence Schemes in the European Union, COM(2019) 12 final

, January 2019

European Commission, Directorate-General Taxation and Customs Union, Evaluation of Administrative

Cooperation in Direct Taxation Final Report prepared by Economisti Associati in collaboration with

ECOPA and Oxford Consulting, April 2019

European Commission. Evaluation of the Council Directive 2011/16/EU on administrative cooperation in

the field of taxation and repealing Directive 77/799/EEC, SWD(2019) 327 final

, September 2019

European Commission, Annual Report on European SMEs 2018/2019 - Research & Development and

Innovation by SMEs, SME Performance Review 2018/2019, Report prepared by LE Europe in collaboration

with DIW Econ, PwC Luxembourg and CARSA, November 2019

European Commission, Communication from the Commission on an Action Plan for a comprehensive

Union policy on preventing money laundering and terrorist financing, C(2020) 2800 final

, May 2020

European Commission. Proposal for a Council Directive amending Directive 2011/16/EU on

administrative cooperation in the field of taxation, COM(2020) 314

, July 2020

European Parliament, The Cum-ex files, Information document, November 2018

European Parliament, Citizenship by Investment (CBI) and Residency by Investment (RBI) schemes in the

EU, State of play, issues and impacts, Study

, EPRS, October 2018

FATCA Annex II to Model 1 IGA, US Treasury website

FATCA Model 1A IGA Reciprocal, Pre-existing TIEA or DTC, US Treasury website

Financial Action Task Force, FATF Guidance: Transparency and Beneficial Ownership, Financial Action

Task Force, 2014

Financial Action Task Force, 4th Round Ratings, November 2020

Foreign Account Tax Compliance Act (FATCA), 2010

EPRS | European Parliamentary Research Service

128

International Capital Market Association, Savings Tax Directive, ICMA website

Kelly, J., ‘International Tax Regulation by United States Fiat: How FATCA represents unsound international

tax policy’, Wisconsin International Law Journal, Vol. 34 Issue 4, May 2017

Knobel A., ‘How to assess the effectiveness of automatic exchange of banking information’, article, Tax

Justice Network, December 2018b

Knobel A., Reporting taxation: Analysing loopholes in the EU’s automatic exchange of information and

how to close them, A Report Commissioned by the Greens/EFA Group in the European Parliament,

October 2018

KPMG, UBO disclosure requirements within the EU, Global Legal Services, April 2019

O’Reilly, P., K. Parra Ramirez and M. A. Stemmer, ‘Exchange of Information and Bank Deposits in

International Financial Centres’, OECD Taxation Working Papers, No. 46, OECD Publishing, 2019

OECD, International Framework for the CRS

, OECD website

OECD, Preventing abuse of residence by investment schemes to circumvent the CRS, Consultation

document, March 2018

OECD, Standard for Automatic Exchange of Financial Information in Tax Matters, Implementation

Handbook, Second Edition, 2018

Rahman Ravelli, Cum-Ex Trading Schemes Explained & FAQs, downloaded on 20 December 2020 from

https://www.rahmanravelli.co.uk/articles/cum-ex/

Rechnungshof Österreich, Internationaler Informationsaustausch in Steuerangelegenheiten, Bericht des

Rechnungshofes, Reihe BUND 2019/33, August 2019

of the Working Group (and its sub-groups) of Administrative Cooperation in the field of Direct Taxation,

República Portuguesa, Relatório sobre o Combate à Fraude a Evasão Fiscais e Aduaneiras 2019, Cabinete

do Secretário de Estado Adjunto e dos Assuntos Fiscais, July 2020

Schaper, M., ‘Data Protection Rights and Tax Information Exchange in the European Union: An Uneasy

Combination‘, Maastricht Journal of European and Comparative Law, June 2016

Seer R. and Kargitta, S., Exchange of information and cooperation in direct taxation, Research Handbook

on European Union Taxation Law, 2020

Statista, Number of small and medium-sized enterprises (SMEs) in the European Union in 2018 by size,

September 2020

Suso, A.M., ‘The boundaries of abusive practices: the grey zone‘, ECJ – Recent Developments in Value

Added Tax: The Evolution of European VAT Jurisprudence and Its Role in the EU Common VAT System,

Linde Verlag, Schriftenreihe, 2014

Tax Free Today, Living tax free in Austria through corporate structure in Cyprus, 23 October 2019

Tax Justice Network, EU Savings Tax Directive, Tax Justice Briefing

, 2008

Tokola, A., ‘Tax evasion in international investment – Messages from a Finnish perspective’, Tax

Administration, Grey Finance Clearing House, September 2019

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

129

Appendix 1: Analytical approach and methodology

The evaluation matrix and information sources

Table 9 to Table 11 below show, for each of the specific questions set out in the ToR for the

assessment of the various aspects of the effectiveness of DAC (Q1), the analysis of the evolution of

tax information exchanges (Q2) and the coherence of DAC (Q3), the key information sources which

the team is relied on to gather the information required to address each of the questions specified

in the terms of reference of the study.

