EUROPEAN COMMISSION

Brussels, 19.6.2023

SWD(2023) 216 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT REPORT

New EU system for the avoidance of double taxation and prevention of tax abuse in the field of withholding taxes (WHT)

Accompanying the document

Proposal for a Council Directive

on Faster and Safer Relief of Excess Withholding Taxes

{COM(2023) 324 final} - {SEC(2023) 243 final} - {SWD(2023) 215 final} - {SWD(2023) 217 final}

Table of contents

Table of contents

1Introduction: Political and legal context

2Problem definition

3Why should the EU act?

4Objectives: What is to be achieved?

5What are the available policy options?

6What are the impacts of the policy options?

7HOW DO THE OPTIONS COMPARE?

8Preferred option

9How will actual impacts be monitored and evaluated

Annex 1: Procedural Information

Annex 2: Stakeholder Consultation

Annex 3: Who is affected and how?

Annex 4: Analytical methods

Annex 5: 2021 JRC Study: Quantifying the Costs of Withholding Tax Relief Procedures

Annex 6: JRC Study – Effect of non-refundable WHT on FPI and macro-economic indicators

Annex 7: Withholding tax systems and reporting obligations in the Member States

Annex 8: Comparison domestic/conventional WHT rate over dividends and interest in 2020

Annex 9: Explanation of the Financial industry value chain

Annex 10: Biannual summary report on the Code of Conduct follow-up

Annex 11: Common Reporting Elements and reporting obligation functioning

Annex 12: Technical specifications for setting up an EU-wide digital Tax Residence Certificate

Glossary

1Introduction: Political and legal context

In the EU, investors are generally obliged to pay tax on the income they receive from the holding of securities (namely dividends on holdings of equities and interest on holdings of bonds) twice. First, taxes could be due in the country of the issuer of the securities (the source country) in the form of a tax withheld from the gross securities income, called a withholding tax (WHT ( II )); and then in the country of residence of the investor (the residence country) in the form of an income tax. To avoid this double taxation, many countries have agreed to share taxing rights between the source and the residence countries by signing double tax treaties (DTTs). Under the terms of these treaties, non-resident investors may be entitled to a lower rate of WHT or to an exemption in the source country ( III ).

This reduction or exemption of WHT may be granted in two ways. Either the reduced tax rate or exemption is applied directly at the moment the dividend/interest is paid (relief at source), or the excess tax withheld is refunded on the basis of a reclaim by the investor (refund procedure). However, the WHT refund procedures ( IV ) that allow non-resident investors to benefit from tax treaty benefits ( V ) are often complicated and vary considerably among Member States. As a result, investors incur high costs, receive late refunds or even forego the right to refund altogether. This translates into provisional or permanent double taxation since investors are not able to enjoy the tax treaty benefits to which they are entitled to. This discourages cross-border investment, hinders the functioning of the Single Market, and represents a barrier to the free movement of capital – hence being an obstacle to the building of the Capital Markets Union (CMU).

On the other hand, there has been an abusive utilization of WHT refund procedures as recently demonstrated by the ‘Cum/Ex scandal’ ( VI ), where WHT refunds have been obtained based on multiple WHT requests arising from short selling practices. In September 2020, the European Securities and Markets Authority (ESMA) released a ‘Final report On Cum/Ex, Cum/Cum and withholding tax reclaim schemes’ ( VII ) which concluded that a legislative and supervisory response should be sought within the boundaries of the tax legislative and supervisory framework. The fact that these fraudulent schemes have strongly affected several Member States justifies more than ever the need for action.

The European Commission has been analysing and trying to solve the inefficiencies and fraud risks associated with WHT procedures since the early 2000’s. The “Giovannini Group ( VIII )”, which was set up to advise the Commission on issues related to EU financial integration and the efficiency of financial markets, published in 2001 and 2003 two reports ( IX ), which identified WHT procedures as a barrier to efficient cross-border investments in securities.

Subsequently, the Fiscal Compliance Experts’ Group (FISCO) ( X ) produced a fact-finding study on fiscal compliance procedures related to clearing and settlement within the EU in 2006 and a report on fiscal compliance barriers related to post-trading in the EU in 2007 ( XI ), in which existing WHT procedures were analysed and improvements suggested. Drawing on this work, the Commission put forward in 2009 a Recommendation ( XII ) to the Member States on the simplification of WHT procedures. The Recommendation suggested ways to make WHT procedures simpler and more efficient, for the benefit of all stakeholders involved. In 2017, the Commission published a Code of Conduct on WHT ( XIII ), which called for voluntary commitment by Member States. The Code of Conduct compiles practical approaches to improve the efficiency of WHT procedures in various areas (notably entitlement, relief at source, digitalisation, refund period, documentation requirements and forms).

At international level, the OECD Treaty Relief and Compliance Enhancement (TRACE) initiative aimed to address the inefficiency of WHT procedures. The TRACE Implementation Package (TRACE IP) ( XIV ), approved in February 2013, provided for a framework to set up an authorised intermediary system whereby WHT agents registered as authorised intermediaries are allowed to claim exemptions or reduced rates of WHT on a pooled basis on behalf of their customers. This initiative was inspired by the ‘Qualified Intermediary’ system utilized in the US ( XV ). In the EU, the TRACE IP has been implemented by Finland since 2021.

Even though these numerous actions at EU and international level resulted in some improvements, cumbersome WHT procedures remain a barrier to a well-functioning EU capital market and are prone to fraud risk.

In her political guidelines, President Von der Leyen identified fair taxation and the fight against tax fraud as key foundations of an economy that works for the people ( XVI ). She also stressed the importance of completing the CMU, notably to support economic growth and finance the green and digital transitions.

In the Action Plan for fair and simple taxation supporting the recovery strategy ( XVII ), adopted in July 2020, the Commission announced an initiative in the area of WHT procedures with a view to lower significantly tax compliance costs for cross-border investors while, at the same time, preventing tax abuse. In particular, Action 8 of the Action Plan states that the Commission will propose a legislative initiative for introducing a common, standardised, EU-wide system for withholding tax relief at source.

In its Action Plan on a capital markets union for people and businesses adopted in September 2020 ( XVIII ), the Commission confirmed its intention to put forward a legislative initiative in the area of WHT procedures by establishing a common, standardised, EU-wide system for withholding tax relief at source. This follows the recommendation from the High-Level Forum (HLF) on CMU to adopt a legislative proposal introducing a standardised relief at source system and harmonising definitions/processes related to WHT procedures ( XIX ).

The European Parliament also echoed the necessity to improve WHT procedures in its non-legislative resolution on a European Withholding Tax Framework ( XX ) adopted on 10 March 2022. This resolution recommended putting an end to profit-shifting practices, increasing efforts to combat dividend arbitrage ( XXI ) and removing barriers to cross-border investments in the Single Market.

In light of the above, this initiative aims at making WHT procedures in the EU more efficient and less prone to fraud in order to ensure fair taxation and support cross-border investment, thus contributing to building the CMU. Since intra-group passive income payments (dividends and interests) are already covered by the exemption at source granted under the Parent-Subsidiary Directive ( XXII ) and the Interest-Royalties Directive ( XXIII ), the current initiative focuses on simplifying WHT procedures for both EU and non-EU portfolio investors (hereafter: investors) investing in EU securities markets through EU or non-EU located financial intermediaries.

2Problem definition

2.1What are the problems?

Many countries around the world, including a vast majority of EU Member States, apply WHT on cross-border securities’ income (i.e. dividends and interest). This means that payments received as remuneration of a cross-border investment are subject to a WHT in the country of the investment (the “source country”). The tax is withheld either by the issuer of the securities or by a financial intermediary that is involved in the payment of the income, also known as the withholding tax agent (WHT agent). The tax withheld is remitted to the tax administration of the source country on behalf of the investor/taxpayer.

However, the WHT is generally levied at a rate which is higher than the reduced rate the non-resident investor is entitled to according to the applicable DTT. The reduced WHT rate could also be provided unilaterally by the domestic legislation of the source country (e.g. exemptions for pension funds or charities).

To avoid double taxation, a non-resident investor who would have been subject to a WHT rate that is higher that the DTT rate would submit ex-post a refund claim of the excess tax withheld in the investment country. Often the non-resident invests by means of a chain of financial intermediaries.    

Figure 1: Simplified illustration of the refund procedure

Source: DG TAXUD

Note: The maximum time period to claim the refund may vary depending on the domestic law of each country

The standard WHT refund procedure as depicted in Figure 1 is not the only existing system. Overall, one can distinguish between three main WHT procedure systems to grant non-residents’ access to tax treaty benefits that can coexist in the same Member State and are applicable depending on the risk strategy pursued by this Member State: (i) relief at source (dividend/interest payments are made after applying reduced WHT rates as defined in DTTs or taking into account the applicable exemptions); (ii) quick refund (income is initially paid after applying the maximum WHT rate, and any tax withheld in excess is then refunded within a short period of time); and (iii) standard refund (income is paid after applying the maximum WHT rate and the refund of tax withheld in excess takes place after the tax withheld has been already transferred to the source tax administration).

Member States will always need to have a standard refund procedure in place; for some Member States it is the only way for taxpayers to obtain their refund, for other Member States it acts as the final resort to obtain a refund. In case of the latter situation Member States provide taxpayers with the opportunity to either claim relief under the relief at source and/or quick refund. Member States may also restrict certain taxpayers from using the faster procedures if these taxpayers are considered to present an higher risk of abuse. Annex 7 describes in some detail the system(s) existing in each Member State.

An important element is that, in some Member States, there is no difference between the domestic WHT rate for cross-border dividend or interest payments set out in national rules and the WHT rates laid down in DTTs. Hence, in those cases, there is no need to ask for any refund. However, currently, thirteen Member States have a domestic WHT rate for cross-border dividend payments that is considerably higher than the WHT rate stated in DTTs: Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Italy, Poland, Portugal, Slovenia, Spain and Sweden (see Annex 8 for a comparison between domestic and conventional rates). When referring to interest payments, the number of Member States with a higher domestic rate compared to their conventional rate, is lower as the majority of countries exempt cross-border interest payments under their domestic rules ( XXIV ). Based on targeted consultations with the financial sector and some Member States, the total amount of WHT levied over interest payments has a lower weight in quantitative terms compared to dividend payments. Similarly, in the 2021 WHT survey, Member States reported that their interest payment WHT refunds were substantially lower than their dividend income WHT refunds. 

In many instances, the current WHT procedures in place in the EU have proven to be inefficient and prone to tax abuse.

2.1.1Inefficient: lengthy and costly cross-border WHT procedures

Current procedures are cumbersome and costly for investors, financial intermediaries and tax administrations due to a variety of factors, which are further explained in the section on “drivers”.

In particular, there are highly divergent WHT procedures applied in each Member State with more than 450 different forms to be filled in by investors according to information provided by the industry ( XXV ). In addition, most of these forms are only available in national languages. This results in high administrative and financial compliance costs and in long and heavy processes (e.g. to have the form translated) for both taxpayers and financial intermediaries in charge of providing reclaim services to their clients.

In addition, this may also result in high compliance costs for source tax administrations that have to handle various tax residence certificates (‘TRC’) -in format and content- and for resident tax administration of its investors (resident in that Member State) to assist them in trying to claim their refunds from another Member State´s tax administration.

In particular, the lack of a standardised and digitalised system prevents tax administrations from creating cost efficiencies and automating certain processes. Nowadays a majority of Member States is still fully or partially relying on paper-based WHT refund procedures. This issue was confirmed by most tax administrations during targeted consultations.

According to the public consultation, 89% of all valid respondents (1495 of 1679) strongly agree that the current functioning of WHT refund procedures in Member States hinders cross-border investment in the EU securities market. The following problems with existing WHT refund procedures were identified as most relevant: lack of digitalisation and user-friendly forms (more than 88% of respondents -1441-), lengthy WHT refund procedures (more than 89% of replies -1451-) and costly WHT refund procedures in monetary terms (more than 80% of the replies -1283-). For example, in one Member State, a certificate issued by financial intermediaries, which costs EUR 20 to EUR 50 depending on the bank, is needed for each dividend payment; in another Member State an ownership certificate from the custodian bank is required (around EUR 70-100). This makes many small refunds unprofitable ( XXVI ).

Retail investors are particularly impacted. In a recent survey( XXVII ), close to 70% of retail investors who would be eligible to a reduced WHT rate did not claim it, citing as the main reason: too complicated procedures (33%), lengthy / cumbersome procedures (26%), lack of knowledge (21%), too costly (17%). As regards retail investors who reclaimed the withholding tax by themselves, 77% found the process “difficult” to “very difficult”, and in fact nearly half of them (46%) failed to recover the excess withholding tax (34% of them still failed when reclaiming WHT through a bank). As a result, 31% of the retail investors surveyed decided to sell their foreign EU stocks. This clearly goes against the objective of the CMU. 

The above-mentioned issues may result in unnecessarily long delays before refund requests are processed and paid out. The time to obtain a refund may vary from a few weeks in some Member States to many years in others; however, according to the public consultation replies, for 56% (31 of 55) of companies/business organisations and business associations, the period of time to receive a refund was longer than 2 years. This leads to a lengthy provisional excess taxation or in some cases a permanent double taxation. As evidenced above, this discourages cross-border investments and leads to a situation where non-resident investors give up on their right to claim for the tax refund they are entitled to, due to the complexity of the procedures and the high compliance costs.

In January 2016, the overall cost of WHT refund procedures within the EU was estimated at EUR 8.4 billion per year ( XXVIII ) by the Joint Research Centre (JRC). The estimate has been updated in 2022 amounting to EUR 6.62 billion per year ( XXIX ). Further details on the WHT cost estimations are detailed in chapter 5 (baseline scenario).

2.1.2Risk of tax fraud, evasion and avoidance

Some Member States have experienced large-scale tax abuse schemes known as ‘Cum/Ex’ and ‘Cum/Cum’.

‘Cum/Ex’ schemes work as fraudulent multiple reclaim schemes when entitled to a single reclaim. Short selling practice around the distribution day creates a confusion between the economic and legal owner of the securities, which enables both parties to claim tax refunds that exceed the amount that was initially withheld by the WHT agent. Lack of transparency and uneven digitalisation gave rise to the issuance of refund certificates without due control of the traceability of the payment by tax administrations. An estimation made by the journalist network Correctiv, working together with the University of Mannheim, estimated the losses from Cum/Ex at EUR 9.1 billion for the years 2000-2020.

‘Cum/Cum’ schemes are dividend arbitrage patterns aimed to minimise or avoid paying taxes on dividends. Those strategies are often structured in a way that an investor lends or sells its shares to a borrower/buyer domiciled in a country that has a lower dividend tax rate, so as to minimise the taxes paid on such dividend. The borrower/buyer receives the dividend paid out by the issuer of the share and then returns it to the lender/seller, minus the dividend tax and a percentage – or ‘cut’ – negotiated between the two counterparties.

The cost for Member States of Cum/Cum trading has, according to the same sources of Correctiv and University of Mannheim, been estimated at EUR 141 billion for the years 2000-2020( XXX ).

2.2What are the problem drivers?

In order to accurately understand the current situation and actual problems involved in WHT procedures, we have to look at it in a broader context, in particular, the functioning of financial markets ( XXXI ).

In a cross-border context, chains of financial intermediaries are generally longer than in a domestic scenario, which makes the process of transferring the information relevant for WHT or other purposes such as, for example, the exercise of ownership rights, more complex and costly ( XXXII ).

Long financial chains make it difficult for issuers to know the identity of the ultimate investor ( XXXIII ) – while they know instead the financial intermediary’s identity. Similarly, tax administrations struggle to gather sufficient information on the beneficial owner/ultimate investor to be able to properly apply the benefits granted under the DTTs.

The current initiative attempts to introduce transparency in the financial chain by providing tax authorities with sufficient information in time to ascertain the entitlements granted under the tax treaty and avoid as a consequence a prolonged or permanent double taxation.

The main drivers behind the core problems of inefficiency and tax abuse are the following.

2.2.1Divergent WHT procedures across Member States

Feedback given by Member States ( XXXIV ) demonstrates that WHT collection and relief/refund procedures vary considerably among them. Besides what has been mentioned above regarding different forms and procedures in place, the main differences in such procedures for cross-border dividends and interests are:

1.Reporting obligations associated with WHT collection ( XXXV ): in several Member States, domestic reporting obligations must be met in connection with taxable income payments. Differences exist in terms of (i) the content of reporting, (ii) the entity required to comply with the reporting obligations, (iii) the person to whom the reports must be issued to, and (iv) the frequency and format of reporting. In most cases, the information to be reported relates to the gross income, the tax withheld, the net income, and the recipient of the income. Although DAC2 ( XXXVI ) has streamlined the procedures to report information for cross-border passive income, it does not encompass all reporting items needed under WHT procedures.

2.Documentation requirements for obtaining tax treaty benefits (reduced WHT rates): generally beneficial owners must provide evidence to prove that they are entitled to WHT relief/refund under domestic law or DTTs. The type of documentation to be provided differs from one Member State to another. At one end of the spectrum, benefits can be provided on the basis of free-format information about the beneficial owners or on the basis of documentation held by the intermediary under Know Your Customer rules (KYC) ( XXXVII ). At the other end of the spectrum, reduced rate can be granted only on the basis of official forms or certificates stamped by the local authorities of the investor’s country of residence or by the local tax authorities of the country of investment.

3.Format and period of validity of tax residence certificates: tax residence certificates are issued with different content and formats across Member States. In addition, in some countries, new forms must be used for each claim, while in others, certificates remain valid for one year or until they are revoked. 

4.Time limitations to refund claims: the period within which WHT refunds must be claimed varies among Member States and even within the same Member State, depending on the treaty under which the refund is claimed.

5.Tax authority arrangements for processing refund claims: in some countries, the processing of all refund claims is centralised in one office of the tax authorities, while in other Member States, refund claims must be filed with the local tax office responsible for the WHT agent.

6.Due diligence procedures: Although financial intermediaries would already need to check customer information under AML/KYC and DAC2 rules, current reclaim procedures do not adequately leverage these rules as there are divergent customer due diligence procedures in place in the EU for WHT procedure purposes.

2.2.2Lack and /or uneven state of digitalisation of WHT procedures

The lack of digitalisation and/or uneven state of digitalisation of WHT procedures across Member States causes complexity and opacity for all WHT-related actors (tax authorities, investors, intermediaries, etc.). Indeed, the use of paper-based procedures considerably slows down the reimbursement process, making it burdensome and prone to fraud.

The survey of the Advisory Group on Market Infrastructures for Securities and Collateral (AMI-SeCo) ( XXXVIII ) highlights that barriers to digital WHT procedures exist mainly due to requirements for physical paper documents, paper-based processes, and wet ink signatures. National tax authorities commonly require paper documents and original wet ink signatures. Although solutions for digital alternatives exist and are allowed by the legislator, they are not always accepted by the tax authorities, they are complementary to physical documents, or they can only be submitted for a limited range of procedures.

2.2.3Insufficient information for tax administrations to properly apply tax treaty benefits to entitled persons

Based on targeted consultations of tax administrations and of relevant stakeholders ( XXXIX ), the problems identified (i.e. inefficiencies and fraud) are largely linked to:

·Difficulties to identify the beneficial owner of the payment (the one that is eligible for the benefits granted under a DTT or specific national rules) by the WHT agent/securities issuer. The lack of transparency in the financial chain is making it difficult for the company in which a non-resident taxpayer invests to identify its actual shareholders. As a consequence, the company of the investment/WHT agent is unable to provide to its tax administration accurate information on the identification of the final investor on behalf of who the tax has been withheld.

Relevant information regarding the investor’s identity, tax residence, etc. is collected and verified when the account is being opened. Since the client’s identity can have significant commercial sensitivity, intermediaries often do not share this information with competitors (i.e. other intermediaries) who also are involved in the payment chain.

·Cross-checking that the amount of tax withheld was effectively paid to the tax administration on behalf of that beneficial owner. One of the main hurdles encountered by tax administrations when processing a refund claim from a non-resident investor is verifying that the tax withheld (up to 4 years ago) was done on behalf of the refund claimant due to identification difficulties. When the security issuer levies a WHT over the gross dividend, a tax reporting form is filed to its residence tax administration. The problem is that the person identified in that reporting form is the intermediary, not the ultimate investor. For this reason, the corresponding tax administration has to request information to all intermediaries in the financial chain to ensure the transactional traceability of the operation.

