EUROPEAN COMMISSION

Brussels, 14.12.2022

SWD(2022) 406 final

COMMISSION STAFF WORKING DOCUMENT

Final evaluation of Regulation (EU) No 1286/2013 of the European Parliament and of the Council of 11 December 2013 establishing an action programme to improve the operation of taxation systems in the European Union for the period 2014-2020 (Fiscalis 2020) and repealing Decision No 1482/2007/EC

Accompanying the document

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND THE COUNCIL

on the final evaluation of the Fiscalis 2020 programme

{COM(2022) 717 final}

Table of contents

Glossary

1.Introduction

1.1.Purpose and scope of the evaluation

2.What was the expected outcome of the intervention?

2.1.Description of the intervention and its objectives

2.2.Point(s) of comparison

3.How has the situation evolved over the evaluation period?

3.1.Current state of play

3.2.Follow-up on the recommendations of the mid-term evaluation and raising programme awareness

4.Evaluation findings

4.1.Effectiveness - to what extent was the intervention successful and why?

4.1.1.To what extent has Fiscalis 2020 enabled the secured exchange of information, supported administrative cooperation and enhanced participating countries administrative capacity?

4.1.2.To what extent has, Fiscalis 2020 helped to implement Union taxation law?

4.1.3.To what extent has Fiscalis 2020 supported the fight against tax fraud, evasion and aggressive tax planning

4.1.4.To what extent have Fiscalis 2020 resources produced the best possible results at the lowest possible cost?

4.1.4.1.Costs compared to benefits associated with Fiscalis IT systems

4.1.4.2.Costs compared to benefits associated with Fiscalis joint actions

4.1.4.3.Training activity costs compared to benefits

4.1.4.4.Scope to maximise efficiency

4.1.5.To what extent is Fiscalis 2020 coherent with broader EU policies

4.2.How did the EU intervention make a difference (value added)?

4.2.1.EU added value of Fiscalis 2020 IT systems

4.2.2.EU added value of the Fiscalis 2020 joint actions

4.2.3.EU added value of the Fiscalis 2020 training activities

4.2.4.Likely impact if hypothetically Fiscalis was discontinued

4.3.Is the intervention still relevant?

4.3.1.Challenges and priorities in the field of taxation

4.3.2.Extent to which Fiscalis 2020 met tax authorities’ needs

4.3.3.Extent to which Fiscalis 2020 met economic operators’ needs

4.3.4.Extent to which Fiscalis 2020 met public needs

5.Conclusions and lessons learned

Annex I: Procedural Information

Annex II. Methodology and analytical models used in the final study

Approach

Evaluation questions

Overall methodological framework

Programme level evaluation

Data collection and analysis

Desk research    

Stakeholder consultations    

Data analysis    

Case study selection    

Triangulation

Conclusions and lessons learned

Limits to the evidence base and robustness of findings

Annex III. Evaluation matrix and, where relevant, details on answers to the evaluation questions (by criterion)

Annex IV. Overview of benefits and costs and, where relevant, table on simplification and burden reduction.

Annex V. Stakeholder consultation – synopsis report

Stakeholder consultation activities – methodology and approach

Interviews

Programme level interviews

Case study consultation activities

Public consultation

Challenges and mitigation strategies

Key findings per evaluation criteria

Effectiveness

Efficiency

Relevance

Coherence

EU added value

Annex VI. Mid-term evaluation recommendations

Glossary

11.    Introduction

This Staff Working Document (SWD) reports on the final evaluation of the Fiscalis 2020 multiannual action programme (herein Fiscalis 2020). The programme, running from 2014 to 2020, was one of the principal tools used to support the implementation of European Union taxation policy. It provided a framework with which to improve the proper functioning of the taxation systems in the single market through enhanced cooperation between participating countries, their tax authorities and officials. It supports the implementation of Union law in the field of taxation and the fight against tax fraud, tax evasion and aggressive tax planning by ensuring the exchange of information, supporting administrative cooperation and enhancing participating countries’ administrative capacity.

The Fiscalis 2020 programme was set up by Regulation (EU) No 1286/2013 of the European Parliament and of the Council of 11 December 2013 establishing an action programme to improve the operation of taxation systems in the European Union. 1 In compliance with Article 17(1)(2) of the Regulation, the Commission, assisted by independent external evaluators, conducted studies to support final and mid-term evaluations of the programme.

21.1.    Purpose and scope of the evaluation

As mandated by Article 17(3) of the Fiscalis 2020 Regulation, the purpose of the final evaluation is to assess the programme’s performance from 2014 to 2020 2  and the extent to which results achieved will be sustained in the longer term. The scope of the evaluation covers the programme’s years of operation (2014-2020). It incorporates the mid-term evaluation findings conducted in 2018 and assesses the degree to which recommendations made then were implemented and whether such follow-up actions brought about improvements. The geographical scope includes all participating countries, i.e. EU Member States 3 , candidate countries (Albania, North Macedonia, Montenegro, Serbia and Turkey) and potential candidates (Bosnia and Herzegovina and Kosovo*). The evaluation takes into account the programme’s full range of funded and management activities and all the relevant stakeholders (national administrations, Commission services and economic operators) from all participating countries (including candidate and potential candidate countries).

This evaluation round focused strongly on Fiscalis IT structure and systems (also referred to in this report as European Information Systems – EIS) as they account for around three quarters of the Fiscalis 2020 budget and were not previously evaluated in-depth. This was done through five in-depth case studies. See Annex II for more on the final study methodology including the case study selection. Other Fiscalis-funded activities are also evaluated, providing a well-rounded picture of the programme performance as a whole. The final evaluation also draws on the experience of its predecessor programmes, as Fiscalis 2020 cannot be evaluated without considering them. In addition, it considers future evolutions for the programme’s continued relevance and changes foreseen for the 2021-2027 programming period.

In line with Article 18(1) of Regulation (EU) No 1294/2013, which mandates an independent final evaluation of the programme, the Commission services (Directorate General for Taxation and Customs Union - DG TAXUD) commissioned an external contractor to perform a study in support of the Commission evaluation (hereinafter referred to as ‘the external study’). The external study was carried out between November 2020 and January 2022 and is a key contribution to this staff working document (SWD).

The evaluation assesses the degree to which the programme achieved its objectives, the efficiency of resources used (including potential for simplification), the programme’s contributions to the EU priorities of smart, sustainable and inclusive growth, the continued relevance of its objectives, and its added value at EU level. On that basis, this evaluation provides the Commission with overarching conclusions and lessons learned to inform the implementation of the recently established successor Fiscalis programme for the 2021-2027 period. The findings are deemed robust and the conclusions accurately drawn. The staff working document is based on the findings and conclusions presented in the supporting external studies (see Annex 1 for organisational details).

32.    What was the expected outcome of the intervention?

42.1.    Description of the intervention and its objectives

As said, the Fiscalis 2020 programme was set up by Regulation (EU) No 1286/2013 of the European Parliament and of the Council of 11 December 2013 establishing an action programme to “improve the proper functioning of the taxation systems in the internal market by enhancing cooperation between participating countries, their tax authorities and their officials”. The general objectives of the EU taxation programme are to facilitate the coherent implementation of EU taxation law and support the fight against tax fraud, tax evasion and aggressive tax planning. This should be achieved, by enabling secure information exchange and storage, supporting administrative cooperation and enhancing the tax authorities’ administrative capacity (specific objectives).

The Fiscalis programme is an ongoing, continuously growing and adapting programme. The Fiscalis 2020 programme was built on predecessor programmes. The first programme was called Matthaeus-Tax, started in October 1993 and ran until end-December 1995. Rebranded Fiscalis, there have now been four Fiscalis programmes, including the one evaluated here: the first Fiscalis programme (1998-2002); the second Fiscalis programme (2003-2007); the third Fiscalis programme (2008-2013) and the fourth Fiscalis programme (2014-2020), or Fiscalis 2020, evaluated here. 35 countries fell within the Fiscalis 2020 framework: the 27 EU Member States plus the UK while it was an EU Member State and 7 candidate and potential candidate countries.

The Commission manages the programme with assistance from the Programme Coordination Group (responsible for Commission internal coordination) and the Fiscalis 2020 Committee 4 . The Fiscalis 2020 Committee is composed of delegates from each EU Member State. The Programme Coordination Group is composed of DG TAXUD units using the programme and contributes to identifying policy priorities for the Annual Work Programmes, reinforcing coherence between the programme objectives and priorities of the European Commission, monitoring of the programme’s performance and increasing the overall transparency and dissemination of information.

52.2.    Point(s) of comparison 

It is difficult to establish points of comparison because the Fiscalis 2020 programme has a long history and builds upon previous iterations of the programme. As it has been in existence for several years, it is challenging for stakeholders to consider the situation before this started. Take for example, the EIS systems, constituting about 75 % of the programme’s funds. These have long been engrained in national and EU landscapes.

In line with the Commission’s commitment to monitor the EU budget 5 and ensure the accountability for value for money, DG TAXUD developed a framework for monitoring the outputs and results of the programme – the Performance Monitoring Framework (PMF). The European Commission PMF monitors the programme’s outputs and results through a results-based monitoring system. It contains an intervention logic, adjustable indicators, data collection schemes and reporting arrangements (see Figure 1). This framework became operational in 2014, following recommendations of earlier evaluations to provide more and better data to assess programme implementation and performance. The PMF outcome has been taken into account to the extent possible in this evaluation. Note that the mid-term evaluation reported then that while ‘…it does already facilitate performance management; the framework is still too fresh to provide insightful benchmarks for the programme’s performance or establish and examine trends’.