Table 9: Information sources to be used for Q1- Effectiveness of DAC

Question Sub-question Information sources

Interviews with

officials from DG

Taxud, National

Tax

Administrations

Stakeholder

on-line

survey

Literature

review

DAC

Did the definition of income

evolve with the adoption of

DAC? Were there

possibilities appearing in

the market to earn and/or

design income in order to

not respect the spirit of the

DAC? Are there some types

of income still not covered

by DAC?

DAC

Report on legal and

practical

obstacles/restrictions

identified in the exchange

of information (automatic,

spontaneous and upon

request).

DAC

Report on whether

information exchange upon

request was hampered by a

lack of availability of

information in the

requested country and if

yes, for which types of

income and whether the

reasons for this non-

availability were provided

and checked.

EPRS | European Parliamentary Research Service

130

Question Sub-question Information sources

DAC2

Circumventions and

loopholes

1. Known examples of circumvention and

loopholes per type of income/tax not covered,

e.g. capital gains/capital gains tax, including

dividend arbitrage (lessons learnt from the

international cum-ex scandal) and similar cases

of tax avoidance (cum-cum, cum-fake, etc.);

real estate; free ports; bilateral special cases:

e.g. AUT-CY; direct investment avoiding the

creation of a “financial account”; family offices

based on structures without an UBO as no

family member holds 25% or more

2. Circumvention and loopholes with respect

to beneficial ownership, e.g. through incorrect

application of the “look through” principle

with corporate entities, trusts or the like.

Reporting on the development over time of

the number of entities registered in the

respective jurisdictions. Second residence or

golden residence schemes

3. Checks and sanctions for the completeness

and quality of reported data

DAC2

Clarify and consider the

appropriateness of

conditions for automatic

information exchange,

especially:

1. Financial accounts: what is seen as a

financial account and what is not?

2. Beneficial owners: who is a beneficial owner

and who is not? Who has to be reported?

- Reporting

thresholds

- Correctness of

tax residence:

Implementation

of the look

through

approach by

obliged entities

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

131

Question Sub-question Information sources

3. Incentives for obliged reporting entities to

report accurately

DAC2

Use of macroeconomic data

such as national accounts,

Eurostat, IMF and locational

banking data (BIS) to

measure

Mapping DAC2

data against the

other data

sources -

checking for

plausibility

DAC2

Scanning tax evasion and

tax avoidance market for

models for circumvention

on offer

DAC2

The contractor shall detail

how information exchanged

under DAC2 have been

used by Member States and

what are the fiscal effects

seen in Member States (e.g.

impact on revenue

collection).

DAC2

(?)

If available, the contractor

shall provide percentage of

the data used by Member

States.

DAC3

Evaluate circumventions

and loopholes for Advanced

Pricing Arrangements and

oral rulings.

DAC3

Mapping of bilateral flows

of information (depending

on data availability)

DAC4

Evaluate circumventions

and loopholes

DAC4

Mapping of bilateral flows

of information (depending

on data availability)

Other

How the cum-ex scandal was possible and how

an improved exchange of information

framework could prevent a similar scandal in

the future?

EPRS | European Parliamentary Research Service

132

Table 10: Information sources to be used for Q2- Analysis of the evolution of tax information

exchanges

Question / task

Sub-question

Information sources

Interviews with

officials from

DG Taxud,

National Tax

Administrations

Stakeholder

on-line

survey

Review of

publicly

available

data

1. Assessment of the

evolution of the tax data

exchanges year by year

to at least 2018 (or

better 2019)

Assessment shall cover at

least automatic, spontaneous

and exchanges on request

Analysis of statistics collected

under DAC and assessment of

whether some improvements

could be made

2. Contractor shall assess

as much as possible DAC

exchanges for group

request by sending and

receiving country and

type of information over

time

Assessment of whether there

are limitations to the

response by requested

Member States to reply to

group requests (e.g. define

these as fishing expeditions).