2.2.4No clear/harmonised liability allocation for intermediaries

There are different actors in charge of withholding the tax at source, which is relevant for the allocation of the liability to pay the tax in case of underreporting (less tax withheld than required). In a large number of Member States, the responsibility for withholding the tax lies exclusively with the issuer of the securities. In other Member States, this is the responsibility of the intermediaries involved in the payment of the income, whereby a distinction can be made between Member States allowing foreign intermediaries to act as WHT tax agents and those where only locally established intermediaries can assume such responsibilities.

Such fragmentation in the rules regarding the allocation of liabilities leads to uncertainty for the WHT agents and other financial actors who may refrain from acting in the interest of the investors under the threat of bearing responsibility for any underreporting of tax and, ultimately, being held liable to pay the outstanding tax debt.

Consequences

A direct consequence of the current inefficient and costly WHT procedures is that portfolio investors are struggling to benefit from the tax treaty benefits to which they are normally entitled, which leads at least to temporary double taxation.

The risk of double taxation and high costs involved in WHT procedures discourages cross-border investments in the European market (in fact, respondents to the public consultation who do not invest outside their home country indicated as main reasons wanting to avoid double taxation and inefficient WHT procedures). It is thus a barrier to the establishment of a well-functioning capital market union, and results in an uncompetitive EU market.

According to the public consultation, the consequence to which citizens give the highest importance in terms of the problems encountered is the permanent double taxation suffered (91.29 % -1425-) followed by giving up the right of submitting WHT refund claims (74.76 % -1164-). As regards companies/business organisations and business associations, the two consequences they ranked most often as of ‘high importance’ are the delays in effectively receiving the WHT refund (cited by 90% of companies/business organizations and associations), and the high compliance costs associated with the WHT refund process (cited by 84% of companies/business organizations and associations).

Another consequence of inefficient WHT procedures in the EU is the loss of tax revenues when tax fraud is perpetrated in the absence of a proper monitoring of the refund and relief procedures. Public revenues are also indirectly affected in the form of a lower taxable base when less investment is attracted into the EU ( XL ).

The dissatisfaction among retail investors about the functioning of existing tax rules is evidenced by the numerous responses and high proportion of individual investors to the public consultation (in total 1682), which is unusual in direct taxation matters: EU citizens accounted for more than 90 % of responses (1547 out of 1682).

Problem tree

The problem tree comprises the above-mentioned problems and consequences that this initiative aims at tackling. The figure below also presents the main drivers and provides for additional context elements that have been developed:

Figure 2: Problem tree

Source: DG TAXUD

Affected stakeholders

The problems and drivers stated above and their consequences affect investors, financial intermediaries (custodians, banks, collective investment vehicles and other financial players), tax administrations and European companies as ultimate recipients of the capital potentially invested by non-resident investors.

Both retail and professional non-resident investors are directly affected by costly, lengthy and burdensome WHT procedures. In particular, non-resident portfolio investors ( XLI ) struggle with recovering the excessive tax withheld. Individual investors seem to be particularly impacted, since 51% (651) of EU citizens responding to the public consultation were not able to get a refund on the tax withheld, whereas 76% (41) of companies/business organisations and business associations responding to that public consultation were able to receive a refund in some cases.

Member States, and more specifically their tax administrations, are also affected by these complex procedures as mentioned before: firstly, because of the revenue loss they suffer due to fraud perpetrated and to less inward investment; secondly, because they have to devote resources and time to verify the refund submissions, often without proper investigative tools.

Other affected parties are the financial intermediaries involved in the procedures. As they provide services to their clients, they might have the right to ask for the refund on their client’s behalf or are even entitled to the reduced WHT rates themselves as investors. Therefore, financial intermediaries face the same hurdles to recover the tax withheld and may cease to offer the WHT reclaim services to their clients given the massive compliance costs, the large range of different procedures involved and the potential liabilities associated with offering this service to their clients.

Ultimately, European companies are less attractive to non-resident investors, which potentially hampers their capacity to invest and ultimately their growth prospects ( XLII ). They cannot exploit fully the advantages of the Single Market by tapping into the retail investor base in other Member States, thus segmenting unnecessarily the shareholder base over national boundaries.

2.3How likely is the problem to persist?

Some Member States have put unilateral measures in place, including developing and introducing digital tools, to address the problem through either a relief at source, refund system, or a combination of both ( XLIII ).

Without any EU action in the WHT field, the fragmentation of regulations and practices across Member States will grow and, therefore, greater inconsistency will continue in the EU with increasing negative consequences for companies, investors and financial intermediaries.

Indeed, in case of divergent measures still proliferating to address existing hurdles regarding WHT reclaim systems, the problem of compliance and administrative costs is likely to remain. For example, for a financial intermediary with clients operating in various Member States, the operational and compliance costs associated with diverse reclaim systems would be higher than having to face one standard refund procedure, or several procedures with common forms and/or requirements to fulfil in order to request the money on behalf of its clients. Chapter 5 gives more information on the impact of the baseline scenario (assuming there is no EU intervention).

In the absence of EU intervention, tax fraud related to WHT will continue to exist, especially in Member States that have not introduced unilateral measures to address the flaws in the monitoring of WHT reclaim systems.

Several actions have been taken by the EU over recent years to tackle tax abuse in general, including the Anti-Tax Avoidance Directive (ATAD), the Council Directive (EU) 2018/822 amending the Directive on Administrative Cooperation in the field of (direct) taxation (DAC6) and the recently proposed Directive laying down rules to prevent the misuse of shell entities for tax purposes (UNSHELL). Nevertheless, none of those initiatives address the specific problem of double refund abuse.

3Why should the EU act?

3.1Legal basis

Article 115 of the Treaty on the Functioning of the European Union (TFEU) constitutes the legal base for legislative initiatives in the field of taxation. Although no explicit reference to direct taxation is made, Article 115 refers to issuing directives for the approximation of national laws as those that directly affect the establishment or functioning of the internal market. It follows that, under Article 115 TFEU, directives are the appropriate legal instrument for the Union in this field. Based on Article 288 TFEU, directives shall be binding as to the result to be achieved upon Member States but leave the choice of form and methods to the national authorities.

This initiative aims to ensure the smooth functioning of the Internal Market by improving WHT procedures in the field of direct taxation, making them more effective and efficient and addressing existing tax abuse.

3.2Subsidiarity: Necessity of EU action

As shown in the targeted consultation of Member States, some of them have introduced provisions attempting, in different ways and by different means, to address the problems of WHT procedures. For instance, Finland has introduced in 2021 a TRACE-like system ( XLIV ) for dividend payments from listed companies. It relies on a relief at source scheme as primary system. Other countries like Germany have approved a tax reform on WHT procedures for non-residents that relies on a standard refund procedure with several extensive conditions to be fulfilled in terms of documentation and other formal requirements.

Keeping an increasingly fragmented framework of WHT procedures across the EU would lead to higher compliance costs for all actors such as financial intermediaries, investors and Member States tax authorities. The prevailing cross-border nature of the problem at stake determines the necessity of EU action in order to simplify administrative procedures and reduce compliance costs. Otherwise, fragmentation of national rules on WHT procedures across Member States would make the effective functioning of refund or relief procedures more difficult to achieve for cross-border operations, hindering the proper functioning of the internal market. Therefore, an EU action is required to level the playing field for national and foreign investors and for domestic and external intermediaries alike.

As mentioned in section 2.1, a number of Member States have domestic WHT rates that do not exceed the rates established in their DTTs ( XLV ). While those Member States do not need to engage in WHT refund procedures, their resident taxpayers investing or wishing to invest abroad are nevertheless exposed to the described problems and may refrain from doing so because of burdensome withholding tax procedures.

Furthermore, based on their current national legislations, some Member States may not be in a position to trace full details / information on passive income payments from cross-border investment throughout the financial chain. This is typically relevant for the taxation of the operations undertaken by institutional investors or with several layers of intermediaries. To address this, a common framework providing a minimum standardised procedure for WHT, introducing transparency within the financial chain, is needed at EU level.

Despite the non-binding actions already undertaken at national, international and European level, the issue persists. In particular, the most recent soft law measure implemented at Commission level was the approval of a Code of Conduct in 2017. Although the implementation of the Code of Conduct helped in raising awareness on the need to simplify WHT procedures for compliant non-resident beneficiaries (the investors), it was not in itself sufficient to fully tackle the existing problems (see the biannual summary report ( XLVI )). Therefore, further EU action is needed at legislative level, as neither national measures nor multilateral soft measures have been effective so far. 

Furthermore, the initiative aims to respond to the recommendations provided by ESMA in the above mentioned ‘Final report On Cum/Ex, Cum/Cum and withholding tax reclaim schemes’ which concluded that specific action at EU level in the field of taxation should be provided.

3.3Subsidiarity: Added value of EU action

Given that there is a need to act in response to the problems set out in chapter 2, it is preferable to avoid a patchwork of refunding requirements, unilaterally implemented by some or all Member States using different procedures.

Solutions could range from improving existing systems to setting up an EU-wide standardised reclaim regime. The solution sought must strike a balance between making WHT procedures more efficient, less burdensome, and less costly for investors, financial intermediaries and tax administrations, while at the same time improving the prevention of tax abuse (Cum/Ex, treaty abuse, etc.) to ensure that everyone pays their fair share of taxes.

A solution at EU level that would digitalise and harmonise key features of WHT procedures while respecting the principle of proportionality, thus avoiding excessive negative consequences, would be expected to result in a lower administrative burden and hence time and costs savings for tax administrations, investors, financial intermediaries and companies.

In light of the above, making WHT procedures more efficient and fighting against fraud are objectives that can be met more effectively and efficiently at EU level.

The added value of EU action is broadly confirmed in the public consultation where the vast majority of respondents (close to 94 %) stated that the same WHT procedures system should be set up throughout the EU. Only 0.4 % of replies find low added value in an EU wide harmonised framework.

4Objectives: What is to be achieved?

Figure 3: Objective chart

Source: DG TAXUD

4.1General objectives

The general objectives of this initiative are to achieve fair taxation – by ensuring that taxpayers do not pay more than they should and that taxes which are due are paid - and to reinforce the capital market union. 

1)Reinforce the capital market union

To ensure free movement of capital, this initiative aims to facilitate cross-border investment. The initiative seeks to make the EU market more attractive to investors by giving taxpayers proper access to DTT benefits, hence creating a competitive Single Market across Europe. The initiative aims to fully exploit larger-scale markets as nowadays EU financial actors are disadvantaged compared to their global peers. The scale of Europe’s capital markets does not match the size of its economy, and there is ample room to increase the euro’s role as an international currency.

Taxation can present a serious obstacle to cross-border investment and thus to creating a genuine CMU. Alleviating the tax associated burden in cross-border investment does not necessarily require harmonisation of tax codes or rates. It can be achieved by providing Member States with the proper tools to effectively monitor the tax withheld quickly and being able to refund the excess withheld when appropriate.

A fully functioning and integrated market for capital with no barriers will support the EU’s economy to grow in a sustainable way and be more competitive. An economically stronger Europe will better serve its citizens and help the EU play a stronger role on the global stage.

2)Ensure fair taxation 

Ensuring fair taxation also requires preventing tax abuse. Providing Member States´ tax administrations with the proper tools to address the current lack of transparency would make circumventing WHT rules more difficult. Therefore, this initiative aims to contribute to safeguarding the tax revenues of Member States, to making tax systems fairer and to ensuring a level playing field among all involved.

Fighting tax fraud and evasion to help secure national and EU revenues is one of the Commission’s priorities, whose political guidelines call for stepping up the fight against tax evasion to make taxation fairer ( XLVII ). Such an initiative would also strengthen citizens’ trust in the fairness of the tax system, as it would address the public concern that income obtained by investing in financial markets will be effectively taxed once in the country of investment.

4.2Specific objectives

Making WHT reclaim procedures for non-resident investors more efficient is one of the specific goals of the current initiative. More specifically, the initiative aims to:

·Shorten the average time for refunding excess tax withheld in the EU.

·Simplify withholding tax procedures.

·Promote digitalisation and harmonised procedures.

·Ensure equal access to treaty benefits for portfolio investors as direct investors have.

Secondly, the initiative is meant to improve the ability of Member States to detect and counter cross-border tax fraud, avoidance, and evasion. In order to enhance such ability, new tools should be provided to Member States in terms of higher levels of digitalisation and transparency. This ability is dependent on the quality, relevance, and timeliness of information to which Member States have access to and on how the tax administrations deal with such information. If tax officers have the information they need at the right time, it will be possible for them to verify that the tax has been properly withheld and perform risk management so as to ensure that their resources are allocated to high-risk cases. By increasing the probability of detecting non-compliance, the initiative aims to have a deterrent effect, discouraging fraudulent actions and incentivising taxpayers to declare and pay taxes owed.

Both objectives that will be addressed by the initiative - improving efficiency and fighting tax abuse – are of equal importance. Current refund procedures are found to be inefficient, which is costly and detrimental to all stakeholders (investors, financial intermediaries, and tax administrations) and weighs on the overall investment climate in the EU. At the same time, WHT refund procedures have proven vulnerable to fraud, which led to severe tax revenue losses in some Member States. Therefore, to cater to the needs of Member States and their tax administrations as well as of investors and financial intermediaries, both objectives need to be met.

5What are the available policy options?

Designing an effective and efficient solution to the depicted problems is challenging. At the highest level of analysis, a choice is to be made between the status quo or baseline, that represents a non-regulatory setting, and a scenario where the Commission would act in a legislative manner. Under a regulatory framework, there are several options for improving WHT procedures in the EU. This chapter describes the options identified.

5.1What is the baseline from which options are assessed?

Under the baseline scenario, the EU would not act. Tax administrations from Member States would retain their existing procedures, which vary in terms of scope and administrative requirements ( XLVIII ). As explained under the problem definition section, the majority of Member States levying WHT at source over dividend and interest payments to non-resident investors have in place at least the standard refund procedure in combination of one or two of the other WHT systems (relief at source and/or quick refund). Member States are likely to continue applying a combination of different WHT systems.

Furthermore, a tendency can be observed that some Member States try to address the problem of fraud and abuse in different ways. This increases the fragmentation of the WHT procedural framework among Member States and further aggravates the existing problem making EU action to standardize the procedures thus more urgent.

As mentioned in chapter 2, under this “no-policy-change” scenario, investors would continue to incur costs amounting to EUR 6.62 billion per year according to an updated estimate performed by the JRC in 2022 ( XLIX ). The total cost estimation is composed of three elements:

1)Foregone tax relief: Due to the complexity and cost of WHT procedures, some investors do not claim their tax refunds. This component is assumed to represent 10% of the total amount of cross-border investment.

2)Opportunity costs: Due to delayed payments of tax refunds, the investor incurs opportunity costs as the money cannot be invested elsewhere and hence generate income. The opportunity costs have been calculated assuming an average delay of one year and a risk-free rate of 0.6% ( L ).

3)Incurred costs: The incurred costs relate to the costs effectively paid under a reclaim procedure, such as the costs associated with the paperwork and different source country requirements, or the costs incurred by custodians or external service providers. It is assumed to account for an average of 2% of the refundable amount.

The estimation of the costs has been assessed in previous JRC studies undertaken in 2009, 2016, and 2021 (see Annex 4). The decrease in the estimated costs for 2021 (see table 1 below) can be attributed to the following factors: decrease in the risk-free interest rate, decrease in the return rate of investments, and a decrease of the differences between national and tax treaty rates of some Member States. Driven by the current economic context and recent evolution of interest rates, the additional calculation of 2022 was recently performed by the JRC using long-run yields (period 2000-2019) and is taken as dynamic baseline for comparison purposes. A summary of the estimates arising from JRC’s subsequent studies is shown in the table 1:

Table 1: Overview estimated costs (EUR bn)

Source: DG TAXUD based on JRC studies

The costs presented in the above table reflect the total costs borne by investors in case of investments in EU companies; with 52% of these costs borne by EU investors.

In addition to the estimated costs for investors, the effects on GDP (and other elements) have also been estimated in case WHT were removed using a Cobb-Douglas function. The elimination of WHT is used as an approximation of fully functioning WHT tax procedures. The table below indicates, based on the CPIS Survey on cross-border portfolio security investments in the EU, the impact that eliminating WHT costs would have. Therefore, it reflects the increasing investments towards the EU that would take place in case of EU action (and conversely gives an indication of the investments that would be missed out in the absence of EU action):

Table 2: Impact of eliminating WHT costs

Source: JRC 2021/2022 study on WHT costs

For robustness purposes, a second simulation was performed by JRC in 2021 using the Commission’s economic model CORTAX which analysed the tax-benefit policies of Member States to assess the effects of removing tax overpayments (defined as non-refundable WHT) on GDP, investment, wages, employment and tax revenues. The net tax revenue figure shows that the increase in tax revenue from the positive change in economic indicators is outweighed by the decrease in WHT revenues due to more efficient reclaim procedures. This leads therefore to a net tax revenue loss. The table below reflects the outcome of this simulation:

Table 3: Macro economic Impact

Source: DG TAXUD based on JRC studies

Therefore, as mentioned above, in case of non-binding EU action, the EU would miss a boost in investments towards the EU that would have resulted in an increase of 0.047% of the EU27 GDP; 0.137% in capital investment and 0,048% in wages as 2nd round of macro-economic effects estimated by CORTAX Commission’s economic model.

As regards fraud, in the absence of EU action, Cum/Ex fraud could still happen - the losses from Cum/Ex were estimated at EUR 9.1 billion for the years 2000-2020. Similarly, safeguards are needed to address Cum/Cum practices, which led to tax revenues losses of EUR 141 billion for the years 2000-2020( LII ).

5.2Description of the policy options

All the options presented below imply a legally binding framework. Options 2 and 3 build on option 1. Prior to entering into a detailed explanation of each option, the following summary of the options gives an overview of the elements further explained in chapter 5.

Three options are envisaged to tackle inefficient and prone to fraud WHT refund procedures. The check-chart below shows the elements included in each of the three policy options:

Table 4: Overview of options and elements

Source: DG TAXUD

All these elements will be further explained under each of the following sections.

5.2.1Option 1 — Common EU digital tax residence certificate (eTRC) + common reporting

This option encompasses the eTRC, common reporting requirements for financial intermediaries within the financial chain vis-à-vis the tax administration of the source Member State, standardised due diligence procedures, common rules for liability allocation and the option for financial intermediaries to file requests on behalf of clients/taxpayers in one digital form with tax authorities. These elements would be implemented by Member States to improve and amend their existing refund or relief procedures.

The features of Option 1 are the following:

1.Introduction of Digital Tax Residence Certificate: eTRC

Each Member State would have the obligation to put in place a system allowing the digital issuance of tax residence certificates using a common EU format in terms of minimum content and a common IT technology. All Member States would have the obligation to recognise these eTRCs as a valid proof of residency. Such certificate would be issued by the country of residence of the investor and would enable the tax administration of the source Member State to examine whether the investor is eligible for a reduced WHT rate.

Setting up an eTRC would not involve a harmonisation of the criteria to determine whether a specific investor is resident for tax purposes in a country ( LIII ). Every Member State would determine the residence according to its own domestic rules. The consistency across Member States would come from a standardisation of the format and of the IT solution. In particular, the eTRC would have the following features:

Consistency in format

1. Common minimum content: identification of the taxpayer, country of residence, date of issuance and validity period.

2. Standardised validity period of one year whereby the eTRC could be used multiple times to reclaim refunds and be potentially used for other purposes insofar as permitted by the national legal framework.

Consistency in technology

1. Available to be requested online.

2. Tax administrations’ capability to issue it electronically in real-time.

3. Common verification method to ensure the reliability of the source and the authenticity of the document across Member States. This common verification method could be achieved based on four potential IT solutions: (i) verifiable credentials (relying on a verification model based on machine-readable, cryptographically verifiable certificates that allow the users to keep the eTRC in a digital wallet), (ii) electronic seals (providing trust regarding the authenticity of the issuer, as well as the integrity of the certificate via electronic signature), (iii) data publication (relying on the issuer pro-actively publishing certificates to all verification service providers, either directly or via a central database) and (iv) data querying (relying on the issuer to provide web services for querying certificates, which can be consumed on demand by verification service providers) – see Annex 12 for more details.

2.Reporting obligations ( LIV ) 

The proposed reporting obligations would set a common standard with regard to the kind of information to be shared with the tax administration and would also create a common standard on how it would be shared.