It is important to note that several issues make it difficult to accurately isolate and estimate the programme’s contribution to the overall EU objectives such as that of smart, sustainable and inclusive growth. First, all programme components play supporting roles. They are not themselves taxation policies and thus disentangling the supporting aspect from the policy implementation aspect and evaluate the programme’s performance vis-a-vis the objectives is challenging. Second, the programme consists of various components, which, while independent in nature, all work towards the same objectives and each of these does not necessarily impact or rely on other components in the same way.

Figure 1: Fiscalis 2020 Intervention logic

63.    How has the situation evolved over the evaluation period?

73.1.    Current state of play

The Commission reports on the implementation and performance of the programme annually, through the Annual Progress Reports. All report for 2014 to 2020 are available. The overview of the programme’s implementation presented here is based on the latest available report. 6

As said, the Fiscalis 2020 programme aimed at improving the proper functioning of taxation systems in the internal market, by ensuring the exchange of information, supporting administrative cooperation and enhancing participating countries’ administrative capacity. In this context, the actions under the programme were varied and included the development/ improvement of European information systems, a variety of joint actions, training and external expertise. The Fiscalis 2020 programme had a committed budget of EUR 222 million for the period 2014-2020 (see Table 1 ).

Table 1 : Committed expenses per budgetary year and main programme action categories, in euros

Source:    Fiscalis 2020 – Progress Report 2020

Notes:    1. The table compares committed amounts since the programme’s launch, as the actual expenses are not finalised for all previous years. Budgetary year X covers from 01/04/X to 31/03/X+1;

2. Appropriations may be higher when using foreseen financial contributions from candidate and potential candidate countries participating in Fiscalis 2020.

The vast majority of Fiscalis 2020 funding, 75.6% (about EUR 168 million), went into the development and operation of the EIS systems, followed by the organisation of joint actions other than expert teams and studies, then studies and communication, then training, then expert teams and finally external experts (see Table 1 ).

European Information Systems (EIS)

Fiscalis supports the development of EIS which in turn can secure information exchange, support administrative cooperation and enhance participating country authorities’ administrative capacity. EIS represent a large share of the funding and comprise many systems. Here is a subset of those that we will be looking at in more detail in this evaluation:

·CCN/CSI - CCN/CSI is the backbone to Fiscalis IT systems. It provides the hardware and software infrastructure to implement multiple IT systems and applications.

·eFCA - The e-forms enhanced tax officials’ capacity by providing national tax administrations a homogenous information exchange format.

·EMCS - With the EMCS, EU customs authorities can monitor the movement of excise goods under duty suspension within the EU.

·TIN - Created to ensure effective use of exchanged information by improving taxpayers’ identification through their TIN is essential.

·MOSS - The MOSS was created to reduce the administrative burden of compliance with VAT rules in cross-border sales for economic operators and eased the logistics of collecting VAT for tax officials.

·DAC 1-6 systems - The DAC systems were implemented to support Member States in the Automatic Exchange of Information.

·TNA - TNA improved tax administrations’ capacities through the Eurofisc network with advanced data analytics and efficient data sharing and its open-source software IT system. It replaced previously labour-intensive exchange via Excel tables.

·TEDB - This database provides information on all major taxes in Member States, facilitating information exchange.

Joint Actions

Eligible joint actions for which the Fiscalis 2020 programme could fund, provided certain conditions were met, were the following:

a. seminars and workshops;

b. project groups, generally composed of a limited number of countries, operational during a limited period of time to pursue a predefined objective with a precisely described outcome;

c. bilateral or multilateral controls and other activities provided for in Union law on administrative cooperation, organised by two or more participating countries, which include at least two Member States;

d. working visits organised by the participating countries or another country to enable officials to acquire or increase their expertise or knowledge in tax matters;

e. expert teams, namely structured forms of cooperation, with a non-permanent character, pooling expertise to perform tasks in specific domains, in particular in the European Information Systems, possibly with the support of online collaboration services, administrative assistance and infrastructure and equipment facilities;

f. public administration capacity-building and supporting actions;

g. studies;

h. communication projects;

i. any other activity in support of the overall, specific and operational objectives and priorities set out in article 5 and 6 of Regulation (EU) 1286/2013, provided that the necessity for such other activity is duly justified.

Over 1500 joint actions and multiplying events took place during the 2014-2020 period as under one action, multiple events may be organised. Project groups, workshops and multilateral controls (MLCs) had the most weight on the budget in the funding period, while working visits had the highest number of actions. The overview also shows that workshops and project groups tend to be more cost-intensive per action, while working visits consume less per action.

In total more than 30,000 participants (instances of participation as the same person can attend more than one action) attended the joint actions (see Figure 2 ). The numbers of participants in the Fiscalis 2020’s joint actions remained at a high constant level especially after 2015. On average for the 2014-2020 period there were above 4300 participations/year. This compares to the evolution of in the previous programming period which saw a gradual dip that spilled into 2014. The low 2014 participation was attributed to the delayed transitioning to the Fiscalis 2020 programme which became operational only from April 2014. 

Following a larger and more stable performance over the duration of the Fiscalis 2020 compared to its predecessor, participation dipped again in 2020 this time due to the COVID-19 pandemic and consequent measures leading to an interruption of activities for some months. These have gradually resumed in a different, on-line format. The stakeholders’ community adapted to the new working methods (but with a lag which naturally had an impact on the budget consumption). Despite the pandemic, on average, the Fiscalis 2020 final joint action attendance for the whole period was about 4300 participants a year and higher than the average attendance observed during the mid-term evaluation (around 4200 participants per year).

Figure 3  shows the total number of participants per action type. Project groups represent by far the largest number of instances of participation of taxation officials in the programme. This is mainly due to the repeated nature of these types of joint actions, where individuals who participate often attend several meetings over time, sometimes lasting the entire funding period. By comparison, the nature of other joint actions, e.g. working visits, is such that they typically involve just a few officials. Workshops and multilateral controls also register a high number of instances of participation, due to both a high number of actions (especially for multilateral controls) and many instances of participation per action.

Figure 2: Evolution of participation numbers in joint actions under Fiscalis 2013 and Fiscalis 2020

Source: DG TAXUD ART data

 Figure 3: Total instances of participation in joint actions according to action type, 2014-2020

Source: DG TAXUD ART data

Participation varies by country (see Figure 4 for 2020), and across the period. In 2020, Germany was in the lead followed by Sweden and Italy. Overall, there is a much lower number of non-EU country participants.

 Figure 4: Overview of joint action participants per country in 2020

Source: DG TAXUD ART data

Training

The Commission procured 267 EU eLearning modules, with courses that mainly focus on training officials to use the EMCS and on various VAT, fraud and refund related topics. The training modules included webinars, eBooks, a Learning Management System ( Customs and Tax EU Learning Portal ), and European Competency Frameworks for Customs and Taxation, the first ‘EU Competency Framework for the tax profession’, the TaxCompEU training material. 7 The Commission also organised 141 skills and competences training sessions for IT systems such as VAT refund, EMCS, VIES or MOSS for tax officials. The majority of the Fiscalis 2020 training budget was allocated to eLearning modules. These modules trained more than 130,000 participants from tax authorities and almost one million additional people from businesses and public. 

83.2.    Follow-up on the recommendations of the mid-term evaluation and raising programme awareness

The Fiscalis 2020 mid-term evaluation 8 outlined 13 recommendations to strengthen and improve the programme and future funding phases (see  Table 2  and Annex VI). Five of the recommendations identified the Commission’s DG TAXUD as the sole responsible actor, one with other Commission services, five as having shared responsibility with national administrations and two where national authorities were the main actor. Of the 11 recommendations where the Commission was asked to act, 8 were to cover both the 2014-2020 and the subsequent funding programmes and 3 were relevant for the following programme.

In response, the Commission services (DG TAXUD) developed an internal action plan setting out 13 action areas corresponding to the recommendations, with 48 sub-actions outlining how to address or, in the case of those identified for national administrations only, support the implementation of the mid-term evaluation recommendations. This section briefly lays out progress made in implementing the Action Plan, based on DG TAXUD interviews.

The mid-term evaluation concluded at the end of 2018 and the action plan was adopted in October 2019. As such, there was not much time for major change to occur and thus many recommendations will relate to the 2021-2027 programme period and will be completed during 2022. Further, many recommendations and actions require active participation and cooperation by Member States’ or national programme coordinators, and thus go beyond the Commission and DG TAXUD’s sole control.

Despite these caveats, the state of play by March 2022 regarding progress in addressing the recommendations and implementing the various actions under each recommendation can be seen below (see Table 2 and Table 3 ). As can be seen the vast majority of the recommendations has been addressed as the vast majority of actions in the action plan have been implemented (37 out of 48 can be considered implemented). Actions in progress or yet to be implemented are largely linked to ‘Monitoring and reporting’ or Communication.