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

133

Table 11: Information sources to be used for Q3 - Coherence

Question / task

Information sources

Interviews with officials

from DG Taxud, National

Tax Administrations

Stakeholder

on-line survey

Literature

review

1. Explain (if any) the differences between

the OECD CRS standard and the DAC as well

as differences between DAC and FATCA (US)

2. How does DAC interact with EU rules on

judicial assistance, money laundering and

asset recovery?

3. Are there restrictions imposed by the

sending countries on the receiving ones not

to use requested information for

AML/financial crime purposes

4. Analysis of the implementation of the

current reporting obligations and tax

information exchange flows by countries

and jurisdictions previously covered by the

EU Savings Tax Directive or under similar

obligations in an EU agreement (Andorra,

Bahamas, Bermuda, Cayman Islands

Liechtenstein, Monaco, San Marino, and

Switzerland)

5. Exchange of information with United

States (how does the information exchange

work with the US, differences FATCA/DAC,

analysis of the reciprocity including in the

IGAs under FATCA, loopholes or obstacles)

Literature review

The search for publicly available literature relevant for this study followed the research team’s tried

and tested three-steps approach to such an exercise.

Step 1. Development of literature search parameters and identification of potential sources. The

objective of this step is the identification of all potentially relevant literature. It involved the

searching of:

appropriate electronic/internet and print sources (e.g. journals, textbooks, research reports, EC

publications, publications from other international organisations (e.g., IMF, OECD, etc.),

government publications and publications from stakeholders); and

'grey' (i.e. unpublished) literature.

EPRS | European Parliamentary Research Service

134

Step 2. Application of exclusion criteria and initial filter of the various pieces of literature. The

objective of this step is to ensure that the existing literature has been sifted for quality and validity

using the criteria below and is entirely relevant.

The criteria used in this step included:

The appropriateness and relevance of the questions/issues covered by the piece of literature;

and

The relevance in terms of timeliness.

During this stage, the various literature pieces were reviewed and only those meeting the three

criteria listed above were used as information sources for responding to the various research

questions.

Step 3. Completion of the search for relevant material. The objective of this step is to ensure that all

the relevant literature has been identified.

To ensure that no important information pieces were missed, the articles/books/documents often

cited by the articles, studies, reports, etc. that had not been excluded for the review in the previous

step of the literature review were also reviewed.

Interviews of officials

Relevant officials from DG TAXUD and a number of Member States were invited to participate in a

phone or video interview. These included officials from the Member States listed in the terms of

reference of the study, namely Austria. Belgium, France, Germany, Latvia, Luxembourg, and

Portugal. Officials from Austria and Germany declined to participate.

In addition, officials from five additional Member States (Italy, Malta, Slovakia, Spain and Sweden)

were also interviewed and the Dutch officials provided a written response.

The interviews were based on the interview guide attached in the appendix. Some officials

requested a shorter interview and both the long and short version of the interview guides are

provided in Appendix 4. The interviews typically lasted from 60 to 90 minutes.

Survey of stakeholders

In addition to consultations of officials, an on-line survey of various groups of stakeholders from civil

society, consumers and various segments of the business and finance community was also run. The

precise list of stakeholders invited to participate in the survey is provided in Table 12.

The survey was rolled out during the weeks of 12 and 19 October and a first reminder was sent on

27 October and a second reminder on 13 November. Qualitative information was received from 3

stakeholders and declines from another 10. Reasons given for not participating included lack of time

(some due to Covid-19 crisis) and having nothing to contribute).

Table 12: List of stakeholders invited to participate in the on-line survey

Economic sector of stakeholder Stakeholder

Accounting /audit

European Associations

Accountancy Europe

European Accounting Association

European Federation of Accountants and Auditors for SMEs

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

135

Economic sector of stakeholder Stakeholder

Individual accounting organisation

Consiglio Nazionale dei Dottori Commercialisti e degli Esperti

Contabili (CNDCEC), the public sector organisation that

represents the Italian accountancy profession

Individual firm

KPMG

Tax advisory

European Associations

CFE Tax Advisers Europe

ETAF - European Tax Adviser Federation

Banking

European Associations

European Association of Co-operative Banks (EACB)

European Banking Federation (EBF)

Individual Banks

BNP Paribas

UBS

Insurance industry

Insurance Europe

Investment Fund/Asset Management

industry

European Association

European Fund and Asset Management Association (EFAMA)

Individual Asset Managers

Aberdeen Standard Investment

Aegon Group

Amundi Asset Management

Natixis Investment Managers

Private capital industry

European Associations

Invest Europe

European Business Angels Network (EBAN)