Reporting by financial intermediaries is required to combat tax abuse in the form of Cum/Ex and Cum/Cum as explained in Chapter 2. For Cum/Ex, the WHT agent or the tax authority would have information available on the dividend to ensure that (i) the WHT has been levied in the first place; and (ii) that only one refund can be requested for the WHT levied in (i). The lack of transparency of information in the financial chain has resulted in the Cum/Ex abuse. For Cum/Cum, the information reported by financial intermediaries, such as the acquisition date of the security, can be used by the WHT agent or the tax authority for risk assessment purposes to prevent the tax avoidance practices related to Cum/Cum.

·Who would share the data? The data would be reported directly, depending on the option chosen by the Member States, to the WHT agent or the tax authority in source Member State by all EU-based financial intermediaries playing a role in the dividend/interest payment chain, in a comprehensive and standardised way. Non-EU-based financial intermediaries would have to report in a similar way in case they are enrolled to a dedicated Register (explained below).

·Which data would be reported? The reported data would equip tax administrations with the information they need to assess whether a taxpayer is entitled to a reduced WHT rate or not. It would notably include identification information about the financial intermediary and the final investors (Taxpayer Identification Number, name, address, etc.) and dividend/interest payment information (issuer, ISIN number, date of payment, WHT rate applied, net dividend/interests received and the WHT amount levied).

·How would the information be shared? The reporting is to be done in a digital format, thus allowing both the provider(s) of the data (financial intermediaries) and the recipient(s) (tax administrations) to compile and process the data in an automated and standardized manner.

·When would the data be reported? All financial intermediaries established in the EU and non-EU financial intermediaries registered would be obliged to report to the relevant tax administration without delay after the dividend/interest payment has taken place. After consulting the relevant stakeholders, including tax administrations, reporting the information within 25 calendar days of the payment date would be a realistic deadline.

Thanks to this reporting, tax administrations would obtain the information they need to apply their own risk management and process refund requests promptly.

Harmonising the data to be reported would result in heavier reporting obligations for financial intermediaries compared to the current situation, both in terms of the number of financial intermediaries that would have to report to the tax authorities, and in terms of the amount of information that would have to be reported (see section 6.1 for more details). Yet as a few Member States are considering introducing additional reporting requirements, a harmonised set of reporting requirements, the digitalisation of the processes and the harmonisation of the format would be expected to avoid a situation where financial intermediaries would face different information requirements established by national law and a variety of forms which might lead to an uneven playing field and an even heavier administrative burden.

3.Due diligence

The financial intermediary closest to an investor that is seeking withholding tax relief would need to perform a set of checks regarding its client, i.e. due diligence procedures. These procedures should build on the due diligence procedures required under DAC2 and on the know-your-customer (KYC) rules that are applicable to financial institutions under the EU Anti-Money Laundering Directive ( LV ). Under these rules, financial institutions are already required to conduct customer due diligence in connection with the opening of accounts and certain specified transactions on an on-going basis. The rules generally require the financial institution to identify the customer and to verify the customer’s identity using reliable and independent documents, data or information.

Such due diligence procedure would be complemented by two actions geared towards ensuring that the relevant investor is entitled to the reduced tax rate:

·checking the authenticity of the content and the source of the eTRC via the common verification method agreed and

·checking the correct applicable WHT rate according to the DTT or domestic rules.

4.Liability allocation 

Common rules would be introduced to define when financial intermediaries would be held liable for the provision of incorrect data, which would lead to loss of tax revenue for the Member State. The liability would be placed at the level of the financial intermediary closest to the investor, who is responsible for performing the due diligence requirements. The intermediary would be liable in case of mis- or underreporting, except if the intermediary proves it has taken reasonable measures to check the taxpayer's entitlement to reduced WHT rates.

5.Multiple requests on a bulk basis

A common standard for financial intermediaries would be set to request a refund on behalf of their clients, so-called ‘bulk’ requests ( LVI ). It implies an agreement on a common refund form or IT scheme for the request to be done in an automated way.

6.Scope

The implementation of the eTRC would cover all Member States. For the reporting obligations and other elements included in this option, the scope covers those Member States where non-resident investors may face double taxation when receiving dividend/interest payments because the Member State of the investment has a domestic WHT rate for cross-border dividend/interest payments which is higher than the WHT rate stated in their DTTs ( LVII ). It will be applied in relation to dividend or interest payments from EU listed companies to portfolio non-resident investors wherever they are located.

7.Non-EU dimension

The eTRC would be potentially scalable to those third countries that are willing to commit on similar level of digitalisation, content and verification method (this would benefit non-EU investors). On the other elements under option 1, non-EU financial intermediaries would be allowed to join the system, but this would be on a voluntary basis. In order to do so, they would need to register in a dedicated register. Registered non-EU financial intermediaries would have the same benefits (provide their clients with swifter procedures) and obligations (reporting and liability obligations) as EU-based financial intermediaries. The authorization process for registration and monitoring of compliance will be under the operational control of the relevant Member State.

This option would imply that Member States will be able to make their existing refund/relief procedures more effective and fraud-proof. Therefore, under this option the existing refund and relief procedures currently in place in every Member State will gain directly in efficiency and security.

5.2.2Option 2 - Relief at source system

This option builds on the elements introduced in option 1. In addition, option 2 would implement at EU level a relief at source system, which is characterized by applying the reduced WHT rate directly at the moment of the interest or dividend payment ( LVIII ). Nevertheless, a standard refund system would remain in place as a back-up option (see section 2.1).

It entails a standardised procedure for claiming withholding tax relief at source on portfolio investments. A relief at source system allows the withholding tax agent to apply reduced rates on WHT pursuant to DTTs or domestic rules on behalf of their customers (portfolio investors) at the moment of dividend/interest distribution.

The relief at source system envisaged by the EU initiative would build on option 1 and would also specify the applicable procedure to allow for the relief. Specifically, it would have the following features:

1.eTRC, common due diligence procedure and fair liability allocation 

All these elements remain the same to what is described under option 1.

2.Reporting obligations

The reporting obligations defined under option 1 would be leveraged under a relief at source system with the only difference that the information would be sent to the WHT agent in charge of paying the dividend instead of the source tax administration. The information would be reported, in a subsequent phase, from the WHT agent to the source tax administration.

3.Procedure

This option provides for a system where the correct amount of WHT is applied at the moment of paying the dividend to the investor. This task is outsourced to a WHT agent or financial intermediary.

4.Scope

This system would apply to payments to non-resident portfolio investors from listed companies located in any of the Member States covered by the reporting obligations under Option 1.

5.Non-EU dimension

In terms of eTRC and reporting, the same approach as under Option 1 would apply.

Under this option, Member States with no relief at source would need to introduce it and those Member States with a relief at source may need to extend the scope and make the necessary amendments in the reporting obligations in order to standardize their systems to the relief at source regulated in the proposal.

This option is the preferred one among investors and financial intermediaries as shown in the public consultation and target meetings with the industry. This option was also identified by the Commission in the Action Plan for fair and simple taxation supporting the recovery strategy and the Action Plan for a Capital Market Union ( LIX ). 

Therefore, even when there may be Member States reluctant to implement it in the short-term given the recent Cum/Ex scandals, setting up a relief at source system is clearly considered as a goal to be achieved in the long-term from the perspective of truly attaining a CMU.

5.2.3Option 3 –Quick refund system ('QRS') within a set time frame or/and Relief at source system (‘RAS’)

Option 3 fully encompasses option 1 as described above with the added requirement that Member States applying a refund system should ensure that the reclaim is handled within a pre-defined timeframe ( LX ), so-called the Quick refund system. Additionally, under this option, Member States can introduce or continue to implement a relief at source system in order not to restrict Member States to relieve double taxation at the moment of payment.

This option ensures that Member States have at least one effective WHT system in place to provide relief, such as QRS. But it shall also allow them to use RAS where they deem it appropriate rather than only providing QRS.

The main features of Option 3 are the following:

1.eTRC, reporting obligations, due diligence procedure, liability allocation and request in bulk basis

These elements remain the same to what is described under option 1.

2.Procedure

Member States will continue to have the choice to either implement a relief at source or a refund procedure within a set timeframe( LXI ). The relief at source would be as stated in option 2. Under QRS, all actors in the process need to report information, perform checks or make payments within set time limits. The only difference between a standard refund procedure and the quick refund procedure is the timing. An overview of the quick refund procedure process is depicted in figure 5 below.

3.Scope

Option 3 covers, similarly to Option 2, dividend and interest income payments from listed companies established in any of the Member States where the statutory rate to non-resident is higher than conventional one and it is paid to portfolio investors ( LXII ).

4.Non-EU dimension

In terms of eTRC and reporting, the same approach as under Option 1 would apply.

Figure 5: Description of the QRS procedure (timeline)

Source: DG TAXUD

(Legend: X+number of days)

The deadlines mentioned in the figure are based on the results of the targeted consultations undertaken mainly with financial intermediaries and tax administrations ( LXIII ) in terms of time needed for each intermediary involved in the financial chain. Existing refund procedures in some Member States may already comply with this pre-defined timeframe ( LXIV ).

Financial intermediaries that do not comply with their obligations, including in terms of timeframe, could be faced with sanctions ( LXV ) such as fines. Tax administrations that would not process the refunds within the set timeframe - without valid reason, such as raising questions to further investigate the request - would need to add interest penalty payments ( LXVI ) to the refund payments. 

An important feature of the QRS is that tax administrations would retain control of the procedure. More specifically, they have the opportunity to check on the basis of their risk management strategy and in an automated way, the requests received before the refund is paid. In case the checks would result in the need for further queries, the tax administration has the option to halt the procedure and investigate further.

It should be noted that under the QRS, the control of the refund process - including the risk assessment- is undertaken by the tax administrations. Under RAS, the ex-ante monitoring task is outsourced to financial intermediaries, while tax administrations would be able to perform checks ex-post to ensure the financial intermediaries perform their task correctly.

5.3Option discarded at an early stage

The option of choosing a non-binding instrument to address hurdles encountered in relation to WHT procedures was discarded. The experience with a soft-law approach (in particular the 2009 Recommendation and the 2017 Code of Conduct) has proven ineffective in the past and would not bring any added value compared to the baseline scenario (see Annex 10 for further explanation on the soft-law measures).

The possibility of introducing a common Quick refund system across Members States as a standalone option has not been considered since it would imply that Member States where relief at source system is already being applied (either as a primary system of for some payments) need to step back and replace this system with a common Quick refund system for every payment. As long as Member States are comfortable to apply the relief at source system they should be allowed to maintain a system that does not produce provisional double taxation. Obliging to establish a Quick refund system for all payments would run against the objective of moving towards a relief at source system (see section 1).

The figure below provides a chart on the intervention logic related to the current initiative where the link between the problem tree, the objectives and the policy options are clearly stated:

Figure 6: Intervention logic

Source: DG TAXUD

6What are the impacts of the policy options?

This section is organised as follows. First, the economic impacts are presented. A summary of the economic impact of the policy options is followed by the effects of the initiative on SME’s. Then, the economic impacts of options 1-3 are considered in terms of: (a) the costs and benefits for investors, (b) the costs and benefits for tax administrations, (c) the costs and benefits for financial intermediaries and (d) the macroeconomic effect for Member States. Finally, there is an assessment of the social and environment impact, and impacts on fundamentals rights, notably on the right to the protection of personal data.

6.1Economic impacts of the policy options

6.1.1Overview of economic impact of options 1-3

Before entering into the detailed assessment of each policy option, hereinafter there is a summary chart of the costs and benefits for the main stakeholders under each option. The table below provides an overview of the benefits and costs for investors, financial intermediaries, tax administrations and Member States.

Table 5. Summary of costs and benefits for stakeholders under Options 1-3

Source: DG TAXUD

As shown above in Table 5, Member States would be expected to incur a net tax revenue loss under Options 1-3. These net amounts reflect two opposite impacts on tax revenues: (i) the loss of tax revenue that Member States would incur due to more efficient WHT procedures resulting in less foregone WHT for investors, and (ii) additional tax revenues due to an increase in GDP, through increased cross-border investment, and increased employment and wages. Member States will collect additional tax revenue as a result of the reporting of information to counter tax abuse, however, it has not been possible to quantify these amounts and it is therefore not reflected in the above table.

Table 6. Summary of net tax revenue for the EU27 under Options 1-3 (EUR billion)

Note: calculations based on 2019 CPIS cross-border securities data.

The amounts related to the benefits for investors are derived from the JRC study for withholding tax costs used for the baseline as outlined in Chapter 5.1. The amounts related to GDP, investment, wages, employment and net tax revenues are derived from the CORTAX model used by JRC and outlined as well in the baseline scenario.

6.1.2Economic impact of the initiative on SME’s

Under the SME test, SMEs are not identified as a relevant affected business by the proposal ( LXVIII ). Any impact of the policy options will not differ between large companies and SMEs if they are acting as portfolio investors and indeed both may benefit from the outcome of the policy options ( LXIX ). As securities issuers, large companies and SMEs that regularly distribute dividends and are traded on the stock markets would be equally affected as well, as they would overall benefit of becoming more attractive investment opportunities with a higher return. However, investment in cross-border securities from listed companies by retail investors is primarily happening in large companies rather than in SMEs.

From the public consultation, no comments received were specific to SMEs as regards their role as an investor, as a financial intermediary or as an investment company.

6.1.3Option 1 Common EU digital tax residence certificate (eTRC) + common reporting

6.1.3.1Setting up a common EU digital TRC

The impact on stakeholders of introducing an eTRC are outlined below, taking into consideration the various technological solutions envisaged for the eTRC (i.e. verifiable credentials; electronic seals; data publication; and data querying).

Costs and benefits for investors

The costs of using an eTRC are expected to be minimal for investors. The digitalisation of the tax residence certificate would mean that a WHT refund request could be submitted digitally facilitating investors’ interactions with national tax administrations and speeding up the overall process of reclaiming WHT. It is expected that, similar to the existing paper-based version, obtaining electronic tax residence certificates would continue to be free for investors. However, investors would also need to have access to a computer and/or mobile device to be able to benefit from the digital certificate, if the IT solution chosen is verifiable credentials and electronic seals as described in Chapter 5.

More specifically, to be able to use verifiable credentials, investors would need to obtain, install, and operate a wallet application (on mobile device, desktop, or even cloud) to hold the eTRC along other credentials and interact with his/her identity provider and with the verifier(s). At worst, for investors as natural persons the price of mobile wallet software is expected to be in line with typical mobile (Android or iOS) applications. Investors as legal persons should use digital wallet software, typically as a subscription to a wallet-as-a-service in the cloud, but it could also be a standalone application deployed on their own infrastructure.

For electronic seals, any electronic storage solution (such as hard drive, USB stick, cloud storage, etc.) would be sufficient.

Costs and benefits for tax administrations

According to a mapping exercise by the Commission services, a number of Member States already provide a TRC in some form of electronic format (e.g. pdf). In terms of being able to submit a WHT refund digitally, 6 Member States reported in a survey undertaken in 2021 that this was the main method used to submit a WHT refund, 2 Member States replied that a digital submission was for a minority of refund claims, while 6 Member States indicated that only non-electronic means of submitting refunds was possible ( LXX ).

Digitalisation of the tax residence certificate would facilitate tax administrations’ interactions with investors and allow for the automation of reliable and secure tax residence certificate validity checks, thus helping to combat fraud.

The costs for tax administrations of the various IT solutions for the eTRC are provided below (and in more details in Annex 12):

Table 7. Summary of annual costs and benefits for stakeholders

Source: DG TAXUD

Therefore, depending on the IT option chosen (through implementing acts), IT development costs for an eTRC for the EU27 are estimated to be between EUR 4.9 million and EUR 54 million.

The estimates for options A and B are based on a similar electronic seal project undertaken by the Commission. The actual costs of implementation, operation and maintenance of the required IT services for options A-D will depend on the specific IT context each tax administration is facing (national IT market, national administration systems and online services already in operation, etc.).

Costs and benefits for financial intermediaries

The digitalisation of the tax residence certificate would allow financial intermediaries to automate their related processes and save potential storage costs of the physical documents. This would speed up the refund process as well as making identification more secure by checking verifiable credentials and thus combatting fraud. Compared to electronic seals (option B) which “only” verify the integrity of the eTRC and the signature by the tax authority, verifiable credentials offer the significant additional benefit that the verification of the identity of the beneficiary (‘holder') is already ‘embedded’, requires no human intervention and is therefore much less error prone.

The verification costs of an eTRC would be expected to be minimal for financial intermediaries. However, if a financial intermediary wants an automatic verification process which allow investors to directly submit an eTRC online (either as verifiable credential or as electronic seal), then the implementation costs are estimated at between EUR 80 000 and EUR 135 000 per financial intermediary for Options A and B (as detailed in Annex 12). The equivalent costs for Options C and D are between EUR 60 000 to EUR 100 000 per financial intermediary. Automatic verification may be an optimal cost/benefit solution for intermediaries like custodian banks with a large number of transactions. This cost is not taken into consideration in the final estimation of costs for financial intermediaries as they are deemed optional.

6.1.3.2Common Reporting

Option 1 entails setting up reporting obligations for financial intermediaries to report WHT data to the tax administrations and/or the WHT agent in the Member State of the securities issuer.

Costs and benefits for investors

No direct additional reporting costs for investors are anticipated as they already have to provide a number detailed information based on the existing EU AML rules KYC and DAC2/CRS which financial intermediaries and withholding tax agents already need to comply with( LXXI ).

Investors would benefit from this option since it should reduce the time to get the refund and lower opportunity costs. Decreasing the foregone tax relief can be considered intermediate in terms of time needed to get the refund for the investors between the current baseline and the relief at source and quick refund options, which decrease significantly or almost completely the time between the launching of the request and the refund. For this reason, the impact of this option has been estimated by relying on the following assumptions: average time of six months to get the refund, a number of people not claiming the refund halved according to the Commission study estimates in Annex 4, and a cost for processing the refund lowered due to digitalisation from 2% to 1% of the refundable amount. Building on the Commission studies, and their assumptions, the total costs incurred by investors are estimated to decrease by 50% under Option 1 compared to the baseline estimation. They amount to EUR 3.31 billion (compared to EUR 6.62 billion in the baseline scenario).

Table 8: Estimated benefits for investors– breakdown cost reduction driver (EUR bn)

Source: DG TAXUD and JRC calculations

Costs and benefits for tax administrations

High benefits are expected for tax authorities compared to the baseline scenario. Thanks to the reporting obligations, tax authorities in the source Member State would be able to use multiple sources of information with which to check eligibility in a relief at source system and to ensure that a WHT refund is correctly granted to the investor. Digital reporting of such information to the tax authority would help Member States to target tax abuse better by ensuring they are able to check underlying securities income and WHT data.

The amounts involved in WHT abuse are significant. An estimation made by the journalist network Correctiv, working together with the University of Mannheim, quantifies the losses from Cum/Ex at EUR 9.1 billion, and from Cum/Cum at EUR 141 billion for the years 2000-2020( LXXII ).

The ESMA report noted a number of jurisdictions that had implemented investigations into WHT abuse ( LXXIII ). In addition to the aforementioned WHT abuse, the ESMA report estimated that between 2012 and 2015, Denmark was subject to a WHT fraud for approximately 12.7 billion DKK (EUR 1.7 billion). 

The benefit of greater transparency in terms of effectively fighting tax abuse would largely depend on the extent Member States already have rules in place at national level to counter tax abuse, such as anti-tax avoidance rules, and systems in place that can be used to detect tax abuse (eg. use of big data for risk assessment purposes). Bilateral discussions indicate that both national rules to prevent WHT abuse, and indeed systems to detect that abuse, vary widely between Member States.

In terms of costs, reporting systems would need to be implemented to allow multiple financial intermediaries to report to the source Member States. Using DAC6 data, the cost for Member States tax administrations is expected to entail EUR 1.4 million for implementation costs and recurring annual costs of 19% of implementation costs. As 13 Member States are currently considered within the scope of the proposal, the estimated implementation costs are EUR 18.2 million ( LXXIV ), with approximately EUR 3.5 million in recurrent annual costs. These estimates are consistent with the costs incurred by Finland for implementing their new relief at source system, which includes reporting by all intermediaries in the financial chain. The Finnish tax administration estimates implementation and recurring costs of EUR 3.7 million over a 5-year period.

Costs and benefits for financial intermediaries

Meetings with business stakeholders and two targeted surveys were sent to business stakeholders, (as outlined in Annex 2) with the objective of quantifying the incremental costs for the new reporting requirements. Most respondents were unable to provide cost estimates. One respondent, however, having a significant market share of the global portfolio custody market in the EU and operating in 12 Member States, provided cost data. Based on the data supplied by this respondent, the EU-27 development cost for financial intermediaries in the EU is estimated to be EUR 54.2 million with recurring costs of EUR 13 million (see Annex 2 for further details) LXXV . Comments from other respondents to the survey confirmed that reporting costs would significantly increase.