Table 2: Mid-term evaluation recommendations by responsibly actor, period and status

Table 3: Implementation state of play of the 48 actions identified for the 13 mid-term recommendations

Note that the Commission also monitors programme awareness. The Commission therefore runs a programme poll every 18 months to measure familiarity and networking levels in the programme. It is conducted in participating countries’ tax administrations and invites programme participants and non-participants to take part. The last programme poll ( Table 4 ), conducted end-2019, shows declining awareness since 2016 (59 %) to 45 % in 2019. The results are also now below the 2011 baseline (66 %) and has some way to go to meet the target (>75 %). Some caution is needed when comparing the results of the pools as the 2019 programme poll targeted a larger audience within the national administrations 9 . This may help explain the lower awareness rate. Several communication actions had already taken place at the initiative of the Commission (these included for example the use of new communication tools and channels and a common effort between the EU and national programme teams to ensure a wider distribution of information on the programme to the potential beneficiaries). Nevertheless, it was felt that there was an important potential for further raising awareness about the programme at the level of national authorities. In this context, in 2019, a new communication strategy was launched by DG TAXUD.

Table 4: Programme Poll baseline, target and outcomes

Source: Fiscalis Annual Progress Report 2020

94.    Evaluation findings 

104.1.    Effectiveness - to what extent was the intervention successful and why? 

When evaluating effectiveness, efficiency and coherence we look at the Fiscalis 2020 as whole and how its components contributed to achievement of objectives. It considers per objective, the IT systems that are part of the EIS, joint actions and training that were implemented and assesses their success, and to what extent observed results could be attributed to the programme rather than external factors.

114.1.1.    To what extent has Fiscalis 2020 enabled the secured exchange of information, supported administrative cooperation and enhanced participating countries administrative capacity?  

This evaluation question covers the Fiscalis specific objectives of (1) securing exchange of information; (2) supporting administrative cooperation; and (3) enhancing participating countries tax authorities’ capacity so as to reduce administrative burden and taxpayer compliance costs. The Fiscalis Performance Management Framework (PMF) includes useful indicators on EIS, administrative cooperation and administrative procedure sharing. The annual progress reports provide indicators on how Fiscalis actions enhanced tax officials’ administrative capacity, through sharing administrative guidelines and procedures as well as enhancing authorities’ staff, skills and competences.

As said, Fiscalis supports the development of EIS which in turn can secure information exchange, support administrative cooperation and enhance participating country authorities’ administrative capacity. In so doing, they enable authorities and economic operators to reduce the time and resources required to carry out activities such as exchanging information or files, identifying taxpayers and collecting and analysing data and provide tax officials time to perform additional activities. Key EISs supporting these objectives and their functionalities are listed in Table 5 .

Table 5: EISs that contribute to enhancing administrative capacity

Take the CCN, for example, as the backbone IT infrastructure securing information exchange. It was available 100 % of the time according to the evaluation. The data exchange volume via the system steadily increased, reaching 32.23 terabytes in 2019 compared to just 4.3 terabytes in 2014 10 . Clearly, secure information exchange increased over time, though this is also because of the growing number of EISs (28 taxation systems in 2020 compared to 20 in 2014). Similar trends were observed for specific EIS (see Box 1 ). Overall, the systems helped Member States communicate with each other securely and efficiently in areas of mutual interest, thereby enhancing collaboration. In addition, 9 out of 11 interviewed Fiscalis coordinators of EU countries explained that IT systems have modernised and harmonised their systems information exchange procedures and in turn improved their work efficiency and capacity. Among the IT systems case studies followed by the external contractors, the MOSS stands out for reducing administrative burden for economic operators and tax officials (see Box 2 ).

Box 1: Case study examples of Fiscalis EIS frequency of use

Box 2: Case study insights of the MOSS role in enhancing participating countries admin capacity

While it is early to make conclusions about the TNA, the majority of surveyed TNA users 12 reported that TNA decreased the administrative burden, associated with the previous way of exchanging data, and that they had more time to dedicate to analysis 13 . Under the DAC, all interviewed tax authorities agreed that the Automatic Exchange of Information under DAC1/DAC2 significantly enhanced their capacity to exchange information with other Member States.

Table 6  reports the joint actions followed regarding these sub-objectives specifically. It can be seen that about 500 events took place on enhancing administrative capacity, more than 300 on secured information exchange and more than 600 on administrative cooperation. Working visits were one type of action often used through which officials could acquire or increase tax expertise or knowledge. Member States also exchanged tax knowledge and experiences through project groups, often used to discuss, develop and implement EIS. Workshops, although used less frequently, sought to improve tax administrations’ procedures. Capacity building actions mainly consisted of technical assistance to a few Member States. They used less to enhance administrative capacity because it is only used in a smaller number of Member States and initiated by the Commission. Multilateral controls (MLCs) and PAOEs are examples of direct administrative cooperation between tax authorities. Note that MLC processes are usually resource intensive and require adequate support to be done efficiently and effectively. Hence, Member States can request funding under the Fiscalis programme budget. In the case of MLCs, the Fiscalis programme finances the collaborative aspect of this type of joint action (e.g. a coordination meeting between participating country authorities or to finance transport to facilitate cross border simultaneous controls). In other words, Fiscalis provides financial support for participating countries to come together – if the action is under the remit of the Fiscalis programme regulation of course as set above – but it is the responsibility of the participating countries to organise the action. At no stage is the Commission involved in such procedures. Nor does it have access to any information exchanged during those procedures. They are conducted solely by tax authorities of Member States. The involvement of the Commission is limited to approving the financing under the Fiscalis programme of collaborative activities supporting the multilateral control proposed by the tax administrations. While less common, there were working groups and seminars for specific IT systems or for sharing EU taxation law information.

On average 95 % of event participants shared working practices, administrative procedures, or guidelines from their national administration programme, and on average 77 % of them declared that an administrative procedure, working practice, or guideline from the programme led to change in their administrations’ working practices.

Participants and action managers of joint actions regarding working practices and administrative procedures/guidelines were positive about their effectiveness and usefulness, and largely agreed that the events achieved intended results. In 2020, 88% of action managers of the 61 joint actions that sought to extend working practices and/or administrative procedures/guidelines indicated that they achieved their intended results 14 . 

Table 6: Events, participants and budget per joint action type to improve administrative capacity

The great majority of interviewed Fiscalis coordinators consider training to be key to enhancing administrations’ administrative capacities. The Commission procured 267 EU eLearning modules, with courses that mainly focus on training officials to use the EMCS and on various VAT, fraud and refund related topics. The training modules included webinars, eBooks, a Learning Management System ( Customs and Tax EU Learning Portal ), and European Competency Frameworks for Customs and Taxation, the first ‘EU Competency Framework for the tax profession’, the TaxCompEU training material. 15 The Commission also organised 141 skills and competences training sessions for IT systems such as VAT refund, EMCS, VIES or MOSS for tax officials. 

The number of tax officials trained is growing, as more and more tax officials have been trained using EU common materials 16 . Participants’ training quality scores are high (average 72.7 score on a 0-100 scale). Similarly, the share of tax officials who found IT training met their expectations or were useful are consistently over 90 %.

Collected evidence suggests that the IT systems that make up the EIS helped Member States secure information exchange, contribute to administrative cooperation and build up administrative capacity. IT tools were also key in enhancing administrative capacity by allowing tax authorities to perform additional actions. The evidence also shows that Fiscalis-funded actions and training enhanced tax authorities’ administrative capacities by enhancing staff skills and competences and sharing administrative procedures and guidelines.

All 11 interviewed Fiscalis coordinators of EU Member States were of the view that IT systems, joint actions and training modules would not be available to the same extent without Fiscalis funding, and their functions would be more difficult to undertake. This was supported in interviews with the developers of the Excise Duty Calculator, as it is largely dependent on input from another Fiscalis-funded IT system – TEDB. While some joint actions could have happened, they would not have been possible to the extent and pace. Thus, it is acceptable to say that secured information exchange, administrative cooperation and enhanced administrative capacity would not have been achieved to the same extent without Fiscalis.

As mentioned in section 2.2, a separate study has been carried out on the Performance Measurement Framework. The outcome and recommendations of that study have been taken into account in the new Monitoring and Evaluation Framework of the Fiscalis programme (MEF) that is currently being put in place. This will allow the Commission to collect additional evidence and improve monitoring and evaluation of the new cycle of the Fiscalis programme vis-à-vis the objectives of the programme.

124.1.2.    To what extent has, Fiscalis 2020 helped to implement Union taxation law? 

This section looks at how the Fiscalis programme contributed to the general objective of supporting the implementation of Union taxation law.

Maintaining Fiscalis EIS systems is a big part of the funding and effort to achieve this specific objective and all IT systems contribute to it. A few are especially relevant ( Table 7 ) because of their role in implementing Union law by enabling secure information exchange, supporting administrative cooperation between tax authorities and enhancing their administrative capacity. Among the joint actions, there were some that were directed at the implementation of Union taxation law specifically (see Table 8 ). Together with the other actions and training they again contributed to a better understanding of EU tax law and better administrative capacity that would in turn lead to a better implementation of EU Law.

Table 7: IT systems that support implementation of Union law

Table 8: Events, participants and budget per joint action type to implement Union law (2014-2020)

Source: ART data on events.

Measuring Fiscalis 2020’s impact on the implementation of Union law in taxation is difficult, as it is difficult to establish a baseline comparison of no Fiscalis programme because of the long history of tax cooperation under previous Fiscalis programmes, and because of the nature of the programme actions vis-à-vis the objective i.e. they are supporting actions.

Joint actions contributed to the objective of implementing Union law, notably through administrative cooperation and administrative capacity building. Joint action created networks between participating tax authorities, which all interviewees considered as having significant added value. Seminars, workshops and project groups that bring together many different Member States were considered useful, as they foster mutual learning and thus better understanding and alignment with legislation. Other actions, such as working visits, were also considered useful, as they tend to be focused on good practices for a specific approach or EIS, thus fostering mutual learning as well.