Individual Private Equity firms

Bridgepoint

PAI Partners

Stock exchanges

European Association

Federation of European Securities Exchanges

World Association

World Federation of Exchanges

Pension provider industry

European Association

Pensions Europe

Individual Pension Funds

Arbejdsmarkedets Tillaegspension (ATP)

Wholesale financial industry

European Association

Association for Financial Markets in Europe (AFME)

Retail investors

European Association

Better Finance (European Federation of Investors and

Financial services Users

Fintech

European FinTech Alliance (EFA)

Business

Business Europe

EPRS | European Parliamentary Research Service

136

Economic sector of stakeholder Stakeholder

(has circulated invitation to participate in the survey to all its

members)

SMEunited (European Association of Craft, Small and Medium-

Sized Enterprises)

Consumers

European association

BEUC

NGOs

Action Aid

European Network on debt and development (eurodad)

Oxfam

Social Platform

Transparency International

Action Aid

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

137

Appendix 2: Comparison of key features of FATCA and CRS

This annex compares the main features of FATCA and CRS with regards to:

Reporting FIs

Types of financial accounts which are reportable if held by one or more Reportable Persons or by

a Passive Non-Financial Entity with one or more Controlling Persons that is a Reportable Person

Reportable Persons (natural or legal) whose account(s) are reportable

Information to be reported and exchanged

The comparison below draws heavily on the OECD Standard for Automatic Exchange of Financial

Information in Tax Matters Implementation Handbook and information available on the IRS

website.

205

Table 13: Reporting FIs

Commentary

FATCA

CRS

Type of financial institution

defined as a Reporting FIs

within the scope of FATCA/CRS

Except for investment entities, the

concepts of the different institutions

are similar.

The definition of an investment

entity in FATCA differs from that in

the CRS. It is less prescriptive.

206

Moreover, the conditions that have

to be met to qualify as a collective

investment vehicle under the CRS

differ slightly from those set out in

FATCA – the FATCA conditions were

amended in the CRS to ‘take

account of the multilateral context,

remove US specificities and the

consequential changes to the

definition of Reportable Persons’.

207

Reporting FIs include

208

• ‘Depository institutions

any entity that accepts

deposits in the ordinary

course of a banking or similar

business’

• ‘Custodial institutions:

any entity that holds, as a

substantial portion of its

business, financial assets for

the account of others. An

entity holds financial assets

for the account of others as a

substantial portion of its

business if the entity’s gross

income attributable to the

holding of financial assets and

related financial services

equals or exceeds 20 percent

of the entity’s gross income

during the shorter of: (i) the

three-year period that ends

on December 31 (or the final

day of a non-calendar year

accounting period) prior to

the year in which the

Reporting FIs include

209

• ‘Depository Institutions:

generally, includes savings

banks, commercial banks,

savings and loan

associations and credit

unions’

• ‘Custodial Institutions:

generally, includes

custodian banks, brokers

and central securities

depositories’

205

See OECD Standard for Automatic Exchange of Financial Information in Tax Matters, Implementation Handbook,

Second Edition, OECD, 2018 (the OECD Implementation Handbook 2018).

206

See OECD Implementation Handbook 2018, p. 127.

207

See OECD Implementation Handbook, p. 128.

208

See US Department of the Treasury, Model 1 Intergovernmental agreements listed on

https://home.treasury.gov/policy-issues/tax-policy/foreign-account-t ax-compliance-act

209

See OECD Implementation Handbook 2018, p. 61.

EPRS | European Parliamentary Research Service

138

Commentary

FATCA

CRS

determination is being made;

or (ii) the period during which

the entity has been in

existence’

• ‘Investment Entities:

Entities that conduct as a

business (or is managed by an

entity that conducts as a

business) one or more of the

following activities or

operations for or on behalf of

a customer: (1) trading in

money market instruments

(cheques, bills, certificates of

deposit, derivatives, etc.);

foreign exchange; exchange,

interest rate and index

instruments; transferable

securities; or commodity

futures trading;

(2) individual and collective

portfolio management; or

(3) otherwise investing,

administering, or managing

funds or money on behalf of

other persons’

• ‘Specified Insurance

Companies:

any Entity that is an insurance

company (or the holding

company of an insurance

company) that issues, or is

obligated to make payments

with respect to, a Cash Value

Insurance Contract or an

Annuity Contract’

• ‘Investment Entities:

generally, includes Entities

investing, reinvesting or

trading in financial

instruments, portfolio

management or investing,

administering or

managing financial assets’