Bilateral discussions with Member States (as detailed in Annex 7) confirmed the replies from business stakeholders. Currently only withholding tax agents have reporting responsibilities with their own tax authority and not with the tax authority of another Member State ( LXXVI ). Reporting would also be required from every intermediary in the financial chain to the source tax authority, which is generally not the case now. Furthermore, the content of data to be provided under the proposal (all data should be transactional) would differ from the data currently reported to the tax authorities. In addition, for the investor to benefit from a quicker refund as envisaged under Option 3, the reporting of the information to the tax authority would need to take place within 25 days of the date of the shares going ex-dividend, while the current reporting is done on a quarterly or annual basis.

6.1.3.3Streamlined standard refund procedure

The streamlined standard refund procedure includes a standard reclaim form with minimum data fields, a standard liability allocation through common due diligence rules and a non-EU financial intermediary Register, and the possibility to perform bulk basis requests.

Cost and benefits for investors

The digitalisation and streamlined standard refund procedure would benefit investors by lowering the cost of processing the claim. As a result, more investors would request their refunds, decreasing the foregone tax relief. It leads to costs saving of EUR 3.31 billion under Option 1, as explained in table 8.

Costs and benefits for financial intermediaries

For financial intermediaries, costs savings would result from the introduction of a standard refund procedure in the EU, mainly due to its digitalisation aspects like the use of an eTRC and being able to submit a digital refund claim. Based on the data supplied by the aforementioned respondent to the targeted survey of business stakeholders, the development costs of updating their existing WHT reclaim systems are estimated at EUR 21.7 million for financial intermediaries in the EU with recurring cost savings estimated at EUR 13.5 million (see Annex 2 for further details).

The other elements of the streamlined standard refund procedure (standard liability allocation through common due diligence rules, a non-EU financial intermediary Register, and the possibility to perform bulk basis requests) are expected to have minimal cost impact on stakeholders.

In terms of due diligence, all financial intermediaries in the EU have already existing reporting obligations under DAC2/CRS with their own tax administration and are required to be compliant with KYC/AML rules. The only addition requirement would be related to providing more transparency on the financial chain and checking the entitlement to reduced tax rates. Therefore, any increase in costs for due diligence are expected to be minimal.

Member States would need to adjust their liability rules to align with the proposal, including ensuring that sanctions are dissuasive and proportionate. Apart from the costs of regulatory change no incremental costs are anticipated, as Member States already have liability rules for investors and intermediaries in terms of abiding by the current rules on WHT refunds.

From the consultations, stakeholder groups for intermediaries acknowledge the need for a robust liability framework. They stress that intermediaries should only be held liable for not fulfilling their own obligations, for example under AML, KYC and DAC2 rules, and not held liable for information provided erroneously by investors or other financial intermediaries. Accordingly, a liability framework should have clearly defined and unambiguous rules.

Establishing a common requesting method on bulk basis would imply reasonable costs as there are currently systems in place for financial intermediaries to submit refund requests on behalf of their clients. Again, we expect that the costs to move to a common format would be offset by the advantages of being able to do it in bulk through a single electronic form.

In order to ensure that non-EU financial intermediaries comply with the obligations of the standard refund procedure, a specially dedicated system can be set up that is similar to the Finnish Authorised Intermediary system for relief at source ( LXXVII ). This would require limited screening criteria for authorisation purposes with implementation and recurring costs being minimal for financial intermediaries in registering with the system and tax authorities for administrating the system. The benefit of using a register is that non-EU intermediaries would be pre-screened with eligibility criteria, for example whether they are already subject to financial regulations and tax rules equivalent to those already in place in the EU. 

Macro-Economic effects for Member States

The benefits of Option 1 for Member States are the streamlined and increased reporting of data for verification purposes for refund claims which otherwise needs to be requested on a case-by-case basis or delivered via the refund request. However, ultimately Member States would still be responsible for verifying, processing and refunding WHT refund claims, therefore the speed of the WHT refund would partly depend on their internal processes.

Therefore, while WHT refund processes are expected to improve, the expected benefits would be less than under Options 2 and 3. As already provided earlier in table 5 the impact of this option would increase EU GDP by 0.016% (50% compared to the full removal of WHT procedures) covering all investments in the EU from EU and non-EU investors.

Table 9: Impacts on GDP

Source: DG TAXUD and JRC calculation

6.1.4Option 2 - Relief at source system

In order for Option 2 to be effective, Option 1 is required to be implemented (that is an eTRC, common reporting, streamlined standard refund procedure, customer due diligence, standard liability allocation, and register for non-EU AI’s). The element of standardised refund requests on an automated basis is not applicable for a relief at source system. The costs and benefits identified for Option 1 above remain relevant for Option 2, with some additional costs and benefits stemming from the new procedure in place.

Costs and benefits for investors

This option would reduce costs the most for investors, with no delay in refund, no cost associated to the reclaim procedure and no foregone revenue. Indeed, a relief at source system is expected to result in even quicker processing of reclaims as treaty entitlement can be applied automatically, so the favourable DTT rates can be applied directly on the WHT payment rather than the investor having to submit a WHT refund claim. It results in an estimated EUR 6.62 billion cost savings annually for investors (see JRC study as explained in the baseline scenario under Chapter 5).

Table 10: Estimated benefits of WHT relief procedures – breakdown by cost driver (EUR bn)

Source: DG TAXUD and JRC calculations

Costs and benefits for tax administrations

Costs for implementing a relief at source system for tax administrations largely stem from the eTRC and reporting requirements, as assessed under Option 1. Although costs for Member States would decrease due to relief at source being applied rather than processing WHT refunds, Member States tax authorities have expressed concerns about the possibility of WHT tax abuse through the application of a relief at source system in particular with regard to Cum/Cum practices. It should be noted that with relief at source the risk of Cum/Ex is almost eliminated as this system, by nature, applies the correct WHT at the moment of the payment and, hence, no double refunds could take place.

The fact that the intermediary is abroad, especially in the case of non-EU financial intermediaries, may also make it more difficult to address non-compliance by intermediaries, including audits and monitoring, despite a statutory liability framework being put in place. Being able to recover tax may also be more difficult when the intermediary is not based in the same Member State. But the concern most often cited by Member States is that they would prefer to retain control of WHT refunds and apply their own anti-abuse tax rules for risk assessment purposes, which is more readily available under Option 3. While there will be benefits from reporting to address WHT tax abuse, however, no quantification could be made to assess the extent of such a benefit. For instance, under the new Finnish relief at source system, the average WHT rate over dividends paid to non-resident portfolio investors raised from 14.30% to 22.73% from 2020 to 2021 (year in which the reporting system was introduced) even when more than 50% of investors had applied relief at source. According to the Finnish tax administration it partially may stem from the fact that the new system hinders the use of abusive practices( LXXVIII ).

Costs and benefits for Financial Intermediaries

Costs for financial intermediaries that are associated to the eTRC and reporting have already been discussed under Option 1. In addition, financial intermediaries would incur costs from setting up and running the verification system that is now undertaken by the tax authorities of Member States under the current refund systems. Financial intermediaries would be responsible of ensuring that the relief at source system is correctly applied. This would mean they would potentially incur greater liabilities than under the baseline scenario. However, financial intermediaries would have all the necessary data and documentation to perform the verification checks under this option.

Under the relief at source system, financial intermediaries would have to report information ex-ante to, the WHT agent in the source Member State to ensure that the correct WHT rate is applied. The reporting mechanism is similar to Option 1 so no difference in costs for intermediaries between Options 1 and 2 are expected for this reporting.

In a targeted survey of financial intermediaries regarding the new relief at source system in Finland, 3 respondents noted that recurring costs of implementing the Finnish system had increased by 61% compared to the existing reporting costs and that they incurred implementing costs of EUR 1.6 million. For benefits, financial intermediaries noted that customer satisfaction improved and also that the liability risk for financial intermediaries had improved. However, as the system was only implemented in 2021, respondents noted that it was difficult to estimate both the benefits and recurring costs of operating the system.

Macro-Economic effects for Member States

Option 2 is expected to increase GDP the most compared to the baseline scenario with a total of 0.031%.

Table 11: Impacts on GDP of Option 2

Source: TAXUD and JRC calculations.

As explained in Annex 4, a CORTAX study by the JRC, taking into account retroactive effects, estimates for this option even higher effects for GDP growth of Member States with an increase of 0.047%. However, full benefits are unlikely to be achieved given that some refunds would still need to be processed, for example, when relief is not possible due to the absence of key documents provided by the investor/financial intermediary, or for risk assessment reasons.

6.1.5Option 3 – Relief at Source System (‘RAS’) or Quick Refund System (‘QRS’) and within a set timeframe

Option 3 will also encompass all the elements covered under Option 1.

Costs and benefits for investors

No additional costs are anticipated for investors as documents supplied by them already need to be supplied for WHT refunds. Investors would benefit from less foregone WHT refunds and quicker refunds. The time to process the refund is expected to be very short and so the opportunity costs would be reduced in Option 3.

In the Economic Impact for the 2009 Recommendation, Member States noted a significant decrease in costs from switching from a refund process to letting the intermediary verify the refund. Ireland moved to a refund system for Qualified Intermediaries in October 2006 that resulted in an increase of refunds issued within 20 days from 23% in 2006 under the old system to 100%. This allowed the Irish tax authorities to concentrate more resources on non-QI refunds and turnaround times have improved from 29% processed within 20 working days in 2006 to 75% processed within 20 working days in 2007. After the Netherlands introduced refund procedures for securities held with intermediaries -the procedures allow foreign and domestic custodians to file refund claims on behalf of their clients in a standardised electronic format- it managed to cut down a very labour-intensive process taking 2 to 3 months on average to 3 weeks.

It is estimated that the cost of reclaiming would be divided by two in Option 3 compared to Option 1 from 1% to 0.5% of the amount reclaimed. Similarly, due to this decrease of costs, it can be anticipated that only half of the investors in Option 3 would not reclaim the refund, so 2.5% compared to 5% in Option 1 and 10% in the baseline scenario estimate of the JRC in Chapter 5.1. Building on the Commission study and the above assumptions, the reduction of costs of the Quick refund systems under this option are estimated to be at EUR 5.17 billion.

Table 12: Estimated benefits of WHT relief procedures – breakdown by cost driver (EUR bn)

Source: TAXUD and JRC calculations.

These costs compared to the estimates outlined for the baseline scenario described in Chapter 5 would result in cost savings of 78.1% as compared to the baseline scenario. In terms of the one In, one out approach, the benefits of a decrease in paperwork and streamlining the different WHT requirements of Member States would lead to a reduction of the incurred costs of EUR 730 million per year (that is 75% of incurred costs of 970 million under Option 2 LXXIX ). As the estimates relate to both EU and non-EU investors  and 56% of the cross-border holding of securities is within the EU (see Annex 5, table 1), the assumption is made that the cost savings is EUR 409 million annually for EU investors.

Costs and benefits for tax administrations

Costs for implementing this option have already been partly assessed under Option 1, as most costs would be related to the reporting under Option 1. However, tax administrations may still need to introduce further automation to perform the checks and the risk management in order to make the required assessment within the short timeframe. Further, amendments might be necessary to accommodate the pay out of the refunds. However, as most refunds will be done via the QRS or relief at source, efficiency gains should arise as well, so that in the longer term less resources would be required, or the resources could be allocated to dealing with higher risk cases.

If they opt for QRS, tax administrations can continue to control the refund process under this option by processing the WHT reclaims submitted by the investor, or the financial intermediary on behalf of the investor. Consequently, Member States tax administrations would be able to apply more easily their risk assessment as long as Member States already have their national systems in place to use data obtained under the reporting mechanisms as described under Option 1. While there are benefits associated to reporting, however, no quantification could be made to assess the extent of such a benefit in countering WHT abuse specifically for Option 3.

Costs and benefits for financial intermediaries

There will be costs associated to the implementation of the IT system and recurrent costs linked to the reporting obligations, as discussed under option 1. The system of streamlining the information to be provided by financial intermediaries, including digitalisation and standardisation at EU level, as discussed under option 1, is expected to produce considerable savings as well.

Macro-Economic effects for Member States

Option 3 is expected to provide a quicker reclaim process compared to options 1, but less (although not much more) than Option 2 Relief at Source system where no refund system is required. The effects on GDP are an increase in GDP of 0.025% compared to 0.031% under Option 2.

Table 13: Impacts on GDP of option 3

Source: TAXUD and JRC calculations.

6.1.6Limitations of data

The analysis in this chapter builds upon limited data. For the calculation of WHT costs for investors and economic benefits for Member States, the Commission studies (as included in Annex 4) rely on a number of macro-economic assumptions, for example the proportion of portfolio securities held by retail investors that may not correspond with microdata like the number/amounts of reclaims on an annual basis, as described below, or dividend income paid to non-resident retail investors in the EU. The Commission study estimate only the direct holdings by retail investors, which tends to underestimate the amount of WHT refunds in the EU.

Due to the non-exhaustiveness of the number of refunds handled by tax administrations, as collected by the Commission in 2021 for the years 2019 and 2020, the assessments of the costs were not based on these data, to approximate the cost for investors and tax administrations.

As outlined in Annex 2, the Commission undertook two targeted surveys with business stakeholders: to assess the reporting costs for business as outlined under Option 1; and the experience of Authorised Intermediaries with new relief at source system in Finland. Although the responses to the reporting cost survey were useful for qualitative aspects, only one stakeholder managed to supply quantitative data for the incremental reporting costs of business which was further assessed under Option 1.

Furthermore, to approach the cost for Member States to put in place a system that would enable a quick refund procedure or a relief at source procedure, the costs incurred by the Finnish tax authority, that developed a system comparable to TRACE, implemented since 2021, were used to approximate such a cost for other Member States, as well as the costs incurred to implement DAC.

6.2General impacts of the initiative

6.2.1Social and Environmental Impacts

The initiative is expected to have a positive social impact. It would ensure a fairer taxation of capital income for investors by reducing double taxation and ensuring that WHT refunds are promptly paid. Furthermore, the initiative would contribute to a level playing field for investors by ensuring that passive income payments derived from cross-border securities in the EU are taxed similarly to passive income payments derived from domestic securities.

The various options are expected to enhance the abilities of tax authorities in the EU to combat tax evasion and tax avoidance, thanks to the introduction of a common reporting system. This information can be used for audit, risk assessment and monitoring purposes. In particular, reporting transaction data to the tax authorities would mean that once the tax certificates or other forms of WHT refund can be directly linked to the underlying distribution, it would be impossible to make multiple repayments under a single distribution as in the Cum/Ex scandals. Data reporting would also help tax authorities to identify unusual transaction flows for risk assessment purposes to combat Cum/Cum abuses like dividend arbitrage.

The proposal would have a small positive impact for the environment given the expected reduction in paper-based refund process through the uses of a common tax residence digital certificate and automated refund requests. Therefore, the current initiative is consistent with the fulfilment of the climate-neutrality objective as requested by the European Climate Law.

6.2.2Fundamental rights

Fundamental rights, in particular the requirements concerning the protection of personal data under the General Data Protection Regulation (‘GDPR’), would be safeguarded. Investors in possession of the electronic tax residence certificate, whether included in a digital wallet or kept in dedicated depository, would decide themselves whether their personal data would be provided to their financial intermediary to submit a WHT reclaim to a WHT agent or tax authority in the source Member State.

The exchange of personal data, as contained in the common reporting and standard refund procedures of option 1, between the investor and the WHT agent or the tax authority in the source Member State would only be transmitted for the purposes of ensuring the information reporting obligations of the financial intermediaries and the tax obligations of the withholding tax agent with respect to the refund to be reclaimed. The amount of personal data to be transmitted will be limited to what is necessary to detect underreporting, non-reporting or tax abuse, in line with the GDPR requirements.

7HOW DO THE OPTIONS COMPARE?

This chapter makes a comparison in light of the impacts identified. The options are assessed against the three core criteria of effectiveness, efficiency, and coherence. The proportionality of the options is also part of this chapter.

7.1Effectiveness

The specific objectives against which the effectiveness is evaluated are as follows: to achieve more effective WHT procedures and to improve Member States´ capabilities to fight against tax abuse. On making WHT procedures more efficient, the following criteria will be assessed: (1) shorten time for refunding excess tax withheld; (2) simplify the reclaim procedures; (3) promote digitalisation and harmonised procedures; (4) ensure equal access to DTT benefits for portfolio investors similar to what direct investors have. These criteria would have an effect to lower tax-related costs for cross-border investors and facilitate the cross-border investments hence attracting investment to the European market.

On improving the ability of Member States to detect and counter cross-border tax fraud, -avoidance and -evasion, the following elements to check are: (1) tools for Member States to obtain more information to tackle tax fraud through a higher level of transparency, and (2) the ability to avoid future tax losses.

It is worth noting again that both objectives – improving efficiency and preventing fraud – have the same importance within the initiative: greater efficiency should help to reduce the period within which the investor will receive his/her taxes back, which should benefit all stakeholders; while fighting abuse will be achieved through greater transparency, which will improve compliance and provide tax administrations with better insight to assess the final amount that should be refunded. The weight of each criteria reflects the fact that each of these two objectives has the same importance:

- Each of the four criteria under making WHT procedures more efficient (table 14 below) is scored from -3 to +3 (-3 indicating the most negative impact, +3 the most positive impact, 0 is the baseline scenario).

-The score of the two criteria related to tackling tax fraud (table 15) will range from -6 to +6 instead.

Note that for scoring purposes, during the whole chapter 7, option 3 is taking into consideration mainly the Quick refund system as it is the procedure that Member States need to implement to comply with minimum standards of efficiency and security. In addition, even those Member States opting for Relief at Source need to have in place a refund procedure as a back-up system, which tends to be quick given the reporting obligations to be put in place.

Table 14: Efficient WHT procedures

Source: TAXUD 

Table 15: Tackle tax fraud

Source: TAXUD

As seen above, both options 2 and 3 perform similarly in terms of efficiency (nonetheless option 2 performs slightly better, that is why they score almost equal: 12 vs 11), while they are more different in terms of the second objective, fighting against tax fraud (9 vs 12).

Option 1 entails introducing an eTRC across Member States, a common standard of reporting for the actors in the financial chain and the possibility to file bulk requests. The reporting obligations provide Member States’ tax administrations with the information needed to properly assess whether the benefits of the tax treaties should be granted and to check that taxes due are paid. Gathering relevant information is crucial to fight tax abuse and to achieve swifter refund procedures. In addition, it will also be possible for the tax administrations to better perform risk management with regard to its taxpayers and resources could be allocated to the high-risk cases. Having in place standard reporting obligations, common due diligence procedures with proper liability allocation and bulk requests may speed up and simplify as well to certain extent the existing WHT procedures, granting better access to treaty benefits. Therefore, option 1 scores slightly above the baseline scenario but lower than option 2 or 3 in terms both of efficiency and tackling tax abuse.

Option 2 consists of implementing at EU level a Relief at Source system (RAS) in addition to the main elements included in option 1( LXXX ). RAS implies that tax administrations would only be able to perform the checks once the payments are done, which makes the system less effective in terms of combating tax fraud and tax abuse than option 3, under which Member States have the possibility to apply a QRS. Based on the abovementioned reasons, some Member States, severely affected by Cum/Ex scandal, are more reluctant to implement a RAS system where the control ex-ante is to be outsourced with financial intermediaries. This makes Member States perceive the QRS more secure in comparative terms.

Regardless of these shortcomings, the procedure itself would eliminate the existing barriers to the free movement of capital regarding the cross-border investments by EU and non-EU investors. Therefore, option 2 performs the best in terms of the objective of making more efficient WHT.

Under Option 3 Member States can implement a quick refund system, but can also implement RAS like in option 2 if they are comfortable to do so.

The QRS removes existing barriers to cross-border investment and ensures that investors have proper access to DTT benefits. Therefore, it improves the functioning of the capital market union. Consequently, even when it scores slightly less than option 2 in terms of efficiency of the procedures (assuming that not all Member States will opt for RAS under option 3), option 3 should as well address the objective of making WHT reclaim procedures more efficient by reducing the time period of the refund and simplifying the procedure.

On the other hand, the objective to improve the capabilities of the Members States to fight against tax abuse and to prevent fraudulent tax schemes in the future is best achieved under option 3 than under option 2. To the extent that Member States apply a QRS rather than RAS, the checks by the tax administrations will be ex ante, which is an advantage in terms of fighting against tax abuse compared to systems where the checks are done ex post such as relief at source systems.