Results from Fiscalis event polls substantiate these generally positive results (each year on average 93% of respondents agreed that their participation in joint actions provided a good opportunity for them to expand their network of (and contacts with) officials abroad). The joint actions that were intended to improve understanding and implementing of Union taxation law were largely considered (very) useful and achieving their intended result.

Interviewed candidate and potential candidate countries also appreciated the joint actions for networking opportunities, but felt they were not sufficiently addressing their needs, as actions prioritise EU countries and centred on EISs to which candidate countries have no access (see section 4.3).

E-Learning modules also played a role in implementing Union law. For example, a range of eLearning modules were aimed at explaining VAT rules (e.g. key principles of the VAT Directive, how to distinguish VAT territories, what constitutes a taxable person, and which goods and services are subject to VAT, etc.), in addition to training modules that explain the use of IT systems. As they explain the legal basis behind the rules, and how to interpret and implement them, the modules support the implementation of Union law. Although there is no measure of the extent to which they have successfully supported Union law, trained tax officials found them useful and were highly satisfied with them. In fact, their impact went beyond tax administrations. An interviewee explained they used the course material in school visits to educate children on tax issues.

It is difficult to fully attribute progress towards implementing Union law in taxation to Fiscalis 2020. However, from the evidence presented above, it can be said that the Fiscalis IT systems, joint actions and training have been used intensively and in turn could have contributed to improving the understanding and the implementation of EU Law, via the secure exchange of information, improving authorities’ administrative cooperation and administrative capacity.

The evaluation process here described also indicates that all 11 interviewed Fiscalis coordinators of EU Member States were of the view that IT systems, joint actions and training modules would not be available to the same extent without Fiscalis funding and thus information exchange and cooperation would be more difficult and so would be the functions undertaken by tax administrations. Without Fiscalis 2020, some degree of coordination between authorities would have been likely. However, all interviewed national Fiscalis coordinators as well as national tax officials interviewed as part of the DAC and MOSS case studies felt the level of investment would have been lower and processes would definitely not be as effective and efficient (see Box 3). Fiscalis 2020 thus may have played a significant role in the effective implementation of EU legislation on tax matters. As the EU’s main financial tool for taxation, it is inextricably linked to policymaking.

Box 3: Case study examples on their contribution to implementing Union law

134.1.3.    To what extent has Fiscalis 2020 supported the fight against tax fraud, evasion and aggressive tax planning 

One key aspect to look at when evaluating the Fiscalis 2020 programme is its potential contribution to the general objective of supporting the fight against tax fraud, tax evasion and aggressive tax planning. A number of EIS and joint actions were implemented that could directly and indirectly strengthen the fight against tax fraud, evasion and aggressive tax planning. Key EIS and their functionalities that can directly and indirectly support this objective are listed in Table 9 .

Table 9: List of EISs implemented to fight tax fraud, evasion and aggressive tax planning

Several indicators suggest that EISs contribute to strengthening the fight against tax fraud, tax evasion and aggressive tax planning. As previously described, the steadily increasing volume of CCN messages exchanged indicates that tax authorities have been active in exchanging data, which is key in this process.

Evidence from two case studies on IT systems suggests that EIS contributed to fighting tax fraud by reducing tax administrations’ administrative burden and facilitating businesses’ VAT compliance (see Box 4 : Case study insights of selected EIS’ contribution to fighting tax fraud, evasion and aggressive tax planning ). The TNA is an automated data-mining tool, which allows quick assessment and reporting of potential cases of Missing Trader Intra-Community (MTIC) 18 fraud. The Eurofisc network uses it and its advanced data analytics and algorithms to quickly and accurately detect potential MTIC fraud for further investigation. The Automatic Exchange of Information (AEOI) under DAC1 provides automatic exchange of five categories of income and assets: employment income, pension income, directors’ fees, income and ownership of immovable property and life insurance products. It also includes provisions for spontaneous exchange of information and exchange of information on request. It also contains provision for Multilateral Controls (MLS) and Participation in Administrative Enquiries (PAOE). DAC2 provides automatic exchange of financial account information on (not corporate) taxpayers’ income earned and accounts held in other Member States. DAC1 is in place since 1 January 2015 and DAC2 since 1 January 2016.

Box 4: Case study insights of selected EIS’ contribution to fighting tax fraud, evasion and aggressive tax planning

Fiscalis-funded expert teams substantially supported the development of IT systems that aim to contribute to fighting tax fraud, tax evasion and aggressive tax planning. Ten expert teams developed the following aspects and tools of EISs: Management of IT collaboration in taxation (MANITC I – IV); Automatic Exchange of Information – DAC2; Transaction Network Analysis (TNA I – TNA II); Mobile Application for EMCS Controls; Excise Duty Calculator; Central Electronic System of Payment for data exchange.

Expert teams can play a decisive role in enhancing the development and implementation of some EIS and indirectly fight tax fraud, evasion and aggressive tax planning. The evidence from the TNA case study suggests that the two related expert teams have been instrumental in developing the IT system, and there is some evidence that some more potentially fraudulent cross-border transactions have been identified. The case study on DAC2 also provides indicative evidence in the form of opinions of interviewed Member States, which indicated that the deterrent effect of the Automatic Exchange of Information under DAC2 contribute to fighting tax fraud and tax evasion.

On average, about 44 % of all joint action proposals submitted were linked to fighting tax fraud, evasion and aggressive tax planning and consumed 46 % of the total joint actions’ budget. Under this objective, workshops had the most participants and highest share of the budget (see Table 10 ). Project groups included actions to enhance the capacity of the Eurofisc network, VAT fraud projects and training for MLCs. Where it concerns MLCs, it should be noted that the Fiscalis programme can only financially support the collaborative aspect of MLCs such as coordination meetings involving at least two Member States. During the coordination meetings, in which the Commission never participates, representatives of participating countries are present and exchange information on specific entities and their business relations, to resolve complex taxation cases. If audits are carried out by a national authority alone, the Fiscalis Programme does not provide the funding, as these do not represent a collaborative activity. Other joint actions were working visits, administrative cooperation, presences in administrations and participation in administrative enquires (PAOE). Working visits enable visiting and hosting tax officials to exchange experiences and enhance participants’ tax skills and knowledge. During PAOEs, a Member State’s officials may be present in another’s office during administrative enquires carried out in its territory. During the presence, they may also examine records and interview individuals in the hosting Member State. The distribution of proposals is not geographically balanced and non-EU Member States submitted a much lower share of proposals for this objective.

Table 10: Events, participants and budget per joint action to fight tax fraud, evasion or aggressive tax planning (2014-2020)

Source: ART data on events.

Figure 5: Share of submitted proposals to fight tax fraud, evasion and aggressive tax planning in percent (2014-2019)

Source: ART data on joint action proposals

MLCs are important joint actions for fighting tax fraud, evasion and aggressive tax planning according to most Fiscalis Coordinators. The overall number of operational MLCs 20 grew from 76 in 2014 to 198 in 2019. The number of operational MLCs dropped to 51 in 2020, due to the COVID-19 pandemic. MLC events also increased in the evaluation period, except in 2020 where they reduced for the same reason. The most active MLC event hosts were Germany and the Netherlands, while Eastern European countries were less active. Note that the evaluation only looked at the collaborative aspects of MLCs which is what Fiscalis supports. MLCs have their own legal basis which also stipulates the way information is accessed by different entities 

Figure 6: Number of events for MLCs by Member State (2014-2020)

Source: ART data on events

Based on a survey, Figure 7 shows MLCs average effectiveness rating by MLC coordinators on a scale from 0 “results not achieved” to 4 “results fully achieved”. Indeed, each year, all coordinators of MLCs that took place that year are asked to provide their own assessment of the degree to which the MLCs achieved the results in a survey. One MLC has one MLC coordinator. The average yearly rating increased up to 3.5 in 2020, which suggests that, in general, MLCs’ effectiveness is positive according to MLC coordinators. 21 There is no data on the monetary benefits of MLCs, but the Fiscalis 2020 Progress Reports suggest that they have consistently led to recovered revenue and increased taxpayers’ compliance. Hence, the external evaluation study concluded that they are in general effective regarding this objective. This study did not however assess the quality of each and every MLC conducted.

Figure 7: Achievement of results, as assessed by MLC coordinators

Source: AFF, as reported in the 2020 annual progress report of Fiscalis 2020 (scale of MLC coordinator assessment: 0: not achieved; 4: fully achieved).

The DAC1/DAC2 and TNA case studies found that joint actions are implemented to support the developing, implementing or improving IT systems through project groups, workshops or training sessions. Under the DAC1/DAC2, joint actions were used to exchange best practices and experience among Member States to improve data matching rates, usage quality and statistical reporting. As part of the process of developing and deploying the TNA, three Fiscalis-funded project groups explored its feasibility and implementation and oversaw and assisted its implementation. The work of the first project group informed the work of the expert team, tasked with developing the IT system. Two TNA training sessions were also organised with 55 participants in total, which survey respondents saw as useful 22 . These two examples highlight the synergies between various Fiscalis 2020 activities in developing or improving EIS.