• ‘Specified Insurance

companies:

Generally, includes most

life insurance companies’

Institutions defined as Non-

Reporting Financial Institution (i.e.

not subject to FATCA/CRS

reporting)

According to the OECD, ‘the entities

listed as Non-Reporting Financial

Institutions in the CSR are largely

consistent with the Non-Reporting

entities listed in the FATCA Model I

Annex II IGA even though Annex II

Non-Reporting FIs include

215

‘1 Governmental Entities

2. International Organisations

3. Central Banks

4. Funds that Qualify as Exempt

Beneficial Owners such as:

Non-Reporting FIs are those

which are at a low risk of being

used to evade tax. They

include

216

‘1. Governmental Entities, and

their pension funds

2. International Organisations

3. Central Banks

215

See FATCA Model 1 IGA Annex II.

216

See OECD Implementation Handbook 2018, Figure 7, p. 61.

Implementation of EU requirements for tax information exchange: Progress, lessons learnt and obstacles to

overcome

139

Commentary

FATCA

CRS

sets out more detailed non-

reporting sub-categories’.

210

The conditions that have to be met

to qualify as a collective investment

vehicle under the CRS differ slightly

from those set out in FATCA – the

FATCA conditions were amended in

the CRS ‘to take account of the

multilateral context, remove US

specificities and the consequential

changes to the definition of

Reportable Persons’

211

The CRS provides for the set of low-

risk financial institutions to be

determined under domestic law

while the set of non-reporting

financial institutions under FATCA is

agreed through bilateral discussion.

Overall, it is likely that the former

group is subsumed by the second

group but there is no guarantee that

this always fully the case.

212

The following FATCA types of

institutions are not specified in the

CRS: ‘Treaty-Qualified Retirement

Fund, Investment Entity Wholly

Owned by Exempt Beneficial

Owners, Financial Institution with a

Local Client Base, Local Bank,

Financial Institution with Only Low-

Value Accounts, Sponsored

Investment Entity and Controlled

Foreign Corporation, Sponsored,

Closely Held Investment Vehicle,

Investment Advisors and Investment

Managers’

213

‘The category Financial Institution

with Only Low-Value Accounts is not

specified in the CRS because it relies

on a threshold of $50,000. Such

threshold is not present in the CRS.

• Treaty-Qualified Retirement

Fund.

● Broad Participation

Retirement Fund

● Narrow Participation

Retirement Fund

● Pension Fund of an Exempt

Beneficial Owner

● Investment Entity Wholly

Owned by Exempt Beneficial

Owners

5. Small or Limited Scope Financial

Institutions that Qualify as Deemed-

Compliant FFI

● Financial Institution with a

Local Client Base

● Local Bank.

● Financial Institution with Only

Low-Value Accounts

6. Qualified Credit Card Issuer

7. Investment Entities that Qualify as

Deemed-Compliant FFIs and Other

Special Rules such as

● Trustee-Documented Trust.

● Sponsored Investment Entity

and Controlled Foreign

Corporation

● Sponsored, Closely Held

Investment Vehicle

● Investment Advisors and

Investment Managers

● Collective Investment Vehicle’

4. Certain Retirement Funds

5. Qualified Credit Card Issuers

6. Exempt Collective Investment

Vehicles

7. Trustee Documented Trusts

8. Other low-risk Financial

Institutions’

The general category of ‘Other

Low-risk Non-Reporting FIs’. is

a list of jurisdiction specific FIs

that are excluded from

reporting provided they meet

certain conditions, including

that their categorisation as

such does not frustrate the

purposes of the Standard.

These jurisdiction-specific FIs

‘can be considered as posing a

low risk for tax evasion and

they have substantially similar

characteristics to the

categories of Non-Reporting

FIs contained in the CRS’.

217

210

See OECD Implementation Handbook 2018, p.128.

211

See OECD Implementation Handbook 2018, p.128.

212

See OECD Implementation Handbook 2018, p.128.

213

See OECD Implementation Handbook 2018, page 129.

217

See OECD Implementation Handbook 2018, paragraph 125.

EPRS | European Parliamentary Research Service

140

Commentary

FATCA

CRS

The categories Treaty-Qualified

Retirement Fund, Financial

Institution with a Local Client Base

and Local Bank are not specified by

the CRS because do not ‘translate in

a multilateral setting’.

The category Investment Entity

Wholly Owned by Exempt Beneficial

Owners is not specified in the CRS

because none of their direct account

holders are persons that trigger a

reporting obligation.