Therefore, the main difference between option 2 and option 3 is that, under option 3, tax administrations apply QRS still have the control as checks would take place ex ante before paying out to the investor the tax withheld in excess. Under option 2, the checks to be performed by the tax administrations would only be done ex post, after the WHT agent has withheld the lower WHT and paid the amount to the investor. Hence, in terms of the fight against tax abuse, the advantages of option 3 outweighs option 2, based on the possibility of maintaining control on the reclaim process. However, option 2 achieves a better outcome in terms of efficient WHT procedures.

7.2 Cost effectiveness

The efficiency (effectiveness in relation to cost) scores from -3 to +3 (-3 indicating the most negative impact, +3 indicating the most positive impact) based on the assessment of the options.

Table 16: Cost effectiveness

Source: TAXUD 

Option 1 mainly brings the eTRC, enhanced reporting obligations throughout the financial chain and bulk requests. Investors would not have additional costs as they already must comply with EU AML(KYC) and DAC2/CRS obligations and enjoy the benefits of a digital TRC; it entails more efficiency for investors than the baseline scenario. The intermediaries also have benefits from the digitalisation and streamlining of reporting, but not to an extent that matches the implementation costs, which makes option 1 the least beneficial in terms of cost-benefit analysis compared with options 2 or 3. The tax administrations end up in a much better position than today, especially when it comes to access to relevant, updated data and in a format that is easy to use. This is in the end positive for the Member States as well to some extent. The main beneficiary from the perspective of efficiency is the tax administrations.

Option 2 consists of implementing the RAS across the EU. This increases the value for the investors based on speed of obtaining the correct amount alone. For the intermediaries, option 2 entails establishing a simple and quick system that is in place in multiple jurisdictions and that removes tax barriers benefiting market growth accordingly. The downside is the cost for operating the relief system as such. The option is still quite beneficial to tax administrations, but they will be faced with costs for conducting checks and audits on the intermediaries. Overall, this is the best solution to investors and intermediaries from an efficiency perspective as well as from a macro-economic analysis point of view.

Option 3 implies a choice between a QRS and/or a RAS. Compared to option 2 and for the Member States that would choose the QRS, it is slightly less positive than option 2 from the point of view of the investors, mainly because it is not as quick as the immediate relief. Here the intermediaries do not face the cost of operating the process, but they still need to afford the implementing cost for complying with reporting obligations and due diligence procedures. The tax administration will be faced with operating the system, but do not have to do costly audits for validation purposes, while they also maintain control of the relief payments. This a good solution for tax administrations that would incur in implementing costs notably outweighed by the advantages achieved in fighting against fraud.

7.3Coherence with general objectives of policy initiatives and EU law

The initiative responds to core EU strategies to fight against tax avoidance and tax evasion, and reinforce the functioning of the capital market union, to the benefit of the single market. Each option is coherent with these EU priorities to different extent:

Table 17: Coherence

Source: TAXUD

The introduction of a common eTRC helps to implement a digitalised and standardised system regarding the request and issuance of tax certificates. The reporting obligations of option 1 enable the tax administrations to receive information about the financial chain when dividends are paid out. These measures are major progress to fight tax abuse and fraudulent tax arrangements. Although the measures are already a huge step forward and are an improvement compared to the current situation in terms of effectiveness and fighting against tax fraud, the ultimate goal is to reinforce a Capital Markets Union without obstacles whereby there is a common system for a swift refund of WHT. This aim can only be reached when measures like those proposed in option 2 or option 3 are additionally implemented in the EU.

In addition, the requirements of the GDPR will be met. Regardless which technical basis will be chosen for the eTRC, the persons concerned decide themselves whether their personal data would be provided to their financial intermediary to submit a WHT reclaim to a WHT agent or tax authority in the source Member State. On the reporting obligations of all options, the information will only be transmitted for the purpose of ensuring the fulfilment of tax obligations. That means that the personal data will be transmitted only to the extent that is necessary to detect underreporting, non-reporting or tax abuse in line with GDPR requirements.

7.4Proportionality

Proportionality refers to the extent to which an initiative is suitable or appropriate for the purpose of achieving the desired objectives, does not go beyond what is necessary to achieve these objectives and limits the scope to those aspects the Member States cannot achieve by their own. The options proposed meet the requirements of proportionality by both improving WHT reclaim procedures and preventing tax abuse. Due to the fragmentation and problems identified in EU WHT reclaim systems, in particular their impact on cross-border investors and financial intermediaries, an EU intervention is considered necessary.

Table 18: Proportionality

Source: TAXUD

Under option 1, implementation costs of national IT-systems will occur when the eTRC is implemented. The implementing and recurring costs of the system are acceptable, taking into account the benefits. A current fragmented system with partly paper-based documents is replaced by a completely digital system. It will increase the digitalisation of administrative processes by Member States and achieve efficiency gains, also enabling the intermediaries to improve their processes. This is one step forward to achieve more efficient WHT procedures.

The implementation of additional reporting obligations proposed under option 1 imposes additional administrative burden and costs for financial intermediaries as the reporting of information must be extended to include more granular data, including the identity of the beneficiary. These costs are, nonetheless, outweighed by the positive impact the information received by tax administration has on the improvement of WHT procedures. However, under this option, financial intermediaries do not see a clear return in the form of a harmonised system. On the contrary, option 3 is setting up at minimum a standardised quick WHT refund system for the sake of investors, keeping similar reporting obligations than under option 1. Similarly, under option 2, whereby the main burden is borne by financial intermediaries, a common system is established as well to get relief at source.

All in all, all the options entail a faster and smoother refund/relief system to a different extent, and they all pursue the aim of preventing future fraudulent tax schemes which were responsible in the past for the huge losses of tax revenue. In light of this, the additional obligations embedded in each option do not go beyond what is necessary to achieve efficiency and fair tax systems. Therefore, the proposed options are deemed as proportional.

8Preferred option

Based on the outcome of the previous section, where a detailed comparison was done across all the options, the scores of each criterion and the overall assessment of the three options are shown in the table below:

Table 19: Overall assessment

Source: TAXUD

Therefore, the above analysis indicates that option 3, which requires to implement either a Quick refund system (QRS) or/and Relief at Source (RAS), is the most appropriate to meet the objectives of the intervention. Overall, we can conclude that setting up a QRS or/and a RAS, is highly effective to tackle the problems identified in the EU in terms of speed, simpler processes and more digitalised procedures.

At the same time, option 3 is safer as it gives Member States the option to retain an ex-ante control. Under QRS, refunds take place only after the dividend/interest income payments have reached the beneficiary which provides more control to the tax administrations in terms of identifying potential abusive cases. This is especially relevant for Member States that have been heavily hit by Cum/Cum and Cum/Ex practices during recent years and that, because of political reasons, might be more reluctant to adopt a RAS system in the short-term. Some other Member States may decide to go for a RAS, with the advantages it entails in terms of timing for the taxpayer.

In terms of cost-benefit ratio, a choice to implement QRS or/and RAS represents a fair compromise solution between stakeholders, as financial intermediaries, in general, prefer a relief at source system and some Member States will rather go for a system whereby they can retain control over refunds. For investors, the QRS entails similar advantages compared to a RAS in terms of reducing compliance costs as the submission of the request will be done on their behalf.

In addition, agreeing on a QRS across the EU and/or RAS for those Member States willing to, is coherent with policies currently been undertaken at EU level. This initiative complements the Unshell proposal (e.g. by including relevant information for “Unshell” in the common eTRC), DAC2 (e.g. by providing for a greater level of granularity when it comes to the reporting of financial income, the reporting of WHT levied and refunded, and by ensuring that all relevant WHT agents/securities are included within its scope) and Shareholders Rights Directive ( LXXXI ) (e.g. by sharing the same aim of providing transparency in relation to the final investor). Implementing a QRS and/or RAS is consistent as well with the Commission’s priorities to combat tax abuse and strive for a strong capital markets union. 

Concerning data protection, the initiative requires the processing of personal data since the investors can be natural persons. However, this can be considered justified as the proposed measures are proportionate and necessary to fight against tax abuse, the personal data processing is limited to what is necessary to achieve this objective and the initiative will fully comply also in other respects with the requirements of EU data protection law/GDPR.

The principle of subsidiarity is met both by QRS or RAS as they both clearly offer the EU added value by introducing an effective and, at the same time, secure system across all Member States. The public consultation made it clear that EU action was needed, with 94% of positive replies, in order to harmonise the patchwork of current systems in place. Fragmentation only drives compliance costs upwards both for investors and financial intermediaries.

It is also important to emphasise that the preferred option is proportionate and does not go beyond what is needed to achieve the goals. The overall balance is positive: the expected return in terms of effectiveness for investors, financial intermediaries and tax administrations are higher than the estimated implementing costs.

Finally, Option 3 complies with the objectives of the initiative and is also the most feasible to achieve in the short-term. It is the also the option that should have the highest acceptance rate among Member States.

One in one out approach

In terms of the ‘one in, one out’ approach, the benefits of a decrease in paperwork and streamlining the reporting WHT requirements of Member States would lead to an estimated reduction of the incurred costs of EUR 409 million per year for private stakeholders and a reduction of the recurring costs of 13,5 million per year for financial intermediaries; on the other hand, reporting obligations and streamlining procedures would lead to implementing costs of EUR 75.9 million and recurring costs of EUR 13 million per year subsequently.

9How will actual impacts be monitored and evaluated

9.1Indicators for monitoring and evaluation

Monitoring should act as an incentive to properly implement the obligations under the proposed framework for the main actors in the current intervention. Option 3 is the preferred option. The table below gives an overview of the objectives, the indicators to measure whether they will be achieved under option 3, and the tool for monitoring them:

Table 20: Monitoring and evaluation framework

Source: TAXUD

9.2Monitoring and reporting 

Data requirements for monitoring purposes under option 3 (the following data should be reported to the Commission on an annual basis by Member States):

-eTRC: details on the implementation and issuance of the eTRC to investors by the Member State of residence of the investor.  

-Enhanced transparency: assessments by the Member State of the compliance of their financial intermediaries with their reporting obligations, in particular in terms of the quality and timeliness of the data exchanged with the tax authorities of the source Member States.

-Quality and completeness of information reported to the source Member States: the number of WHT refunds per Member State of the financial intermediary that cannot be refunded within the 40 days due to incomplete or poor-quality reporting by a financial intermediary, including the reclaims.

-Liability: Member States should collect information on number of liability actions against financial intermediaries (resident and non-resident) subject to the proposal– opened, closed and pending liability cases.

-Enforcement activities: The type and number of enforcement activities undertaken by Member States to ensure that their financial intermediaries comply with the customer due diligence and reporting obligations; and statistics on the registration of both EU and Non-EU authorised intermediaries as well as de-listings.

-Sanctions: the number and value of sanctions imposed by Member States on their financial intermediaries for failing to comply with customer due diligence and reporting obligations under the proposal.

-Payment deadlines: The number of days that it takes to make a WHT refund by the source Member State – WHT refunds made within 40 days and beyond 40 days. Reasons for WHT dividend payments outside the 40 days should be reported including the reasons for i.e., potential tax abuse cases and incomplete information. Further, statistics on the number and amount of interest payment penalties paid to investors where the Member State has failed, without a valid reason, to refund the WHT within the payment deadlines.

-Relief at source: The volume of dividends/interests paid with the correct WHT rate applied at the moment of payment in relation to the whole amount of dividends/interests paid to non-resident portfolio investors.

-Reported information: The use of data reported to the tax authority of the source Member State for risk assessment purposes – the number and amount of WHT abuse detected due to information received.

9.3Monitoring and evaluation by the Commission

The data collected under 9.2 will be published by the Commission on annual basis in a report. The economic effects of the proposal will also be assessed in the report. Member States and the Commission will examine and evaluate the functioning of this Directive every 5 years based on the data provided. A report on the evaluation of the Directive will be submitted to the European Council.

Member States will be asked to communicate to the Commission any relevant information necessary for the monitoring and evaluation of the Directive in improving procedures in the European Union to reduce double taxation related to withholding taxes as well combating tax abuse of withholding tax systems in the European Union.

Annex 1: Procedural Information

1)LEAD DG, AGENDA PLANNING AND WORK PROGRAMME

The proposal for the New EU common system for the avoidance of double taxation and prevention of tax abuse in the area of withholding taxes was prepared under the lead of Directorate-General for Taxation and Customs Union (DG TAXUD) in close collaboration with Directorate General for Financial Stability, Financial Services and Capital Markets Union (DG FISMA).

In the Agenda Planning of the European Commission, the project is referred to under item PLAN/2021/10794. The 2020 Commission ‘Action Plan for Fair and Simple Taxation supporting the recovery strategy’ includes this project in the action eight under the header to increase the efficiency of tax payment procedures during 2022/2023.

2)ORGANISATION AND TIMING

Work on the preparation and fact-finding phase of this policy initiative started in June 2020. An Inter-service Group set up in June 2021 assisted DG TAXUD in the preparation of this impact assessment report. The Inter-service Group included colleagues from the following Commission services: DG FISMA, JRC, DG ECFIN, DG BUDG, DG COMP, DG GROW, DG JUST, OLAF, DG TRADE, DG CNECT, DG DIGIT, the Commission's Legal Service and the Commission's Secretariat-General.

The Inter-service Group met on four occasions between June 2021 and November 2022 (24 June 2021, 14 June 2022, 6 October 2022 and 8 November 2022). During the first Inter-service Group meeting an overview of the initiative was shared with the participants and the Inception Impact Assessment coupled with the problem tree paper and the consultation strategy document were discussed. Introduction, problem statement, objectives and the need of EU action of the initiative were covered by the second Inter-service Group meeting held in June 2022. In October 2022, the available policy options and their impact were addressed during the third Inter-service Group meeting. The last meeting of the Inter-service Group took place on 8 November 2022 whereby the draft of the entire Impact Assessment was discussed before submitting it to the RSB. After a negative opinion of the RSB a revised version of the Impact Assessment was discussed on 7 March 2023. At each occasion, the members of the Inter-service Group were given the opportunity to provide comments orally and in writing on the draft versions of the documents presented.

3)CONSULTATION OF THE REGULATORY SCRUTINY BOARD

The draft impact assessment report was submitted to the Regulatory Scrutiny Board on 16 November 2022. The impact assessment report was reviewed by the Regulatory Scrutiny Board on 14 December 2022. The RSB delivered a negative opinion. The revisions introduced in response to the first RSB opinion are summarised in the table below.

The impact assessment report was resubmitted to the RSB on 20 March 2023 with revisions introduced in response to previous opinion, the RSB issued a positive opinion with reservations on 21 April 2023 to this resubmitted report. The following chart summarizes how RSB’s comments were addressed this second time:

4)EVIDENCE, SOURCES AND QUALITY 

The Commission has been consulting widely and has received input from various sources on this impact assessment work. The following expert advice has fed into the Impact Assessment:

·ESMA: ‘Final report On Cum/Ex, Cum/Cum and withholding tax reclaim schemes’, September 2020, the European Securities and Markets Authority (ESMA)

·ESMA: ‘Report on preliminary findings on multiple withholding tax reclaim schemes’, July 2019, the European Securities and Markets Authority (ESMA)

·ESMA: ‘EU Securities Markets ESMA Annual Statistical Report’, 2021

·JRC studies as explained in Annex 4

·‘European Post Trade Forum Report’ and ‘Detailed analysis of the European Post Trade Landscape’, 2017

· The Giovannini Group. ‘Cross-border clearing and settlement arrangements in the European Union’ (Nov.2001)  

· The Giovannini Group. ‘EU clearing and settlement arrangements’ (April 2003)

·FISCO Group: ‘Fact finding study: Fiscal compliance procedures related to clearing and settlement within the EU - Fiscal Compliance Experts' Group - 19 April 2006.

·FISCO Group: ‘Fiscal compliance barriers related to post-trading within the EU - Fiscal Compliance Experts' Group - 23 October 2007

·FISCO Group: Booklet: Simplified withholding tax relief procedures - Fiscal Compliance Experts' Group - 9 February 2010

·‘Workable solutions for efficient and simplified fiscal compliance procedures related to post-trading within the EU’ report from the Tax Barriers Business Advisory Group – T-Bag

·OECD: ‘Treaty Relief and Compliance Enhancement (TRACE)- Implementation Package’ approved by CFA, 2013

·Public Consultation: 1682 replies and 51 position papers from associations, companies in general, banks, consulting firms/advisors and academics.

·Target consultations via physical/online meetings and in written feedback from: OECD on TRACE experts, GLEIF (Global Legal Entity Identifier Foundation), EBF (European Banking Federation), AGC (Association of Global Custodians), EFAMA (European Fund and Asset Management Association), EY Blockchain experts for taxes initiative, ICI Global (Investment Company Institute), Euroclear, EBSI experts (European Blockchain Service Infrastructure), AFME (Financial Markets association), JITSIC at OECD (Joint International Taskforce on Shared Intelligence and Collaboration) and several experts from Member States.

The Commission services have taken into account all of above-mentioned sources of expert advice in the present impact assessment.

Annex 2: Stakeholder Consultation

1)Introduction

Table Overview of consultation activities

2)Stakeholder views of the Options presented in the Impact Assessment

2.1. Option 1 – Setting up a common EU digital Tax Residence Certificate and common reporting

The proposed introduction of a common eTRC is supported by all stakeholders. For investors, who need to provide evidence of their tax residency, this is part of a faster and more uniform procedure and will also be part of improving the Capital Markets Union. The financial intermediaries are primarily in favour of an investor self-declaration of residency instead, but they do not oppose the digital solution, which enables automation for them. The self-declaration was not considered as a way forward in the proposal as Member States considered that they required evidence of residency from the tax jurisdiction of the investor. Overall, Member States are also positive with the introduction of an eTRC although some Member States have concerns with regard to its implementation costs.

Financial intermediaries are aware that enhanced transparency is required, however, compliance costs should be kept to a minimum. Further, they consider that having an EU reporting standard would ensure unform reporting throughout the EU as opposed to national reporting solutions. Financial intermediaries have some concerns on the amount of information to be sent as well concerns regarding sharing customer information, in particular for customers who do not wish to claim a refund.

In general, Member States are in favour of the reporting of WHT information by financial intermediaries to their tax authority. They consider that such information could be useful in combatting tax abuse. With respect to the refund procedure, Member States consider that standardising the reclaim form could be beneficial although there are minor differences between Member States on what should be included and how the reporting should take place.

In regard to a standard liability procedure financial intermediary have stated that they should only be held liable for their own errors and that if information is supplied to them that is erroneous, for example by the investor, then the intermediary should not be held responsible if due diligence procedures have been correctly carried out. Most Member States recognise this apportionment of liability in their own Member States national rules. However, one large Member State has rules which imply strict liability of information handled by their withholding tax agent and that the withholding tax agent should seek their own redress from other intermediaries in the financial chain or the investor for information that is incorrect. With regard to a standard streamlined refund procedure, including bulk requests investors are only indirectly affected by this as they already are required to supply information for reclaims and for existing customer due diligence procedures. Rather, a streamlined refund procedure (quicker assessment of correct tax) will rather ultimately benefit investors. Reduction in the burden of filing reclaims has been noted by investors in the Public Consultation. Financial intermediaries have also welcomed the opportunity to streamline refund procedures as much as possible to reduce their administrative. Member States recognise the benefit of streamlining refund procedures, however, the information to be supplied should reflect their national requirements to ensure a correct application of WHT relief.

2.2 Option 2 - Relief at source system

For the investors this is the quickest method and consider this as the optimal option. The Financial Intermediaries are also in favour of this option although they acknowledge that some form of refund is also necessary.

Most Member States are favourable for this option although would require safeguards to ensure the system is compliant as well being able to detect abuse promptly in order to prevent WHT abuse. A smaller group of Member States, including a large Member State, are currently not in favour of this option as they consider that all WHT claims need to be controlled by their tax administration, rather than WHT relief being provided by a financial intermediary, in order to prevent tax abuse and would prefer to retain a WHT refund system. This second group puts more emphasis on extensive data gathering, appropriate WHT refund deadlines, and due diligence and liability rules for financial intermediaries. In the latter group, which is smaller, there tends to be Member States that suffered more losses from the Cum/Ex and Cum/Cum issues.

2.3 Option 3 - Relief at source system or Quick refund system

The Quick refund system was not part of the Public Consultation unlike the Relief at source system. From the replies to the Consultation, the dominant opinion of investors was that the reclaim systems should be reformed to provide prompt WHT relief therefore the Quick refund system would achieve this although less effectively compared to the Relief at source system.