To see if the Fiscalis 2020 programme has supported the fight against tax fraud, tax evasion and aggressive tax planning by improving the functioning of taxation systems in the internal market, one can look at the ability of tax systems to collect tax revenue which is the primary source of income to support Member States public policies. On average, and as a share of GDP, tax revenues (which are significantly higher in the EU compared to other advanced economies such as Canada, the US or Japan), constituted about 40.2 % of EU-27 GDP in 2020, following a steady increase since 2015 (see Figure 8 ).  23  This could be evidence that the programme contributed to improve the ability of tax administrations to collect taxes though of course it is difficult to establish causality as increased tax revenues also depend on several other variables including economic growth, tax rates and the tax base.

One indicator often used as a measure of the effectiveness of the tax system and tax administrations to collect revenue is the VAT gap indicator 24 for which most EU countries have comparative estimates. It can be seen that there has been a steady decline of the VAT gap from 2014 to 2019 (see Figure 9 ) and a related decline in VAT revenue loss. Such positive developments could be related to policy actions and increased capacity by tax administrations that were supported by the Fiscalis 2020 programme though as said it is difficult to establish the accurate contribution of the programme alone.

In other words, evidence shows that tax collection improved and tax fraud and aggressive tax planning levels may have declined on the basis of these (limited number of) indicators. Therefore, it may be the case that the Fiscalis 2020 has supported such developments indirectly through the securing of exchange of information, improving cooperation and administrative capacity. As said, it is difficult to connect trends and developments in tax fraud to Fiscalis 2020 in an exact manner, though qualitative analysis via interviews and case studies suggest that Fiscalis had a positive impact in this regard (see below). Note too, that, in 2019, EU countries lost an estimated EUR 134 billion in VAT revenues which suggests that there is still room for improvement.

Figure 8: EU27 tax revenue compared to Canada, Japan and USA (% of GDP)

Source: European Commission, Directorate-General (DG) for Taxation and Customs Union, based on Eurostat and OECD data

Figure 9: Evolution of EU28 VAT Gap 2014-2019

Source: Centre for Social and Economic Research & Institute for Advanced Studies: Study and Reports on the VAT Gap in the EU-28 Member States: 2020 and 2021 Final Reports 25  

All 11 interviewed Fiscalis EU national coordinators 26 believed there is tangible evidence that Fiscalis contributed to the objective of fighting tax fraud, tax evasion and aggressive tax planning. More than half of them saw MLCs as an effective action that produces tangible outcomes. As was explained eight out of fourteen Fiscalis coordinators indicated that MLCs often generate additional tax revenues or uncover tax fraud cases, therefore producing tangible results in the form of additional tax revenue and/or prevented/uncovered tax fraud. Three national coordinators mentioned specific EISs as very effective. In their view, the Automatic Exchange of Information under DAC was very effective as it is being used in MLCs to solve complex cross border business taxation case relations. Data on incomes or financial bank accounts of entities in other countries help national tax administrations perform joint MLC audits.

All 14 interviewed EU and candidate countries Fiscalis coordinators stated that the same results could not have been achieved without Fiscalis. They especially highlighted EISs as the main activity that Member States could not have implemented on their own. Two interviewees mentioned the Automatic Exchange of Information under DAC, stressing that a unified approach to sharing data would have been impossible without Fiscalis’ support.

On the basis of the evidence collected, it can be concluded that overall, Fiscalis has contributed to fight tax fraud, evasion and aggressive tax planning by financing EISs, joint actions and training that helped tax authorities better avoid, detect and combat fraudulent activities. EIS play an important role in fighting tax fraud, evasion and aggressive tax planning. MLCs also generated tangible effects according to the MLC coordinators, and their effectiveness and efficiency is improved by the EISs for information exchange. Thus, Fiscalis via EISs complemented by joint actions has helped contribute to the fight against tax fraud, evasion and aggressive tax planning though the extent of this contribution is not possible to quantify as it is a supporting instrument and it is difficult to disentangle from policy action.

144.1.4.    To what extent have Fiscalis 2020 resources produced the best possible results at the lowest possible cost?

When evaluating efficiency, the external study looked at the extent to which allocated programme resources were worthwhile in achieving the intended results. The programme-level analysis relied on cost data (for EIS, joint activities, training activities) provided directly by DG TAXUD or extracted from the Annual Progress Reports, which was compiled and used to answer the relevant evaluation questions. It was complemented with evidence on perceived benefits collected through the programme-level stakeholder consultations and the case study findings.

The methodology for evaluating the efficiency of each specific IT system was adapted in in each case study to the specificities of each system (based on their theory of change). Depending on the availability of reliable data, an attempt was made to estimate the costs and benefits, and where possible to monetise them. The latter was only possible to a certain extent for the MOSS, DAC1/DAC2 and TNA case studies. DG TAXUD primarily supplied data on costs. In the case of MOSS and DAC1/DAC2, it was complemented with cost data from previous studies and data provided via interviews, albeit by a small sample of stakeholders. To report on the benefits of each IT system, the analysis relied mainly on interviews with relevant stakeholders. In the case of TNA, this was complemented by a survey, and in the case of DAC1/DAC2 by results of the yearly assessment questionnaires of the Automatic Exchange of Information. The assessment on the ratio between costs and benefits was made by relating to the likely costs of a counterfactual scenario in which the IT systems did not exist, mainly quantitatively. More detail on the methodology followed in each case study is presented in each case study report.

As shown in Figure 10  below, EIS made up a large majority (around 75 % per year) of the Fiscalis 2020 budget. Overall, the resource allocation and division remained relatively stable, and most resources budgeted were committed as expected (see Table 11 ). The total budgeted expenses for the 2014-2020 period were EUR 222,666,000 and those committed – EUR 222,085,746.

Figure 10. Committed budget per type of activity, in EUR million

Source: Fiscalis 2020 Annual Progress Report 2020

Table 11: Committed vs budgeted expenses in EUR - Fiscalis 2020 27  

Source: Fiscalis 2020 programme Progress Report 2020, Annual Work Programmes 2014-2020

154.1.4.1.    Costs compared to benefits associated with Fiscalis IT systems 

The overall cost for Fiscalis 2020 EIS was EUR 168 million over the period 2014-2020. The development costs (i.e. software development) amounted to EUR 19.6 million (14 %, Fiscalis EIS budget). Additional system costs for testing, infrastructure, maintenance, operations and quality assurance are estimated at EUR 93.2 million (65 %, Fiscalis EIS budget). This is much higher than development costs because many of the systems existed before 2014 and operating costs (infrastructure, data centres maintenance and other expenses, testing and quality assurance) are often more resource intensive. Note that cost details are only available as total costs for the Fiscalis and Customs programmes together. During 2014-2020, these were about EUR 233 million (EUR 33 million for functional and conformance testing and Member State support; EUR 155 million on infrastructure, maintenance and operational costs; and EUR 45 million on quality assurance activities). As 27 % of the IT systems are for Fiscalis 2020 and 26 % are common to both programmes, one can associate around 40 % to Fiscalis 2020. Due to this sharing of contracts between Fiscalis 2020 and Customs 2020, there is no breakdown of costs per IT system, and we cannot provide an overview of the resources allocated to Fiscalis 2020 IT systems per specific objective.

Member States also incur costs for acquiring, developing, installing, maintaining and day-to-day operating of the non-Union components of the Fiscalis IT systems. These costs vary significantly between countries and systems, based on the IT infrastructure already in place, the size of the country where it is being implemented and how their tax administration is organised, the local costs of IT support, and so on. It is therefore difficult to estimate the costs associated with this aspect of participating in the Fiscalis 2020 programme, and interviewed coordinators were also unable to provide such estimates. Box 5 presents IT systems cost that were assessed in case studies. They cannot be taken as a representative for other systems and should be seen as examples due to the considerations outlined above.

Box 5: Examples of Member States’ Fiscalis IT system costs, based on case studies

Other stakeholders (e.g. businesses, financial institutions) can indirectly incur costs for Fiscalis IT systems. For the MOSS, in addition to compliance costs for the 2015 VAT rule changes, a 2016 study estimated the cost to start using the MOSS system. The costs resulted from required changes to billing software logic, capturing additional transactional information, website changes and separate report requirements for MOSS returns 29 . Based on the answers received, the study estimates that the average cost for setting-up the IT systems amount to about EUR 1.2 million. However, there are very large variations across businesses, as the costs identified vary from EUR 8 000 to EUR 10 000 000. As most of the data were provided by large businesses, the relatively high average costs (as well as some of the higher values reported) are most likely related to the size of the businesses. Indeed, some of the businesses provided aggregated data for the entire group (thus including group companies). Such aggregated figures could explain the high variations across countries and some of the highest figures.

Under DAC2, financial institutions are obliged to share their clients’ bank account details with the national tax authorities, which required setting-up systems for collecting the data and clients’ agreement to share it. This enables an automatic information exchange on financial account data among EU Member States. The costs borne by financial institutions cannot be provided, as there is no available data, but estimated for the Polish financial sector at EUR 4.4 million.

As reported previously, there is clear evidence of the IT systems studied have contributed to the Fiscalis 2020 objectives. Many of the benefits are inherently difficult to quantify (e.g. the extent of tax fraud reduction and Fiscalis IT systems contribution; or tax authorities’ increased administrative capacity). Nevertheless, the case studies conducted as part of this evaluation revealed concrete examples of benefits outweighing costs for specific systems (see  Box 6 ). Taken together, the evaluation concludes that there is clear evidence of Fiscalis 2020 IT system benefits outweighing costs.