Although financial intermediaries would prefer the Relief at source system, they understand that a WHT refund system will still need to operate, in particular to address the concerns of some Member States on WHT tax abuse.

As with Option 2) Member States are split in two groups. Most Member States are open to an initiative to provide prompter WHT relief either through a quick refund or relief at source system. However, a minority of Member States emphasise the importance of a system which has sufficient controls to prevent tax abuse over the need to have a system which provides prompt WHT reclaims. Therefore, a refund system would allow such control to be exercised by Member States before granting WHT relief.

3)Consultation Participation

3.1 Inception Impact Assessment (IIA)

Category of respondent

The consultation period was between 14 May and 11 June, 2021 via the Commission website ( LXXXII ). 33 Responses were received. As shown in Figure 14, most of the respondent were business associations and companies from the financial sector (financial institutions and big scale investors):

Figure 20: Overview respondents

Source: TAXUD

Figure 21: Overview respondents

Source: TAXUD

A range of options were presented in the IIA:

Option 1: Improving withholding tax refund procedures to make them more efficient: This option entails the implementation of several measures, the objective of which is to simplify and streamline withholding tax refund procedures by making them quicker and more transparent. These measures are not limited by but could include: the establishment of common EU standardised forms and procedures for withholding tax refund claims irrespective of the Member States concerned and the obligation to digitalise current paper-based relief processes.

Option 2: Establishment of a fully-fledged common EU relief at source system: This option entails the implementation of a standardised EU-wide system for withholding tax relief at source whereby the correct withholding tax rate, as provided in the DTT is applied at the time of payment by the issuer of the security, to the non-resident investor thereby not incurring double taxation.

Option 3: Enhancing the existing administrative cooperation framework to verify entitlement to double tax convention benefits: This option envisages a reporting and subsequent mandatory exchange of beneficial owner-related information on an automated basis, to reassure both the residence and source country that the correct level of taxation has been applied to the non-resident investor.

Key results:

There was broad support for all three options with agreement on the urgent need for EU-wide regulatory harmonisation of forms and processes in the WHT sphere and emphasis on digitalisation. In terms of content preference, the responses split between option 1 (digitalisation) and option 2 (relief at source). Further, many of the responses suggested that all three options are complementary so all of them should be addressed.

3.2 Public Consultation

Overview

A total of 1683 responses were received, one of which has been deemed invalid due to containing profanity and has been discarded, leaving 1682 valid responses. Regarding stakeholder types of respondents, EU citizens accounted for 92.0% of responses (1547 of 1682), 3.5% of responses (59 of 1682) came from companies/business organisations and 2.4% of responses (40 of 1682) from business associations.

Figure 22: Stakeholder type – all respondents

In regard to country of origin, 57.2% of responses (962 of 1682) came from Spain, 20.1% (338) from Germany, and 10.0% (169) from Portugal. As for the remaining responses, 10.5% (177) originated from other EU countries and 2.2% (37) were from non-EU countries.

Figure 23: Country of origin – all respondents

When only looking at the country of origin for companies/business organisations and business associations, 17.2% of responses (17 out of 99) originated from France, 15.2% (15) from Belgium, 11.1% (11) from Germany, and 21.2% (21) from non-EU countries.

Figure 24: Country of origin – companies/business organisations & business associations

Main issues

89.0% (1495 of 1679) of respondents strongly agree that the current functioning of WHT refund procedures in Member States hinders cross-border investment in the EU securities market. 92.0% (1430 of 1555) of EU citizens strongly agreed, compared to 50.0% (49 of 98) of companies/business organisations and business associations.

In terms of payments where the issue of inefficient WHT procedures is relevant, dividends from listed companies received 98.93% of responses (1662 of 1679). More than 78% of companies/business organisations and business associations also deemed dividends from unlisted companies and interest related to debt instruments in listed companies as relevant, compared to less than 35% of EU citizens for those two types of payments.

Figure 25: Payments affected by inefficient WHT procedures – all respondents

In the opinion of respondents, the nature of the problems with existing WHT refund procedures considered to be of “high importance” include the following: lengthy WHT refund procedures were cited by 88.69% of respondents (1451 of 1680), lack of digitalisation and non-user-friendly forms by 88.13% (1441 of 1680), and costly WHT refund procedures in monetary terms by 79.59% (1283 of 1680).

Figure 26: Nature of the problems with existing WHT refund procedures – all respondents

As for the consequences due to the problems with WHT refund procedures, the two most important matters for EU citizens include the permanent double taxation suffered cited by 91.29% of respondents (1425 of 1555) and giving up the right of submitting WHT refund claims by 74.76% of respondents (1164 of 1555). On the other hand, the two issues most often ranked as ‘high importance’ by companies/business organisations and business associations are the delays in effectively receiving the excessive WHT refund, cited by 89.80% of respondents (88 of 98), and the high compliance costs associated with the WHT refund process cited by 83.67% of respondents (82 of 98).

Figure 27: Consequences of the problems encountered with WHT refund procedures – all respondents

58.7% of respondents (969 of 1651) regularly invest in securities in EU countries that are not their home country, 24.7% (407 of 1651) do so occasionally, while 15.0% (248 of 1651) have never done so. Those who do not invest outside their home country indicated wanting to avoid double taxation and inefficient WHT procedures as main reasons.

Figure 28: Investing in securities in an EU country that is different from the home country – all respondents

Additionally, 62.3% (857 of 1376) of the respondents who invest in securities in EU countries outside their home country regularly encounter problems with the refund of the excess tax withheld when the rate of the applicable Double Taxation Convention is levied. 50.8% (651 of 1282) of EU citizens were never able to get a refund on the excessive tax withheld, whereas 75.9% (41 of 54) of companies/business organisations and business associations were able to receive a refund in some cases.

Figure 29: Receive excessive tax withheld back –
EU citizens (left) vs. companies/business organisations & business associations (right)

The countries with which respondents have regularly encountered problems with are France with 51.6% of respondents (867 of 1680) and Germany with 39.3% (661 of 1680), while the Netherlands has posed the least problems according to 16.6% (279 of 1680) of respondents.

More than 50% (of 1555) EU citizens answered with “don’t know” or did not give an answer on the questions regarding the time or monetary costs for claiming refunds. For 56.4% (31 of 55) of companies/business organisations and business associations, the period of time to receive a refund was longer than 2 years. Additionally, 83.3% of this stakeholder group (45 of 54) reported that collecting the required documentation for one refund claim takes more than three weeks with the monetary cost of the refund procedure being a medium percentage of the refund amount for 33.3% (18 of 54) of them.

Need for EU Action

94.1% of contributors (1583 of 1675) strongly support the need for EU action to make WHT procedures more efficient; with 94.3% (1576 of 1672) responding that action at EU level would add high value as an EU-wide framework would be in place.

Figure 30: Need for EU action – all respondents

Policy options

From an investor’s point of view, 83.9% (1295 of 1544) of EU citizens have a preference for a harmonised relief at source system. Many companies/business organisations and business associations share this view with EU citizens; 42.2% (35 of 83) of this stakeholder group thought that a harmonised relief at source system should be accompanied with a harmonised refund system.

The three most preferred measures to streamline WHT refund procedures are the creation of a single web portal for refund claims, cited by 72.17% of respondents (1193 of 1653), standardised and same language forms by 55.78% of respondents (922 of 1653), and the ability to e-request tax residence certificates along with a digitalized verification system by 52.57% of respondents (869 of 1653).

Under a potential EU relief at source system, 61.34% (1022 of 1666) of respondents think that dividends, interest, royalties, and other passive income payments should be covered. Additionally, 81.6% (1297 of 1589) of respondents think that there should be a fully-fledged relief at source system (covering both low and high-risk payments).

In the context of a relief at source system, 44.4% (24 of 54) of companies/business organisations and business associations consider a joint limit of a minimum WHT rate and a maximum amount of payment as a low-risk payment. While 29.6% (16 of 54) of this stakeholder group consider payments as “low-risk” if the applicable WHT rate is above 15%. EU citizens had mixed opinions on this question and their responses did not reveal a clear preference. 65.6% (1066 of 1666) of respondents believe that investors from EU and non-EU countries should benefit from a potential relief at source system. 74.7% (1194 of 1666) of respondents think that both EU and non-EU financial intermediaries should be obliged to report relevant information on the correct WHT rate to be levied on the dividend payment (or other passive income payments) to withholding agents.

In terms of enhancing the existing administrative cooperation framework, 63.8% (952 of 1492) of EU citizens strongly agree that it would be appropriate to broaden the Directive on administrative cooperation (DAC) framework in the EU, compared to 23.3% (21 of 90) of companies/business organisations and business associations. Overall, the majority of respondents agree to some extent. Of those who agreed that the administrative cooperation framework within the EU should be broadened, 63.9% (821 of 1284) are of the opinion that this should be done in combination with other measures (i.e. a harmonised relief at source system).

Regarding the country in which the financial intermediary closest to the investor should report relevant information, 86.85% (1314 of 1513) of EU citizens are of the view it should be the residence country of the investor. 29.47% (28 of 95) of companies/business organisations and business associations share this view, while 45.26% (43 of 95) of this stakeholder group believe that information should be reported to the source country of the investment. Additionally, 71.8% (1129 of 1573) of respondents agree that identification information, treaty residence status of the beneficial owners of the income paid, and the nature and amount of income earned by those investors should be reported to ensure tax treaty benefits entitlement. 2.0% (32 of 1573) of respondents disagreed with this approach while 26.2% (412 of 1573) answered with “don’t know.”

53 position papers were submitted to this public consultation, 2 of which have been deemed invalid and therefore have been discarded, leaving 51 valid position papers remaining. The position papers especially stressed the importance of harmonising and digitising WHT procedures along with implementing standardised forms to ease the filing and processing of refund claims.

3.3 Meetings with Member States

3.3.1. WPIV of 30 September 2022

The meeting was set up with Member States to discuss key aspects of the Commission initiative: the EU Digital tax residence certificate (eTRC), and the Quick refund system (QRS). A presentation was also given by Finland on the results of the first year of its relief at source system.

The Commission introduced the results of the survey sent to Member States on the eTRC and their current tax residency procedures. The eTRC proposal was broadly supported by Member States, however additional work would be required by them to assess each option further as well as ensuring compatibility with their existing TRC systems. In terms of information contained on the eTRC there was agreement that certain information should be mandatory but that the form should also have optional fields which could be used by Member States for their specific requirements, including their own residence criteria. For the validity period, the need for a revocation mechanism should also be included in the system.

Member States raised a wide range of issues on the QRS proposed. The scope of the proposal (portfolio versus other types of investor), the possibility of introducing risk management tools, the importance of clarifying liability rules (contractual agreements or a statutory basis) as well as the reporting obligations. Member States were concerned about control issues citing that the system should be robust against tax abuse like Cum/Ex and Cum/Cum frauds. No conclusive outcome on the QRS proposal was possible, however, the majority of Member States were willing to analyse its functioning.

Finland reported positive outcomes from their new relief at source system: 82% of dividend payments with lower rates were paid to individual investors while the quality of data and controls to prevent tax abuse had improved. Member States were interested on how Finland dealt with issues like the scope, the reporting obligations (use of a standardised xml reporting schema), the controls undertaken in their system to prevent tax abuse, and the liabilities of financial intermediaries.

3.3.2. WPIV of 17 March 2021

The OECD presented the TRACE relief at source system. In terms of liabilities, the country where the Authorised Intermediary (AI) administering the relief at source system is based, would be responsible for ensuring that the Intermediary is liable. However, the sanctions (including the revocation of the AI status) could have effect for all the governments involved. Further forms of control could be considered for the EU (e.g. joint audits). The OECD considers that the financial intermediary closest to the investor is best placed to check the validity of information submitted by the investor for treaty eligibility, including the self-declaration of tax residence, through the use of information obtained for AML/KYC purposes.

Finland gave a presentation on its new relief at source system of WHT which came into force as of 1 January 2021. The new system is based on the OECD TRACE system. Finland clarified that the system is limited to dividends paid on nominal shares by publicly listed companies to non-resident investors. Collective investment vehicles like transparent funds are excluded from the relief at source model, as the matter would involve changes to tax treaties, and not only changes to relief at source procedures.

The biggest change compared to the previous relief at source system in Finland, is that under the new system information on the Beneficial Owner (BO) is passed on from AIs to the Finnish tax administration, while before it was the responsibility of the paying company to apply the correct level of taxation (if allowed). Hence, under the new system the AI is responsible for the accuracy of the information provided. Finland relies on a strict liability approach for AI.

In terms of resources, as the system has been recently implemented additional resources were required but Finland expects that the workload for the administration will decrease. In terms of liability, the Finnish tax administration can audit the AI and risk analysis is used for audit selection purposes. Tax collection is always primarily targeted at the Authorised Intermediary in those cases of under withholding where the AI has not used reasonable measures to determine the beneficial owner of the dividend.

Germany presented its new WHT refund system. Given the Cum/Cum and Cum/Ex scandals, Germany wanted to ensure that the tax administration had full control over WHT refunds and that the proceed is centralized in Germany. The main elements of the pending reform, currently under discussion in the Parliament, are: 1 - procedure to become fully centralized; 2 - all steps to be digitalized (although this will take some time, until 2024); 3 - the refund system will remain the rule and tax administration will remain the only one in charge of checking the entitlements; 4 - investors and operators will be required to provide additional information justifying their claims with a considerable increase of the administrative burden. On the other side, the digitalized procedure will ease the burden.

The Commission presented the outcome of the questionnaire sent in advance of the meeting for the collection of quantitative and qualitative information on the current situation regarding withholding tax procedures and position of Member States on how to improve the current system. 20 Member States provided replies. No conclusive outcome can be observed from the overview of the replies. According to the 20 replies to the questionnaire received so far, the majority of countries rely on a combined relief and refund WHT system. Another conclusion drawn from the feedback received is that most of respondents have ongoing developments (unilateral domestic measures) on WHT procedures or foresee them for next years (mainly in the area of digitalisation and enhancing an effective relief at source system). Finally, the possibility of setting up an EU relief at source system mirroring TRACE, supporting the work strand of making simpler and more efficient WHT refund procedures and reinforcing the administrative cooperation framework was mentioned by respondents among the potential optimal solutions for tackling inefficiencies of WHT procedures in the short and medium term.

3.3.3. WPIV of 13 November 2020

The Commission presented to Member States the CMU Action Plan. The three core objectives underlying the CMU Action Plan were highlighted: (i) making financing more accessible to European companies; (ii) making the EU an even safer place for individuals to save and invest long-term; and (iii) integrating national capital markets into a genuine Single Market. In the furtherance of these objectives, it was proposed to simplify and standardize WHT procedures by putting forward a relief at source system as these are divergent across member states, lengthy, costly and prone to fraud. The aim of the WHT initiative is not to harmonize rules in the EU, including residency, rather it is to simplify existing WHT procedures in the EU.

ESMA presented its final report on the Cum/Cum and Cum/Ex scandals. Their main conclusion is that addressing these abuses should primarily lie with the tax authorities of Member States, for example through cooperation between tax authorities and financial authorities in the EU. A Member State was concerned that a relief at source system would not address the Cum/Cum and Cum/Ex frauds, in particular for enforceability of EU rules on non-EU Intermediaries, ESMA concurred that these were indeed issues to be addressed, and noted that the final report focusses on Cum/Ex and not Cum/Cum schemes like dividend arbitrage.

The Commission’s Joint Research Centre (JRC) presented the methodology used to analyse the costs of inefficient WHT procedures. The methodology was set out in the Economic Impact of the Commission Recommendation on WHT Relief Procedures (2009). According to an update of this study in 2016, direct costs of inefficient WHT procedures are around 8.4 million euros per year. These costs include three specific sources related to WHT relief procedures: (i) compliance costs; (ii) forgone relief (i.e unclaimed tax refunds); and (iii) opportunity costs due to delayed claims and payments of tax refunds. Further, it was indicated that an update to this study was underway for the new initiative. Methodological improvements on the previous approach are based on the use of the Computable General Equilibrium model CORTAX. This will allow to capture behavioural responses of relevant economic agents such as firms and governments as well as general equilibrium effects of tax policies on economic activity like GDP and cross-border financial investment flows.

On the way forward, the Commission presented Action Plan for fair and simple taxation supporting the recovery strategy (the Action Plan) published in July 2020. The Action Plan proposes to introduce a common, standardised EU-wide system for WHT relief at source coupled with a new exchange of information and cooperation mechanism between tax administrations and financial competent authorities. To deliver these goals, OECD TRACE initiative will be taken into account and both legislative and non-legislative measures are envisaged. The Commission’s preferred approach consists of a system that allows a smooth and correct application of tax treaty benefits, through a relief at source as a primary mechanism, and a digitalized/standardised refund system as back-up process.

A few Member States declared lack of support for the implementation a fully-fledged relief at source system as they consider still premature to explore such a solution. However, they were not against the idea of standardizing and digitalizing WHT procedures. Member States indicated that the following issues should be taken into account in the initiative: the lack of harmonised interpretation of certain rules such as tax residency across the EU; the need to coordinate the EU initiative with the work of the OECD; an accurate delineation of the scope of the initiative; the liability of intermediaries should be defined; triangular cases where the authorized intermediaries are not resident in the same country as the investor (including outside of the EU) should be addressed; appropriate IT systems to be in place to handle the proposed system; and concerns that removing WHT refund system in place of relief at source might increase tax abuses.

3.3.4. Bilateral meetings with Member States

Several bilateral meetings took place with Member States in 2021-2022 to understand their current WHT claim systems and the issues that are important for them in respect of possible reforms.

Further, for the preparation of the Impact Assessment, in depth bilateral meetings were held with Member States where withholding tax refund systems appear to play an important role ( LXXXIII ). The following issues were discussed in these meetings: the digital tax residence certificate, enhanced transparency via increased common reporting, and a proposal for a quick refund procedure.

3.4 Meetings with stakeholders

Meeting with business stakeholders on 28 September ( LXXXIV )

A meeting was held with key stakeholders to discuss the main elements of the Commission’s WHT initiative, in particular the main elements the Quick refund system aims to introduce to improve the current framework on WHT procedures. In addition, a presentation was given by Finland on the results of their new relief at source system. The OECD also participated in the meeting, sharing their experience of the TRACE system.

Stakeholders raised the following issues which they considered required further development:

·Confidentiality issues in relation to the system proposed: Preserve clients´ data from competitors is one of the main concerns of the industry when setting up a system departing from TRACE.

·External dimension of the initiative: important for the industry that non-EU located financial intermediaries can participate in the new system.

·Clear preference for an investor self-declaration instead of tax residence certificate. It has the advantage of involving an active statement from the investor himself.

·Identification of beneficial ownership: issue to be addressed in a further stage of the initiative.

·Reporting obligations seem burdensome, preference for TRACE-like information transmission method on pool basis to the WHT agent and directly to the source country on yearly basis.

·In favour of introducing the relief at source system at least for low-risk payments (threshold below 15% WHT rate).

Further, the stakeholders had further questions on the compulsory/voluntary nature of the Quick refund system, the elements to be included in the due diligence system, and the timeline for the adoption of the proposal by the Commission.

The Commission requested participants´ support to assess the economic costs this new system would imply for the financial sector.

Bilateral meeting with stakeholders

In order to gather the views of financial intermediaries several meetings were held with associations representing intermediaries involved in the WHT reclaim systems in the EU and with other interested stakeholders (such as, consulting firms, international organizations, etc.) and also to inform the Commission about WHT systems in the EU. Further, the Commission stressed the need to gather data for the Impact Assessment include costs and benefits of recent WHT reforms of Member States, and key industry data, for example, on type of investors (retail/companies/institutional investors), and direct/indirect holdings of portfolio investments.

EFAMA gave an overview of the asset management industry. We touch upon their main concerns such as the lack of certainty about the entitlement to DTT benefits of investment funds and paper-based/cumbersome WHT refund procedures. Blockchain was pointed out as a technology that could help on WHT procedures (EFAMA focused mainly in blockchain as way forward to solve the issue of transparency within the financial chain and the digital TRC). Further, it could also provide visibility in spot transactions to help identify dividend arbitrage and fraud. For entitlement to treaties, especially relevant for transparent funds, a matrix could be developed to properly address CIVs entitlement to DTT benefits.

AGC ( LXXXV ) outlined how the US QI relief at source system operated. AGC welcomed a Quick refund system. Some elements should be clearly defined for AGC to have a final position: liability allocation system and information requirements feasibility. AGC requested that any system the Commission comes up with regarding WHT should be easily replicable by third countries (OECD level). Compatibility at international level is key point for them.