Box 6: Example of costs vs benefits for Member States of specific IT systems, based on case studies

164.1.4.2.    Costs compared to benefits associated with Fiscalis joint actions 

Fiscalis 2020 joint actions costs are available at a much more granular level. Expenditure on joint actions (seminars, workshops, project groups, bilateral or multilateral controls, other administrative cooperation activities, working visits, PAOEs and capacity building activities) excluding expert teams and studies amounted to about EUR 33 million. The largest share of resources was spent on project groups, followed by workshops and multilateral controls.

As the majority of the costs associated with participation in these joint actions is covered by the Fiscalis 2020 budget (travel costs, costs linked to the organisation of events, daily allowances and accommodation costs), Member States incur very little costs. This is slightly different for third country working visits, where Fiscalis covers travel and subsistence, but organisation costs are borne by the host countries.

Because of this, arguably, Fiscalis joint actions benefits clearly outweigh the costs incurred by Member States. However, this might not be the case for candidate and potential candidate countries, as two interviewees suggested. One explained that an evaluation showed their financial contribution to Fiscalis as high considering the extent to which they can participate. Although they see value in participating, they would like access to more aspects of these actions.

In general, the average cost per participant has remained relatively stable with a slight overall decrease, which might signal improved efficiency (see  Figure 11 ). There are no clear differences for different types of joint actions.

Figure 11: Average cost per event and per participant

Source: ART data

Fiscalis programme poll results each year are overwhelmingly positive, suggesting actions’ benefits are high. The majority of tax authorities interviewed for this evaluation agreed that the benefits outweighed costs. Therefore, it is safe to conclude that joint actions investments were considered worthwhile and cost-effective when considering the costs versus the outcomes achieved.

Expert teams, in particular, seemed highly cost-effective. Participation in expert teams is not fully reimbursable by the Fiscalis 2020. While travel and subsistence costs are fully reimbursed, personnel costs are only partly covered (the Fiscalis budget covers 50 % of administrations costs). These expert groups were crucial in the development of some of the Fiscalis IT systems which might not exist in the current setting without them (e.g. TNA, TEDB).

To illustrate further the cost-effectiveness of the Fiscalis joint actions, concrete examples from some of the case studies are presented in Box 7 below.

Box 7: Costs vs benefits of specific joint actions

174.1.4.3.    Training activity costs compared to benefits 

In the period, EUR 6.9 million was committed for common tax training, compared to EUR 8.7 million budgeted (absorption rate, 80 %). The majority of the Fiscalis 2020 training budget (EUR 4.7 million) was allocated to eLearning modules. These modules trained 130,470 individuals from tax authorities and 941,938 additional people from businesses and public. Thus, the training modules cost EUR 4.9 per person trained.

Based on the most recent eLearning satisfaction survey, the courses are highly appreciated. Both national administrations and individual users have reported high quality scores on the content, methodology and technology 32 . This was confirmed in this evaluation. The majority of interviewed tax authorities described eLearning as useful and effective in increasing their capacity, as well as being helpful for other stakeholders. More broadly, other Fiscalis training activities taking place as part of project groups or capacity building activities (notably IT trainings) were also considered to have directly increased their administrative capacity and facilitated tax authorities’ coordination and cooperation.

184.1.4.4.    Scope to maximise efficiency 

Fiscalis 2020 mid-term evaluation judged the overall structure and processes for managing the Fiscalis programme as broadly appropriate. Given that they have not recently changed, it can be assumed that this holds true, and it was not a focus of this evaluation. Further, the mid-term evaluation identified ways in which Fiscalis 2020 actually reduced the Commission and participating countries’ administrative burdens, e.g. introducing unit costs for accommodation expenses, which resulted in cost and time savings.

However, it identified concerns about the paperwork required to join, initiate and provide feedback on joint actions. Also, limited human resources at national level to adequately respond to requirements led to recommending DG TAXUD to optimise procedures and resources for joint actions and keep workload manageable. This appears to have improved in the latter half of Fiscalis 2020, as no administration considered the burden too high compared to their benefits.

No concrete inefficiencies or possibilities to further simplify or reduce the burdens were identified. There are always certain barriers or burdens, for involved, but generally, tax authorities feel that these are balanced with the programme outcomes.

194.1.5.    To what extent is Fiscalis 2020 coherent with broader EU policies

Under the coherence criterion, we analyse the extent to which the Fiscalis 2020 programme is coherent with broader EU policies. As a first step, this section addresses Fiscalis 2020 coherence with the Commission broader strategic framework, especially in light of changes in Commission priorities since the mid-term evaluation. We then further analyse the relationship between Fiscalis and similar EU funding programmes, as well as its coherence with measures taken in non-EU Member States.

As noted during the Fiscalis 2020 mid-term evaluation, its programme action was aligned and coherent to the Commission overall strategic priorities, especially to the Europe 2020 strategy.

Looking at the Commission revised objectives for the 2019-2024 period, a similar conclusion can be drawn. Under the Commission priority “An economy that works for people”, the Commission aims at “…tackling tax abuse, ensuring sustainable revenues and supporting a better business environment in the Single Market.” The Fiscalis achievements under its specific objective to fight tax fraud, evasion and aggressive tax planning (see section 4.1.3) clearly aligns with this Commission strategic objective. Further, the Fiscalis 2020 funded actions are coherent with the Commission priority “Europe fit for the digital age”, under which the Commission aims to modernise public administrations as in the Commission’s eGovernment Action Plan. Actions under the Fiscalis specific objective to enhance administrative capacity of participating countries, especially the development, improvement and implementation of EIS will support this priority.

In 2020, the Commission released the Package for fair and simple taxation containing 25 taxation initiatives to support, the recovery from the economic disruption caused by the COVID-19 crisis and the transition into a green and digital economy 33 . Most of the initiatives are still in their preparatory phases and thus fall out of Fiscalis 2020 scope. The new Fiscalis programme will be able to contribute via joint actions to many of these initiatives. Below are relevant initiatives planned for 2021-2023:

·Amendment of 2011/16/EU to include exchange of information on income generate through share and gig platforms (DAC7) and crypto assets (DAC8);

·EU Initiative on VAT in the digital age, to amend the existing VAT directive in line with digital developments 34 ;

·Actions to further develop the Eurofisc network.

Fiscalis contributed to implementing two of the Commissions’ 2019 political priorities, demonstrating its continued coherence with the Commissions’ revised political priorities. While the context of the new Commission has changed since the mid-term evaluation, the overall conclusion that it corresponds to the overall political strategy and objective of the Commission remains valid.

There also appears to be coherence between Fiscalis 2020 and other similar instruments. Complementarities and synergies exist between the Fiscalis 2020 and the Customs 2020 programme. As established in Article 5 of Regulation (EU) 1294/2013, the general objective of the Customs 2020 Programme is to “support the functioning and the modernisation of the customs union in order to strengthen the internal market […]”. Since DG TAXUD manages both programmes with similar management structures, this achieves efficiency gains in managing the Commission resources. Further, both programmes share many operational features that are co-funded by both programmes and available for customs and tax administrations. The programmes jointly funded key IT infrastructure such as CCN and EU common training materials. This leads to efficiency gains, as it avoids implementing separate systems, and enables customs and tax administrations to exchange information and share knowledge better.

The Structural Reform Support Programme (SRSP) 2017-2020 35 is complementary to Fiscalis 2020. Until 31 December 2019, the Structural Reform Support Service of the Commission managed it, replace by DG REFORM as of 1 January 2020. The SRSP offered tailor-made expertise to all EU Member States to design and implement institutional, administrative and growth-enhancing reforms in different policy areas including revenue (tax and customs) administration. The support ranged from strategic and legal advice, to studies, trainings and expert visits on the ground in any phase of the reform process. The support providers under the SRSP consisted of the European Commission in-house experts, EU Member State public organisations and individual experts, international organisations and private consultancy sector.

Regardless of the different objectives, and implementation modalities, synergies exist between both programmes. SRSP complemented Fiscalis 2020 by deploying tailored technical support for the programmes and bringing expertise from external providers such as International Monetary Fund (IMF), World Bank (WB) and Organisation for Economic Co-operation and Development (OECD), private consultancy companies and public organisations. Commission interviewees confirmed that synergies between the SRSP and Fiscalis 2020 exist. In their view, the SRSP could assist a country to undertake more structural reforms to tackle problems discovered during Fiscalis joint actions. For example, Fiscalis 2020 was used to kick-off technical assistance for the Greek tax administration before the launch of SRSP in 2017, which was later continued under Technical Support Instrument (TSI).

Complementarities between Fiscalis 2020 and the Hercule III programme managed by the EU Anti-Fraud Office (OLAF) also exist. Article 4 of Regulation 250/2014 laying down the provisions establishing the Hercule III programme indicates that its specific objective is “…to prevent and combat fraud, corruption and any other illegal activities affecting the financial interests of the Union”. Several of the actions eligible for funding under Fiscalis are similar to those under the Hercule III programme. An in-depth analysis and comparison of the Hercule III and Fiscalis’ work programmes revealed no overlap between the two, as the former did not include activities focusing primarily on tax fraud, tax evasion, and aggressive tax planning. On the contrary, one could argue that Hercule III is complementary to Fiscalis, as the fight against fraud conducted with Hercule equipment contributes to the fight against tax fraud and evasion. Despite the complementarities discovered between the programmes, there is no evidence of active cooperation or actively sought synergies. In line with the mid-term evaluation findings, synergies could be further explored through aligning management and planning of the two programmes more closely. Both OLAF and TAXUD receive each other's draft annual work programmes for interservice consultation, as a means to avoid or identify potential overlaps and to seek synergies wherever possible within our respective mandates.

The Fiscalis programme also plays a role in and is coherent with programmes for candidate and potential candidate countries. Seven non-EU Member States participated in Fiscalis 2020 via joint actions and training activities. During Commission and Fiscalis coordinator consultations, in three non-EU Member States, aspects of coherence between Fiscalis and other programmes in non-EU Member States were identified. Similar to the Commission’s TSI, the European Commission Technical Assistance and Information Exchange instrument (TAIEX) provides EU neighbourhood country administrations with support for their approximation or accession process to the EU. Through workshops, expert missions or study visits, knowledge and best practices are exchanged between candidate and EU countries, to drive reforms in non-EU Member States. Candidate country interviewees highlighted the benefits of access to both instruments, as Fiscalis is more flexible and provides short-term outcomes, while TAIEX enables non-EU Member States to pursue more long-term reform processes.

All non-EU Member States’ representatives emphasised the usefulness of Fiscalis activities in supporting accession negotiations with the EU. They highlighted the importance of its joint actions to gain EU taxation policy framework knowledge and experience, which is important for the accession negotiations. DG NEAR interviewees added the important symbolic value of including non-EU Member States in Fiscalis 2020 as providing an opportunity to establish trust and goodwill between the negotiation parties.

204.2.    How did the EU intervention make a difference (value added)?

This section reports on what Fiscalis achieved that likely would not have been achieved, at least not as effectively or efficiently, without it. In other words, to what extent is the Fiscalis 2020 programme a vehicle to achieve more than the Member States would do alone? As said, it is difficult to compare Fiscalis 2020 achievements to a counterfactual scenario without it, as its predecessor programmes have been laying the foundation for the achievements for a long time. This aspect is therefore covered from a qualitative, hypothetical point of view.

214.2.1.    EU added value of Fiscalis 2020 IT systems 

Fiscalis 2020 was a fit-for-purpose instrument (see section 4.1). Fiscalis funded IT systems supporting the implementation of legislative changes and harmonisation in rules, to help Member States complying with new legislation. Without it, Member States would have had more difficulties. As discussed, IT infrastructure and systems such as the CCN, DACs, e-Forms, and MOSS made complying more effective and efficient than it likely would have been without an EU-wide coordination mechanism. With EU tax law becoming increasingly important with increasing globalisation and movement of workers (see section 4.3), the programme’s added value is also likely to continue increasing.

Relatedly, without cooperation and coordination between Member States, unfair tax competition and tax shopping would likely increase, with fraudsters exploiting the lack of coordination across borders as ‘loopholes’ to avoid taxes. Tax fraud challenges cannot be tackled if Member States do not look beyond the borders of their administrative territories or cooperate intensively with counterparts, e.g. through MLCs, which tax authorities found particularly helpful (see section 4.1.3). It is likely that they would have a degree of bilateral or multilateral cooperation without Fiscalis 2020 and its IT systems. However, such approaches would be much less cost-effective, as this EU-wide approach created economies of scale with interconnected systems and consistent information exchange. Without the CCN system, for instance, Member States would have to link national systems 27 times, and each time potentially with different specifications, implying additional time and financial resource investments and lower security levels when exchanging sensitive data. In sum, without Fiscalis they would need to invest more resources and may have produced weaker results.

Box 8: Specific IT systems’ EU added value

Another, perhaps unintended added value of the Fiscalis IT systems is that it helped national tax administrations to modernise and digitalise their own internal systems and tools (see section 4.1.2), notably for countries which have more recently acceded to the EU.

224.2.2.    EU added value of the Fiscalis 2020 joint actions 

Joint actions provided considerable EU added value. Expert teams such as those on the Excise Duty Calculator and TNA (Operation teams I and II) were crucial for preparing and implementing the related IT tools. Without Fiscalis financial support for Member States for these initiatives, they likely would not have been developed. The programme generated innovative ideas, developed by experts in Member States, who were largely compensated for their time via grants 36 . Other joint actions such as workshops and seminars involving many actors, but also smaller scale ones like working visits and MLCs, helped tax authorities establish networks across Member States, and facilitated knowledge sharing. Judging from the consistently positive feedback on joint actions’ ability to meet their objectives (see section 4.1), it is safe to assume that they have had a positive impact. Without Fiscalis 2020 funding and (perhaps more importantly) coordination, most likely there would have been fewer events. Tax authorities voiced being faced with limited resources and would not have had the time or money to invest in planning, organising and carrying out such actions without Fiscalis.

In fact, the networks and knowledge sharing benefits go far beyond Fiscalis-funded joint action confines. One Fiscalis coordinator explained the networks created through joint actions have also opened up informal cooperation. In their country, there were ongoing tax reforms designed with the help and advice of coordinators from other countries. An interviewee from a non-EU country shared that they hired an Austrian tax expert they met through a collateralised loan obligations (CLO) joint action, as a consultant to help solve internal issues. These connections, especially with non-EU countries, would not have been likely without Fiscalis-funded actions.

234.2.3.    EU added value of the Fiscalis 2020 training activities 

As common training activities are closely related to Fiscalis IT systems, and understanding and implementing taxation Union law, their EU added value lies primarily in how they improved national tax administration officials’ knowledge and capacity and thus the effectiveness and efficiency of administrations. Fiscalis-funded eLearning modules added value beyond the participants as they were also used internally to teach for example. The large number of people trained (e.g. just below 350 thousand, in 2020) 37  is evidence that they were considered useful. These training activities and modules would likely not have been developed if Fiscalis had not been in place.

244.2.4.    Likely impact if hypothetically Fiscalis was discontinued

Based on the evidence presented, it is fair to say that without Fiscalis we would not have the IT systems, joint actions and training to the extent that we have and that were instrumental in implementing EU taxation law and fighting tax fraud, evasion and aggressive tax planning. While there may have been bilateral or multilateral efforts, they would not have achieved the same level of results as Fiscalis and they would not have benefitted from the economies of scale generated by Fiscalis or the timeline would have been very different and slower.

These findings are supported by tax authorities across the board. When asked to imagine a hypothetical situation where the Fiscalis programme was stopped or discontinued, all agreed that it would have significant negative consequences. This includes discontinuing IT systems and joint actions due to a lack of resources. In their view, if Member States continued IT systems themselves, to develop and innovate them would be much slower. Joint actions would also not be possible on this scale if they relied on Member States’ own resources. Existing (bilateral) networks between officials in different countries would continue to be exploited, but not as easily renewed in case of e.g. staff changes.

Some interviewees pointed to collaboration with the OECD or the International Organisation of Tax Administrations (IOTA) as an alternative framework. However, they stressed that it would reduce harmonisation and EU-specific benefits such as implementing common EU tax law, because there would not be the crucial IT systems to support it.

254.3.    Is the intervention still relevant?

This section looks at the Fiscalis programme’s relevance by exploring ‘To what extent does the Fiscalis 2020 continue to respond to tax administrations, businesses and the general public needs. It elaborates the degree to which the Fiscalis 2020 proved and continues to be relevant in achieving its overall aim of improving the proper functioning of taxation systems in the internal market by enhancing participating countries tax authorities’ and officials’, cooperation. It explains the extent to which Fiscalis 2020 was able to tackle new and future challenges and needs stakeholders faced.

264.3.1.    Challenges and priorities in the field of taxation 

The fight against tax fraud, tax evasion and aggressive tax planning is an ongoing challenge that results in significant revenue losses in the EU every year as the VAT gap figures attest. Globalisation and digitalisation combined create new challenges to tax systems ability to collect tax revenue by creating potential for tax avoidance or evasion if tax administrations and systems do not keep up with change.

A range of taxation legislative change was adopted during the 2014-2020 period to keep up with the increasingly digitalised, globalised world. Fiscalis 2020 actions were closely linked to these developments. For instance, the 2015 VAT place of supply rule changes were accompanied by the introduction of the MOSS system to facilitate cross-border trade in digital services. The DAC directives were introduced to facilitate the exchange of information between tax administrations, to ensure the fair payment of taxes for taxpayers (individual or entities) who generate income in cross-border activities. Technological advances such as the rise in Artificial Intelligence led to innovations such as the TNA.

Looking to the future, a relevant trend related to all these aspects is the rise in e-commerce. As part of Fiscalis 2020 and building on lessons from its implementation, notably the MOSS system, preparatory work was done to introduce OSS and the IOSS, which became operational on 1 July 2021.

Moreover, the ongoing COVID-19 pandemic has brought numerous changes in existing working patterns, such as normalising remote work. Hybrid remote working arrangements seem to be here to stay 38  which could potentially lead to an increase in the movement of workers across Europe in coming years.

While tax matters remain Member States’ competence, such structural changes accentuate the need to reinforce tax administration cooperation and national authorities’ administrative capacities. In other words, EU action to facilitate Member States’ communication, cooperation and coordination for the internal market effective and efficient functioning remains relevant in this context. Given Fiscalis 2020 success in addressing these aspects in the past, the continuous improvement of IT systems and providing joint actions and training will continue to be relevant.

As discussed in the following subsection, there is wide consensus among key stakeholders that the Fiscalis 2020 is necessary, relevant and corresponds to key actors’ needs. Findings confirm that the programme is engrained in the EU tax policy landscape and is now one of tax administrations’ and economic operators’ main work-tools.

274.3.2.    Extent to which Fiscalis 2020 met tax authorities’ needs 

The evaluation confirms Fiscalis-funded activities relevance in addressing challenges tax authorities identified in the mid-term evaluation.