EUROCLEAR made a presentation stating how Member States deal with WHT procedures in terms of actors involved and systems in place. For 15 Member States Euroclear provides with specific information about the WHT system in place (pro rata temporis vs. record date holding) and who is acting as WHT agent (issuer or professional WHT agent). The main conclusions drawn from the meeting were: most countries rely on record date holding system (where the entitlement to receive the dividends is given by the settlement date) and most countries rely on professional WHT actors that are in charge of all operations regarding dividends payment (via agreements or registers). In those countries which keep relying on securities issuers, most of those issuers outsource WHT tasks on a paying agent. Therefore, it is an extra burden for them to assume the role imposed by the Quick refund system but proportionate to the benefit the harmonisation of the procedure might bring to the investors and ultimately to them (since the new system aims to attract more investment to those EU companies).

ESMA explained in detail their ‘Final report On Cum/Ex, Cum/Cum and withholding tax reclaim schemes’ released on 23 September. On the reference to TRACE made in the report (ESMA has been in contact with OECD), ESMA confirms it is conceived as a preventive tool to avoid Cum/Ex fraud from happening. Even though it does not completely address the problem, it definitely is a step ahead in fighting against it. However, he highlighted that assessing the usefulness of TRACE’s implementation is out of ESMA’s remit. On the use of Central Security Depository’s (CSD) data in the detection of WHT reclaim schemes, they concluded that it is an interesting tool to explore but not the definitive solution. The reasons are that the information that could be retrieved might not be used in a systematic way in order to build a surveillance tool and that information about beneficial owner is not always available. The option of converting WHT refund rights into a kind of new shares to be traded in the securities market was disregarded as recommendation by ESMA as it was not explored further and they didn’t have the feeling of being an effective solution. About the recommendation on enhancing the exchange of info among financial and tax authorities, it was confirmed as the main way forward.

On a follow-up meeting, ESMA pointed out they are in regular contact with FISMA´s unit in charge of the initiative regarding lifting the restrictions for being able to exchange information among financial and tax authorities. . They made a clear explanation of the reporting frameworks ESMA has in place to help MS detecting any market abuse. The main reporting scheme is the one based on transaction-by-transaction recording reported by each investment firm on daily basis and close to real time (the second following day) to the national financial supervisory (NCA) with high level of granularity (place, timing, parts involved, beneficiaries, amount, etc). Afterwards, this huge amount of info is exchanged among NCA within the EU. ESMA´s role in the procedure is to act as supervisor that the system is properly working, as facilitator for a successful exchange of information among NCA.

EBF signalled that there have been several changes in last years that might make easier relief at source initiative to succeed, such as, quite advance tax transparency agenda, leveraging digitalisation as one of the priorities at Commission level, COVID crisis make more urgent than ever put an end to paper-based procedures, Cum/Ex scandals make more evident the need of another WHT system and urgent need of harmonisation/standardization at WHT procedures level according to the industry.

OECD, in particular we met with persons involved in the TRACE IP initiative, shared with us as well some technical ideas:

1.Risk management might be an element to introduce in a potential relief at source system

2.WHT agent doesn´t need to know the identification data about the beneficial owner of the investor (final investor).

3.Some commercial reasons are behind the fact that information flow through the financial chain in TRACE system is done in pooled basis.

4.Liability regulation is a key point to take into account for the success of a relief at source system

5.Digital solution such as eTRC would give comfort to MS and transparency to the system but would not be an actual game changer

AFME shared its main concerns regarding potential harmonisation of WHT systems, definition of beneficial owner, certainty on the regulation of the liability of FI, clear due diligence and potential synergies with Common Standard Reporting.

Stakeholder targeted surveys

In October 2022, the following targeted surveys were launched:

1)Survey of business stakeholders on the impact of costs on the reporting obligation sub-options (i) to (iv) under the initiative

2)Survey of Authorised Intermediaries on the costs and benefits of the new Finnish relief at source system ( LXXXVI ) 

For 1) This survey was intended to measure costs for financial intermediaries of the potential reporting obligation options under the WHT initiative in comparison to costs of their existing reporting systems. 7 replies from business stakeholders were received.

The first section of the Survey asked the financial intermediaries what and how information to be collected for the relief/refund processes under the proposal would compare to the costs which they already collect for the same processes. 3 respondents provided current costs but only 1 respondent could provide both current and estimated costs for the new refund system. This financial intermediary, operating in 12 Member States, replied that digitalisation and streamlining of WHT reclaim procedures would result in cost savings. This estimate was based on the following assumptions:

1)Cost estimates are based on the model that would be very similar to the automated model in Netherlands.

2)Fixed cost is done on the assumption that each country would adopt the Directive in their way and as such there would be as many variations as there are countries, based on its past experience. This estimate is done on the basis of implementation in 20 countries.

Based on shares of the financial intermediary in key portfolio securities markets in the EU, an extrapolation for the EU was made. For the streamlined standard refund procedures, the new proposal is estimated to decrease the costs of submitting a WHT refund. The respondent noted that the incremental recurrent refund costs would decrease (EUR 2.5 million) which would result in a decrease of incremental refund costs of EUR 13.5 million for financial intermediaries in the EU. Development costs to implement the new refund system is estimated to be EUR 4 million by the respondent, resulting in total development costs for financial intermediaries in the EU of EUR 21.7 million.

The second section of the Survey asked respondents to estimate costs of reporting information to their tax authority The same financial intermediary estimated an increase of 200% of current reporting costs to the tax authorities (that is EUR 2.4 million), due to the new reporting obligations of the proposal. Based on this figure, the annual incremental recurrent reporting costs are estimated to increase by EUR 13 million for financial intermediaries in the EU. Based on the development costs for implementation of the reporting system by the respondent (EUR 10 million), it is estimated that total development costs for financial intermediaries in the EU will be EUR 54.2 million.

The third and final section of the Survey asked how costs for the WHT agent/securities issuer would change in the source Member State if the WHT agent would perform the WHT refund and verification process. No respondent could provide current and estimates costs for this section. However, one respondent noted that while there would need to be implementation costs, recurring costs would be similar to the current systems.

Although the respondent has significant market shares in key EU markets, it cannot be assumed to be representative of the portfolio cross-border security market in the EU and can only provide an indication of the incremental changes in costs. The number of financial intermediaries in the final chain, and indeed the size of the financial intermediary, would also need to be taken into account to arrive at more robust conclusions. However, this data is not available. Nevertheless, the qualitative responses of other respondents of the survey confirmed the trends in the incremental changes quantified.

From the replies it appears that financial intermediaries in the EU only report data to tax authorities on client investors if the intermediary operates in the same Member State as the investor, or the financial intermediary has withholding tax obligations in that Member State. Reporting on a cross-border basis by a financial intermediary (albeit directly or via the tax authority of the intermediary) to a source Member State tax authority rarely takes place and only for a limited number of relief at source systems (Ireland and Finland were noted). The absence of these cross-border reporting obligations was also confirmed in the bilateral discussions with Member States.

For 2), relating to the new Finnish relief at source system, the Commission received 6 replies, from which 3 respondents ( LXXXVII ) could provide both the previous and current costs of operating the new Finnish relief at source system. These 3 respondents noted that recurring costs had increased by 61% and incurred implementation costs of EUR 1.6 million. In terms of benefits, respondents had noted that risk for intermediaries had declined, and that customer satisfaction had increased. However, some respondents noted that it was too early to draw a robust cost-benefit analysis as the new system had only been implemented in 2021.

3.5. TADEUS meetings (Tax Administrations European Union Summit)

3.5.1. 5 May 2022

During this meeting the main elements of the WHT initiative were presented to the heads of tax administrations. The main elements outlined during the meeting were: Creation of an EU digital tax residency certificate and/or exchangeable directly among tax administrations; Setting up a common EU standard of information sharing throughout the chain of financial intermediaries about the identity of the beneficial owner; Improve the exchange of information between tax administrations in order to allow them to check ex-post the correctness of the tax withheld; And establishment of an EU common due diligence standard for financial intermediaries to check beneficial owner’s data and/or creating an EU wide registry of accepted financial intermediaries. To wrap-up the meeting the delegates of Member States were encouraged to participate on the, at that moment, ongoing public consultation on the WHT initiative and were informed about next steps of the initiative.

3.5.2. 6 May 2020

It represented a good opportunity to raise awareness among tax administrations’ heads about the magnitude of the problem regarding WHT procedures across the EU and how it entails a barrier to the free movement of capitals, thus, with negative consequences over the Single Market.

3.6. High Level Working Party on Taxation (HLWP)

Held on Toulouse the 5 March 2022, during this meeting with high level representatives designated by Member States, the willing to standardize WHT procedures across the EU was presented. In particular, the issue at stake was explained as well as the background of the initiative, then the potential solutions were outlined stating that they would imply greater digitalisation and transparency for all stakeholders. Finally, participants were updated on the state of play of the initiative and several questions were posed on Member States’ representative for better understanding of the Commission of their respective positions.

4.    Conclusions

There is large consensus among stakeholders about the problems arising with current different WHT procedures and on the need of EU action to tackle such a fragmented and inefficient scenario. However, there were noticeable differences between the main stakeholder groups on possible options to address WHT reclaim problems in the EU. In absence of harmonising WHT rates in the EU or abolishing withholding taxes altogether, investors regarded the relief at source system as the most optimal solution in the EU citing systems like the US QI system as best practice.

Business associations and financial intermediaries also considered that relief at source could provide the most optimal results but acknowledged that WHT refund systems would likely continue in future, at least as a back-up system, and as such the WHT initiative should look to standardise the current WHT refund processes and forms, in particular through the use of digitalisation tools, in order to reduce WHT refund delays.

Member State tax authorities indicated a broad range of views about the possibility of standardised WHT procedures across the EU. Overall, there are two group of Member States:

·Firstly, those where WHT procedures are not a problem either due to a low amount of refunds given the internal rate applied to non-resident investors or either because they already have in place a well-functioning system such as relief at source or quick refund; these countries, in general, indicated the importance of ensuring that any solution should complement their current WHT reclaim processes without making them step back on what they already have in place.

·Secondly, there are several Member States that are aware of some existing inefficiencies related to their current WHT procedures and are open to hearing how possible reforms could improve their current systems. These Member States have, nonetheless, raised the importance of striking a balance between making efficient WHT procedures and keep the control over processes to prevent previous recent tax abuses like Cum/Cum and Cum/Ex.

Annex 3: Who is affected and how?

Practical implications of the initiative: The initiative will impact portfolio investors, financial intermediaries and tax administrations directly and Member States indirectly by attracting more investment to the companies located in the EU and ultimately impacting Member States’ GDP.

1)Portfolio investors will be impacted in the sense they will have easier and more direct access to tax treaty benefits and, in particular, to reduced tax rates on the WHT levied on cross-border passive income payments. 

2)Financial intermediaries will be impacted since they will not need to interact with diverse rules in the EU as a Quick refund system or relief at source will function with a standardised procedure. Further, financial intermediaries will be impacted with regard to their reporting obligations to the tax authorities in the source Member State as information reporting methods will be streamlined and standardised in the EU. Although reporting obligations are already in place between withholding tax agents in the Source Member State and their tax authorities, all intermediaries in the financial chain would need to report to tax authorities in the Source Member States. Further, the information to be reported to the tax authority would need to include, more granular information on a transaction basis in order to counter tax abuse like Cum/Cum and Cum/Ex. 

3)Tax administrations will need to adapt their internal regulation to embrace the changes envisaged under the new WHT system. In particular, tax administrations will ensure that they have available the technology needed to allow eTRC to be requested and issued online and external eTRCs to be properly verified. Reporting systems would need to be adapted by Member States to take into account the new reporting obligations of financial intermediaries under the refund or relief proposal

4)Member States will be impacted by making their capital market more attractive to investment and, therefore, by corresponding macroeconomic effects over GDP and wages. Despite the wider economic benefits, tax revenues for Member States will be expected to decrease as the more efficient WHT refund/relief system means that the foregone WHT refund amounts repaid to investors will decrease.

1)Summary of costs and benefits of the preferred option

2)Relevant sustainable development goals

Annex 4: Analytical methods

In order to support chapters 5 and 6 of this Impact Assessment a number of studies were provided by the JRC to measure the costs incurred by investors of the current WHT reclaim systems in the EU as well as the potential economic benefits of a WHT reclaim reform in the EU.

JRC has provided us along the years with different studies in relation to the estimates of the WHT costs and the dynamic effect of potentially removing all WHT costs. The relevant economic studies for this Impact Assessment are:

In relation to WHT costs:

a.In 2009 the Commission put forward a Recommendation to the Member States on the simplification of withholding tax relief procedures that was accompanied by an impact assessment (IA). The economic analysis in the IA was prepared by the JRC (first study on cost of WHT relief procedures, where the costs amount to EUR 8,4 bn). This study can be consulted online in the following link: 10-2009_STUDY.doc (europa.eu)

b.In 2016, the JRC produced updated estimates of the costs associated with the procedures for withholding tax relief, following the methodology adopted in the 2009 IA (second study on costs of WHT relief procedures, where the costs amount equally to EUR 8,4 bn but with different breakdown among components). A summary of this study is compiled in this Annex 4.

c.In 2021, after the publication of the Action plan for fair and simple taxation supporting the recovery strategy, JRC was requested again to update the figures on WHT relief procedures costs (third study on costs of WHT relief procedures, where the costs amount to EUR 5,52 bn). Attached to Annex 5 of the current IA.

d.In 2022, given the changing economic scenario regarding interest rates, JRC updated again its study (fourth study on WHT costs amounting to EUR 6,62 bn). A summary of this study is compiled in this Annex 4

In relation to the dynamic impacts:

e.Empirical econometric model to test sensibility of foreign portfolio investment to non-refundable WHT. Compiled in Annex 6

f.Macroeconomic impact of reducing the cost of WHT in terms of GDP, wages, capital, etc. Compiled in Annex 6

In this Annex 4 we give the reader an overview of all above mentioned JRC studies:

1)Quantifying the costs of withholding tax procedures for investors in the EU

For the purposes of this Impact Assessment, an estimation was made by the Commission services of the various costs that are incurred by investors regarding their current withholding tax reclaim procedures based on the methodology used in the economic impact SWD attached to 2009 Recommendation on WHT LXXXVIII . Such estimation represents current costs according to baseline scenario.

Using cross-border securities holdings in the IMF ‘Coordinated Portfolio Investment Survey’ (CPIS), the methodology for the estimation includes three different types of costs:

·The opportunity costs due to delayed claims and payments of tax refunds: it is estimated as the average delay period times the risk-free rate times the total amount of tax relief (i.e., the theoretical entitlements). An average delay of one year and a risk-free rate of 0.6% is assumed, applied to an estimated average rate of withholding tax rate relief of 7.4%;

·The foregone tax relief is attributable to retail investors ( LXXXIX ) who do not claim their tax refunds and is assumed to amount to 10% of the total cross-border investment. This amount is derived from the CPIS sectoral data for retail investors (NHN sectors);

·The actual costs related to the reclaim procedures are assumed to account for an average of 2% of the refundable amounts. The calibration of the parameter is the same as in the 2009 economic impact SWD. This third factor is the actual amount of occurring costs related to reclaim procedures (paperwork etc.). The cost drivers relate to intermediaries providing the paperwork and diversity of source country requirements, as well as costs of the source Member State of investment for verifying investor refund claims.

Table [1] below, using a reference period of 2008-19 CPIS data, indicates the total cost estimates broken down for both within-EU and total cross-border investments, including both EU and non-EU cross-border investors. The figure for total cross-border investments includes both investors from EU and non-EU countries.

Table [1]. Estimated cost of WHT relief procedures – breakdown by cost driver (EUR bn)

Source: JRC calculations.

The earlier estimation from 2016 using the same methodology amounted to total costs of approximately EUR 8.6 billion. Although the amount of cross-border security holdings has actually increased between both reference periods, the decrease in costs between the reference periods 2003-14 and 2008-19 can be attributed to the following factors:

·the return rate on investments has significantly dropped in recent years (in particular the interest rate).

·the risk-free interest rate has dramatically decreased in recent years, which explains the significant reduction in opportunity costs (opportunity costs for not having the money back yet due to an ongoing WHT refund procedure).

·In addition, the differences between the national rates and the tax treaty rates have been reduced from 2014 to 2019 in some Member States, which reduces the impact of inefficient WHT procedures.

These scenarios assume of the introduction of a relief at source system, and the assumption that the need for refunds will be eliminated. In that sense it can be argued that these estimations are higher end estimates. The extent to which this assumption is valid is assessed in chapter 6 of this Impact Assessment. At the same time, given the data limitations, a number of assumptions had to be made which are likely to underestimate the actual costs of the current WHT procedures. In particular, the estimations are based on the CPIS Survey, which captures direct holdings for retail investors but not indirect holdings ( XC ). Indirect security holdings are significant in the EU with investors facing the same refund issues as direct holding investors in terms of refunds. These estimations have not been able to be precisely quantified for this Impact Assessment. However, some approximate estimations are available. Data from the ECB indicates that approx. 31% of investment funds in the EU are indirectly held by retail investors ( XCI ). EFAMA indicates that retails investors account for 19% of fund sales in the EU compared to the 10% estimation used for the JRC data ( XCII ). Therefore, the estimated costs of EUR 4.9 billion are likely to be substantially understated.

Furthermore, the current increase of interest rates and among them the free-risk interest rate would mean an increase of the costs of the base line option if this increase were to be maintained in the future.

The JRC provided also an estimate taking into account a long-term free interest rate over the period 2000-2019 of 1.7%, instead of 0.6%, used for the risk-free interest rate in the 2021 study. With such an increase of interest rate, total costs remain quite similar at EUR 6.62 bn with only an increase of 19.7%.

Estimated cost of WHT relief procedures using long-run interest rates

– breakdown by cost driver (EUR bn)

Source: JRC calculations.

In addition to the effects of costs of WHT relief procedures, the simulation also estimated what effects the procedures would have on GDP in the EU using a Cobb-Douglas function. Table [2] below indicates that for all investment in the EU, the costs of WHT relief procedures represent a cost amounting to 0.031% of EU GDP.

Table [2] Impacts on GDP of eliminating the cost of WHT relief procedures

Source: JRC calculations.

2)The effect of non-refundable withholding taxes on FPI and macro-economic indicators

Further, the following economic simulations were prepared by the Commission services for the purpose of this Impact Assessment:

üAn empirical econometric model to test the sensitivity of foreign portfolio investment (FPI) to non-refundable withholding taxes;

üThe macro-economic impact of reducing the cost of withholding tax relief procedures making use of the CGE model CORTAX ( XCIII ).

1)Sensitivity of foreign portfolio investment to non-refundable withholding taxes

Table [2] above indicates costs arising from the current refund withholding tax procedures from the perspective of the investors. From a macro-economic perspective, opportunity costs also arise as inadequate refund processes hold back full integration of the securities markets in the EU and thus hinder their functioning as the main channel of cross-border private risk sharing for firms and households.

As with a) above, this simulation also uses the IMF CPIS Survey as input data but further analyses the data by holding sector (that is financial corporations, non-financial corporations, and households etc.) and by type of security (that is debt and equity). Using a gravity-style model the simulation aims to infer the effects of non-refundable withholding taxes on FPI (foreign portfolio investment) using a panel of 83 countries, including all Member States, for the period between 2005 and 2019.

For all FPI stock of equity and debt holdings (that is not sector specific), the simulation found that a 10pp reduction in non-refundable withholding tax increases FPI stock of equity holdings by between 5.2% and 9.3% (point estimate 8.2%). The corresponding effect on FPI stock of debt holdings is slightly lower and less precise, estimated between 2.2% and 8.1% (point estimate 6.2%). Household sector data is available under the CPIS Survey which could potentially help to identify the specific impact on retail investors. However, data used for the sector analysis had a much smaller sample size which precludes sufficient within country-pair variations over time to infer the impact of non-refundable withholding taxes on FPI stock holdings. A second more fundamental issue with using CPIS household sector data is that it does not include indirect holding of securities to retail investors through funds, as discussed above, therefore both limitations mean that the total stock of FPI data should be used instead of sector specific FPI data.

2)Macro-economic impact of reducing the cost of withholding tax relief procedures making use of the CGE model CORTAX

The Commission economic model CORTAX which analysis tax-benefit policies of Member States was used to assess the effects of removing tax overpayments (defined as non-refundable withholding taxes). The assessment is made on the impact on GDP, investment, employment and tax revenues. Using data from all Member States, the effect on 4 macro-economic indicators is estimated: EU27 GDP, capital, wages and employment. As indicated in Table [3] below, EU27 GDP is projected to increase by 0.047%, capital/wages would rise, respectively, by 0.137% and 0.048%, and employment would increase marginally too as compared to the current status quo. Tax revenues are expected to decrease because even though increase investment and employment will lead to higher tax revenues, the correct refunds of WHT by tax authorities will lead to an overall decrease of 0.039% of tax revenues.

Table [3]. Macro-economic impact on removing tax overpayment

Source: JRC calculations.

For robustness purposes, given that they use different methodologies, the change of 0.047% in GDP according to this simulation closely matches the 0.031% increase in GDP estimated in simulation in Section 6.1.4.4 Figure [13] above.

These scenarios in (i) and (ii) assume a relief at source system being put in place which implies that the need for refunds will be eliminated. Furthermore, given that Foreign Portfolio Investment (FPI) is not explicitly modelled in CORTAX, in this exercise it is assumed that FPI and FDI (foreign direct investment) are of comparable size in each source country. However, FPI levels are typically smaller than FDI and reflect large cross-country heterogeneity. To the extent that countries receive less FPI than FDI, the national effects are likely to be smaller than simulated here. Conversely, the effects would be larger in those cases where FPI is greater than FDI.

Annex 5: 2021 JRC Study: Quantifying the Costs of Withholding Tax Relief Procedures

25 February 2021

Quantifying the Costs of Withholding Tax Relief Procedures

Authors: JRC

Introduction and policy background

Procedures for granting withholding tax relief on cross-border investment are a major and long-standing barrier to the free movement of capital that hinders the good-functioning of the Single Market. The Commission has been trying to tackle this issue for a long time, both from a taxation perspective and also in the context of the Capital Markets Union (CMU). After the so-called Giovannini reports of the early 2000s, in 2009, the Commission put forward a Recommendation to the Member States on the simplification of withholding tax relief procedures that was accompanied by an impact assessment (IA) ( XCIV ). The economic analysis in the IA was prepared by the JRC.

In 2016, upon request by DG FISMA, the JRC produced updated estimates of the costs associated with the procedures for withholding tax relief, following the methodology adopted in the 2009 IA ( XCV ). Given its relevance for the CMU, the Commission (DG FISMA, together with DG TAXUD) was tackling the issue again under objective 6 of the CMU action plan "remove cross-border tax barriers" ( XCVI ). The results were presented during the 2nd Meeting of the Expert Group on Barriers to Free Movement of Capital in the context of the Capital Markets Union, held in Brussels on 26 January 2016. Recently, the Commission published an Action plan for fair and simple taxation supporting the recovery strategy, which includes an initiative for introducing a common, standardised, EU-wide system for withholding tax relief at source ( XCVII ).

Against this policy background, this study provides a methodological discussion and an update of the quantification of the economic impacts presented in chapters 2 and 6 of The Economic Impact of the Commission Recommendation on Withholding Tax Relief Procedures and the FISCO Proposals (see https://ec.europa.eu/info/sites/info/files/booklet-fisco-09022010_en.pdf , henceforth referred to as the IA).

While the overall methodology used is the same, relevant changes in data and parameterization to obtain the results are spelled out in the following. In particular, the starting point of the analysis is cross-border holdings of equity and debt and all the bilateral withholding tax rates applicable in the EU. Then, the study simulates different scenarios of reduced withholding tax barriers on changes in the cost of capital and ultimately the aggregate effects on output. As a general caveat, it should be borne in mind that the whole exercise is performed under certain assumptions on the incidence of the costs of relief procedures, because of significant data limitations. Moreover, it gauges the aggregate output effect using a very simplified and static framework, which does not capture all the macroeconomic spill overs and feedback loops at play in a fully-fledged macroeconomic model. Therefore, the results should be taken with caution. A more detailed critical discussion of the results and underlying assumption is provided in section 3.

1.    Basic facts

This section first describes the data for cross-border holdings of equity and debt securities. Next, we focus on the associated yields. Finally, the withholding tax rates in different countries are presented.

1.1. Cross-border holdings of equity and debt securities

Data on cross-border portfolio holdings are obtained from the IMF “Coordinated Portfolio Investment Survey” (CPIS). The time series is currently updated up to December 2019. The aggregate data for the EU shows that the amount of equity securities held by nondomestic investors in the European Union (EU27) was 10.7 trillion US$ in 2019, up from 6.9 tn in 2014 and 5.1 tn in 2006 ( XCVIII ) Debt securities amount to 11.5 trillion in 2019, the same value recorded for 2014, and up from 8.8 trillion in 2006. . Cross-border holdings of EU investors amount to 5.6 and 6.7 trillion US$ for equities and debt securities respectively in 2019 (see Table 2).

As is apparent from Figure 1, the upward trend for stocks of debt securities has reached a plateau in the years after the global financial crisis. By contrast, cross-border holding of equity securities, on a similarly strong upward trend before 2008, have been steadily increasing also afterwards, although much less markedly.

Table 1. Cross-border holdings of securities in 2019 (in millions of dollars)

Source: IMF CPIS

Figure 1 Cross-border portfolio investment in the EU

Source: IMF, Coordinated Portfolio Investment Survey

Figure 2 Within-EU cross-border portfolio investment

Source: IMF, Coordinated Portfolio Investment Survey

1.2. Yields

While holdings provide a picture of investors’ cross-border positions in the EU, the corresponding yields are informative on the associated returns to the investment. In turn, these define the cost of capital to issuers. The cost of capital consists of two components: (i) the cost of equity capital and (ii) the cost of debt. As for the cost of equity, we consider the dividend yield, which measures the realized return to the shareholder once earnings are paid out. The cost of debt is proxied using bond yields. Figure 3 plots the relevant yields. For bond yields, we plot government debt with 10y maturity. Excluding peaks around the outbreak of the global financial crisis, dividend yields are roughly stable over the 2001-2019 period, hovering around 3%. As expected, bond yields are clearly trending downward, and markedly so as from the sovereign debt crisis.

Figure 3 Debt and equity yields in the Euro area

1.3. Withholding taxes

Following the established methodology, we define the “actual” statutory withholding tax rate as the minimum between the domestic tax rate and the tax rate applicable under the relevant DTA (if in place). Hence, the actual withholding tax rate is a theoretical rate that accounts for the fact that if no DTA is available or the tax rate applied within the Member State (domestic rate) is lower than that in the DTA, the domestic rate is relevant for the foreign investor. All the tax data are obtained from the International Bureau of Fiscal Documentation (IBFD). Domestic rates within the EU are reported in table 2 ( XCIX ).

Taking into account the amount of cross-border security holdings within the EU, the weighted average domestic tax rates for interest income and dividends are 13.0% and 18.6% respectively. The corresponding weighted averages for the “actual” statutory withholding tax rates are 3.4% (interest) and 13.9% (dividends). Compared to the domestic rates this implies an average reduction of roughly 7.4 pp (combined for dividends and interest payments) for non-domestic portfolio investors.

Table 2. Domestic tax rates (in percent)

Source: IBFD

2.    Quantification of the costs associated to WHT relief procedures

This section describes the calculation of the potential costs associated to inefficient withholding tax relief procedures. It then quantifies the impacts on the cost of financial capital and, consequently, on aggregate output.

The costs related to the current system of WHT relief procedures depend on the total dividends and interest payments received by cross-border portfolio investors in the EU. The total amount of dividends received in 2019 can be proxied as the average dividend yield (3.0%) ( C ) times the amount of equity securities held by non-domestic investors. The total amount of interest payments received is estimated accordingly as the average interest rate (3.7%) ( CI ) times the amount of cross-border debt securities. We consider two alternative aggregates for the quantification: i) within-EU cross-border investment; ii) total cross-border investment in the EU. The latter aggregate comprises both within-EU cross-border investment and foreign (extra-EU) investment. Based on the observed stocks and yields, estimated within-EU cross-border payment flows amount to 417 billion US$ or 330 billion € (using an average 2008-2019 USD/€ exchange rate of 1.264). Similarly, total cross-border payments, including to extra-EU investors, are in the range of 746 billion US$ or 590 billion €.

As detailed in Box 1., three different cost sources can be identified, as follows:

1)The opportunity cost due to delayed claims and payments of tax refunds: it is estimated as the average delay period times the risk free rate times the total amount of tax relief (i.e., the theoretical entitlements). We assume an average delay of one year and a risk-free rate of 0.6% ( CII ), applied to the estimated average rate of withholding tax relief of 7.4%.

2)In line with the IA study, the foregone tax relief attributable to (small) investors who do not claim their tax refunds is assumed to amount to 10% of the total cross-border investment.

3)The actual costs related to the reclaim procedures are assumed to account for an average of 2% of the refundable amount. The calibration of the parameter is the same as in the IA study ( CIII ). 

Table 3 reports the total cost estimates, broken down by the different components, for both within-EU and total cross-border investments. The latter scenario implicitly assumes that the calculated gap for withholding tax rates applies also to investment from extra-EU countries.

Table 3. Estimated cost of WHT relief procedures – breakdown by cost driver (EUR bn)

Source: JRC calculations.

2.1 Sensitivity analysis

As Table 3 shows, the main cost driver from withholding tax relief procedures is the foregone tax relief stemming from the disincentive effects and other complexities linked to the existing relief procedures. In the baseline calculations, it is assumed that these affect 10% of relevant flows. The fraction of affected payments plays a crucial role in the estimation of the cost component. In Table 4 we show how the cost estimates (including for the total cost aggregate) evolve under different parameterization of this variable ( CIV ).

Table 4. Foregone tax relief (EUR bn) – sensitivity analysis for different values of payments for which claims are foregone

Source: JRC calculations.

2.2 Impacts on the cost of capital

Following the analysis carried out in the IA study, the potential change of the effective tax burden is calculated as the ratio of the total annual costs to the corresponding estimated total amount of dividends and interest payments. The gains from eliminating these costs translate into a reduction in the cross-border cost of financing of equal magnitude. However, one needs to take into account the fact that cross-border portfolio holdings constitute only a fraction of all financing flows. In this respect, in line with the IA study, we assume that the share of cross-border holdings is 25% for within-EU investment. This calibration is consistent with the available evidence ( CV ). When it comes to total cross-border investment to the EU, we use Eurostat data to calibrate the relevant parameter, set at 35% ( CVI ). 

This reduction in the cost of financing influences only that part of real investment that is financed via debt and new equity. If investment projects are financed by retained earnings, dividend taxes have, under certain assumptions, no impact on investments. We use data from the ECB report on Corporate finance and economic activity in the Euro area (2013) and calculate a fraction of 30% for external financing sources of non-financial corporations (the value used in the previous study of 20% was taken from an empirical cross-country study by Corbett and Jenkinson (1997)).  

Finally, we perform a simple static exercise to get a sense of the aggregate economic impacts. Specifically, we plug the estimated reduction in the cost of capital figure into a Cobb-Douglas production function (with a calibrated capital share of 0.35 and a substitution elasticity of -1), we ultimately obtain the corresponding increase in GDP.

Table 5 reports the results for the different simulated scenarios.

Table 5. Impacts of eliminating the cost of WHT relief procedures

Source: JRC calculations.

3.    Discussion

As mentioned in the introduction, there are a number of caveats with respect to the analysis which suggest caution in the interpretation of the results. These are spelled out in the following:

1)Given the important data limitations, the methodology relies on aggregate financial data. In this respect, it is worth noting that stocks are quite persistent, and on a slight upward trend over the medium term. By contrast, yields are on a marked downward trend, particularly in the more recent years. The lower interest and dividend rates are only partially compensated by the increased stock of cross-border holdings. Moreover, in times of strongly declining rates, the choice of the reference period (2008-2019 in the current study, 2003-2014 in the previous update) affects significantly the estimated values. For instance, the value of the risk free interest is assumed equal to 0.6% (3-m Euribor) in the current exercise, whereas it was 1.9% in the previous update.

2)In the previous study the potential impact on the cost of capital was obtained as the ratio between the change in the flow of payments induced by the removal of the tax barriers to the effective flow of payments. Arguably, one could compute instead the ratio of the estimated change in the flow of payments to the stock of capital measuring the financial effort of the investment.

3)The GDP result is significantly affected by two parameters, namely the percentage of cross-border holdings and the percentage of external financing sources. Values for the Euro Area (based on ECB data) are used in the exercise. A sensitivity analysis could be provided on this. Moreover, the overall impact would be very different (i.e. lower) once we choose to adopt an approach where the impact on the user cost of capital is computed with respect to the investment (see point above).

4)When calibrating the production function, the user cost of capital (instead of the financial cost of capital) could be used, which would take also corporate tax provisions (statutory tax rates and tax depreciation for physical capital) into account. This would require substantial additional data collection and modelling complications.

4.    Further JRC work on withholding taxes and cross-border investment ( CVII )

In addition to the present note, the JRC has planned further analyses to shed light on the effects of withholding taxes (and related relief procedures) on cross-border investment, and aggregate economic variables. The envisaged timeline for these analyses is consistent with that of the policy process, to allow the results being incorporated in the Impact Assessment for the forthcoming Commission proposal for introducing a common, standardised, EU-wide system for withholding tax relief at source.

4.1 Econometric modelling (JRC.B1 and JRC.B2)

First, JRC.B1 and JRC. B2 will collaborate on an empirical study with the goal of estimating the elasticity of bilateral portfolio investment holdings to cross-border withholding taxes. Preliminary analysis points to substantial variation in the country-pair withholding taxes on dividend and interest payments. This, combined with data on bilateral stocks of equity and debt securities available from the IMF CPIS, makes it possible to use panel data gravity-style regressions. The results from this analysis should make it possible to estimate the impact of withholding tax relief procedures on bilateral portfolio investment holdings.

4.2 General equilibrium modelling (JRC.B2)

Second, JRC.B2 will improve the current calculations for the costs of capital by using the general equilibrium model CORTAX. The model has already been used in past IAs for assessing reform proposals for corporate and digital taxation. Interestingly for this analysis, CORTAX captures cross-border flows across member states. The current model version will be extended in order to capture bilateral withholding tax rates in the after-tax rate of return on debt and equity investment. Furthermore, in the calibration, the rate of return on debt and equity will have to distinguish between domestic and cross-border sources of finance. This explicit way of modelling will allow us to observe the impact on cost-of-capital endogenously in the model. Then, by simulating the withholding tax relief reform, we can observe the macro-economic effects of the reform proposal, including its effects on GDP and welfare.

Appendix 1 – Variables and data sources

Annex 6: JRC Study – Effect of non-refundable WHT on FPI and macro-economic indicators    

Summary

The present note describes the analytical contribution of JRC.B1 and JRC.B2 in support to the IA for the Commission initiative for a new EU-wide system for the avoidance of double taxation and prevention of tax abuse in the field of withholding taxes. The note is structured in two parts. The first part proposes an empirical (econometric) model to test the sensitivity of foreign portfolio investment to non-refundable withholding taxes. The second part quantifies the macroeconomic impact of reducing the cost of withholding tax relief procedures making use of the CGE model CORTAX.

Contacts: JRC

1)Introduction

Procedures for granting withholding tax relief on cross-border investment are considered an important barrier to the free movement of capital that hinders the good-functioning of the Single Market. The return to portfolio investment in equity or debt securities made by a non-resident investor is normally subject to withholding tax in the country of the investment (source country). To avoid double taxation, whenever the source withholding tax rate is higher than the reduced rate applicable according to the relevant double tax treaty, the non-resident investor can claim ex-post the refund of the excess tax withheld by the source country. However, the current system for withholding tax refund in place in the EU proves cumbersome and costly, and prone to fraud.

Lengthy and inefficient withholding tax relief procedures can give rise to three different sources of costs that ultimately discourage cross-border investment. First, claiming the refund entails direct financial costs arising from processing fees paid to the custodian or to an external service provider, paperwork, and diversity of source country requirements. Second, the delays – often in years – in refunding investors bring about implicit financial losses for the investors, compared to the ideal scenario with immediate refund. Finally, the fact that some investors, particularly smaller ones, may decide to forego the relief for a number of reasons, makes the current system costly from a macroeconomic perspective, as it holds back cross-border investment. By hindering full integration of securities markets, this hinders their functioning as the main channel of cross-border private risk sharing for firms and households.

Against this background, the European Commission has adopted several actions to address tax barriers to cross-border investment. In 2009, a recommendation to the Member States was adopted on the simplification of withholding tax relief procedures. It was followed by the publication, in 2017, of a Code of Conduct, which put forward new guidelines on withholding tax to help Member States reduce costs and simplify procedures for cross-border investors in the EU. In the context of the renewed efforts to foster capital markets integration under the Capital Markets Union (CMU) initiative, alongside the promotion of fair taxation and the fight against tax fraud in the post-pandemic recovery, the Commission is tackling withholding tax relief procedures as one of the remaining obstacles to cross-border investment in the EU. The Action Plan for fair and simple taxation supporting the recovery strategy proposes to introduce a common, standardised EU-wide system for withholding tax relief at source coupled with a new exchange of information and cooperation mechanism between administrations. The initiative is also one of the actions proposed in the Capital Markets Union for people and businesses new action plan.

This note illustrates the analytical contribution of the JRC in support to the IA for the Commission initiative for a new EU system for the avoidance of double taxation and prevention of tax abuse in the field of withholding taxes. It is structured in two parts. The first part puts forward an empirical (econometric) model to test the sensitivity of foreign portfolio investment (FPI) to non-refundable withholding taxes. The second part quantifies the macroeconomic impact of reducing the cost of withholding tax relief procedures making use of the CGE model CORTAX.

2)Measuring the effects of non-refundable withholding taxes on FPI

1.Empirical strategy

In order to evaluate the potential impact of the policy reform on FPI, the sensitivity of FPI to non-refundable withholding taxes is estimated empirically. The available panel data covers 83 countries, including all Member States, between 2005 and 2019. Following an established practice in the research literature, a gravity-style model is adopted with fixed effects to infer withholding tax elasticities. In particular, the following baseline specification is estimated:

where is the stock of FPI from home (investor) country i in Member State (source) m in year t; is a time-variant vector of non-refundable withholding taxes in Member State m towards an investor from country i; is a dummy indicator taking the value of unity if there is a tax treaty between investor country i and Member State m in year t; and , and are respectively vectors of home-year, source-year and time-invariant country-pair fixed effects. A unit increase in a covariate will lead to a 100*( −1) percentage increase in the FPI stock.

The baseline model is estimated using the Poisson Pseudo Maximum Likelihood (PPML) estimator ( CVIII ). The PPML estimator is chosen as it allows to account for zero FPI stocks and heteroscedasticity in FPI data ( CIX ).

The rich structure of fixed effects allows to infer the effects of non-refundable withholding taxes on FPI from within country-pair variation over time. Specifically home-year and source-year fixed effects control for any country-year specific covariates that may affect cross-border investment, such as GDP, GDP per capita, (financial) market attractiveness, or access to financial markets. Time-invariant country-pair fixed effects account for any covariates that are specific (and constant) for a given pair of countries over time, including geographical distance, common language and past colonial relationship. Therefore, it is only needed to control for covariates that change within a given country-pair over time, for example, the signature of a tax treaty and our main variable of interest, the non-refundable withholding taxes.

2.Data

The main source of data on Foreign Portfolio Investment (FPI) is the “Coordinated Portfolio Investment Survey” (CPIS), conducted by the IMF. Portfolio investment statistics report the international investment position of participating countries, that is their holdings of portfolio investment assets in the form of equity and investment fund shares, long-term debt securities (i.e., debt securities with an original maturity over one year), and short-term debt securities (i.e., debt securities with an original maturity of one year or less). The statistics are reported on an annual basis and are broken down by counterpart economies (those whose residents have issued the securities).

Separate reporting of debt and equity holding allows to analyse the behaviour of the two series separately. Likewise, the breakdown by holding sector (financial corporations, non-financial corporations, households) potentially allows one to detect differences in the cross-border investment behaviour of these agents.

Data on withholding taxes and tax treaties comes primarily from country-specific withholding tax tables prepared by the IBFD and shared with JRC by DG TAXUD. The data has been extensively cleaned in order to identify the applicable withholding tax rates. In particular, considerable efforts have been made to identify the withholding tax rate applicable to negotiable (i.e. tradeable) debt securities owned by the sector of non-financial corporations, households and non-profit institutions serving households (NPISHs).

3.Main results

The main results on the effects of non-refundable withholding taxes on the total stock of equity and debt holdings are presented in Table 1.