Firstly, EISs are reported to be highly relevant for tax officials’ daily work, as they are the main information exchange platforms, which in turn fosters mutual assistance and coordination. In light of the developments just outlined, national tax administrations emphasise coordination between EU Member States as important for adequately applying EU law and fighting tax fraud and evasion. In this regard, the majority of interviews with tax authorities and national coordinators confirmed the relevance of Fiscalis-funded EISs in setting up common platforms. They serve as communication platforms through which Member States can exchange relevant data to fight tax fraud and evasion and implement EU law.

Training and joint actions are also considered relevant to enhance administrative capacity such as knowledge building and networking. Joint actions, particularly working visits and project groups, were mentioned repeatedly as highly relevant. Through the exchange of information and good practices, Fiscalis supports policymaking and the equal and predictable application of EU law, as tax administrations can work together to develop and implement taxation policy. For instance, the Fiscalis tax gap project group (FPG/041), established in 2016, gathered 15 national experts from different Member States to pool knowledge and share experience of estimating tax gap, essential for shedding light on tax evasion and tax fraud. The project group’s work resulted in reports 39 to guide tax gap estimations and enhanced coordination. An interviewee also highlighted Fiscalis role in developing dialogue spaces where EU tax officials responsible for operational tasks, harmonising procedures and audits could exchange good practices and guidelines. The TNA also offers another good example of where an identified need and idea were developed and materialised through several project groups. Overall, the evidence suggests that the Fiscalis 2020 programme allowed tax administrations to have a space to exchange ideas with counterparts in other countries, identify needs and develop solutions in a joint manner.  

While all interviewees acknowledged the programme’s relevance, two interviewed stakeholders mentioned a number of gaps that could be addressed. For example, one coordinator mentioned that Fiscalis could support the creation of a database on Member States’ legislative developments in the tax field, and that it should be adjusted to better cover emerging developments affecting taxation such as the new OSS and IOSS schemes, virtual currencies, digital solutions to combat fraud, etc.

Overall, Fiscalis 2020 is highly relevant to addresses tax administrations’ major challenges and meet their needs. However, Fiscalis coordinators in candidate countries pointed out their limited capacity to benefit fully from Fiscalis-funded actions due to its EU-oriented nature. This adds to the fact that these countries do not have access to all IT systems, which affects their ability to take part in related training components.

284.3.3.    Extent to which Fiscalis 2020 met economic operators’ needs 

Securing a level playing field and ensuring fair competition is key for economic operators to perform their activities. Moreover, excessive administrative burden and tax compliance costs can prevent some businesses from carrying out economic activities, particularly across borders. For example, some economic operators, particularly SMEs, can potentially be discouraged from operating beyond their borders due to different VAT compliance requirements across countries. As above, recent legislative reforms were established to update and adapt VAT rules on cross-border business-to-consumer e-commerce activities.

In particular, Fiscalis IT systems are especially relevant for economic operators. For instance, the MOSS proved to be very relevant in alleviating the administrative burden on companies providing telecommunications, broadcasting and electronic services by avoiding filing VAT returns in each country where services are supplied, which improved tax compliance. The SEED facilitated economic operators’ activities by simplifying compliance via making information easily available.

294.3.4.    Extent to which Fiscalis 2020 met public needs 

Although Fiscalis was not designed explicitly to address citizens’ needs, one could say it is indirectly relevant to the public. Fighting tax fraud and evasion and ensuring tax compliance is a shared concern. As illustrated by the 2016 Eurobarometer survey, 75 % of respondents reported a preference for major EU intervention in fighting tax fraud while only 5 % reported that the EU should intervene less.

This should not be surprising, if EU intervention in fighting fraud is more successful, tax revenue will increase (a trend observed in recent years). Moreover, supporting tax administrations cooperation and strengthening their capacities will result in more effective and efficient procedures and potentially reduce citizens’ compliance costs.

305.    Conclusions and lessons learned

This chapter presents key conclusions and lessons learned based on the findings outlined in the preceding chapter. It entails a discussion of the evaluation findings and the potential link between the results achieved and the more long-term impacts.

Fiscalis 2020 contributed to enhancing tax authorities’ administrative capacity through IT systems, joint actions and common training activities, harmonising approaches, simplifying procedures and lowering administrative burdens.

Fiscalis 2020 enhanced tax authorities’ administrative capacity through IT systems, joint actions and common training activities. Joint actions focused on sharing administrative guidelines and procedures as well as working practices. Participants reported a high degree of satisfaction with these actions and considered them useful. On average 77 % declared that an administrative procedure, working practice, or guideline derived from Fiscalis 2020 led to a positive change in their administrations’ working practices. More than 130,000 tax officials were trained through Fiscalis common training activities. The numbers increased over time, and participants reported a high degree of satisfaction with the usefulness of these trainings.

IT systems used for information sharing also played a role in this, by reducing administrative burdens. They reduced the time and resources needed to carry out activities such as exchanging information or files; identifying taxpayers or economic operators; collecting and analysing data. The IT systems therefore enabled efficiency gains and the shifting of scarce resources to other important tasks. An example of this is the TNA - by making use of advanced data analytics to spot potential cases of fraud, TNA reduces the manual work associated with exchanging data and detecting potential cases of fraud for Eurofisc Liaison Officials. Although they still need to qualify and follow up on those potential cases and may not have less work than before, their resources are now more focused on qualified investigations, and the potential to identify actual fraud cases in the same amount of time is maximised.

Fiscalis 2020 reduced the administrative burden and compliance cost for taxpayers through its IT systems

The main target group for Fiscalis is the national tax administrations in the Member States, and most of the activities, and the IT systems, serve to improve exchange of information between the authorities. Some Fiscalis IT systems are directly relevant to taxpayers by improving access to information such as TEDB, VIES and TIN. Using Fiscalis 2020 funding, the MOSS was developed and implemented to enable businesses supplying telecommunications, broadcasting and electronic services to file only one VAT return in one Member State, rather than with several different tax administrations in different countries. Although they incurred compliance costs initially, those were largely outweighed by the benefits from using the system. It has been estimated that the costs for businesses using MOSS are about 95 % lower than for businesses not using it (EUR 2,200 vs. EUR 41,626 per business per year). Even if this is overestimated, the cost savings are likely significant. The same trend will likely be observed with the new OSS, introduced on 1 July 2021. Furthermore, the experienced gained from MOSS have been instrumental to the new OSS, thus supporting the practical implementation of and compliance with the rule change taking place.

Through IT systems and a variety of joint actions including training, Fiscalis 2020 was assessed to have contributed to the improved functioning of the EU internal market by supporting the fight again tax fraud, tax evasion and aggressive tax planning.

Estimating and measuring tax fraud, tax evasion and aggressive tax planning is inherently difficult since taxable persons or entities deliberately try to hide the information to avoid paying taxes due. Still, the research undertaken for the evaluation study showed that several of the key indicators on tax fraud, tax evasion and aggressive tax planning show signs of improvement. Since 2014, the VAT gap has reduced. The focus on addressing offshore wealth has increased as can be seen through the large volume of leaked documents (as recently as October 2021 with the leak of the ‘Pandora Papers’).

These developments are difficult to link directly to actions funded by Fiscalis, but there is an indirect link, and the evaluations findings show that Fiscalis contributed towards these observed trends as exemplified by the following evaluation findings:

·The TNA IT system supports the identification of potential fraud cases, making use of advanced data-analytics so the process is more efficient than manual checks. Initial estimates show that the fraudulent or suspicious transactions uncovered amounted to EUR 5.3 billion in 2020 and a total of 1814 missing trader fraud cases were identified 40 . It is worth noting that TNA supports Eurofisc Liaison Officials to detect potential cases of fraud. The collected evidence suggests that TNA enables more potential fraud cases to be identified, which can then be subsequently addressed; 

·The MOSS facilitated the filing of VAT returns for businesses providing telecommunications, broadcasting and electronic services. The system significantly reduced the time and effort required for this, compared to a situation in which VAT returns would have to be submitted in several Member States. Although difficult to make a direct link, this reduced burden could have increased compliance with VAT rules, thus reducing the incidence of tax evasion by businesses supplying telecommunications, broadcasting and electronic services; 

·Through the existence of the DAC framework, there is more sharing of information on income declared and taxes paid across borders, making it more difficult to evade taxes. The system also acts to increase compliance, thereby preventing tax avoidance. As confirmed during the interviews with tax authorities, at least two Member States pre-fill tax returns based on DAC information, thus reducing the time needed for taxpayers to complete them and avoiding mistakes in filling in tax returns; 

·In addition, capacity building activities such as workshops, seminars, trainings and working visits which target topics and activities aiming to prevent the occurrence of fraud and tax evasion through knowledge sharing.

Fiscalis 2020 contributed to a coherent implementation of EU tax law through information sharing and knowledge transfer

Fiscalis is a financial instrument that aids understanding and implementing EU taxation policy and not a policy programme. Evidence suggests that Fiscalis 2020 contributed to helping Member States understand and implement rule changes, as well as learn and share best practices for harmonisation. For instance, MOSS was introduced simultaneously with the 2015 VAT place of supply rules on telecommunications, broadcasting and electronic services. It introduced a system through which tax authorities could share VAT information in a coordinated and centralised way, rather than exchanging information bilaterally. MOSS thereby facilitated and simplified the implementation of the 2015 rule changes; this would have been less efficient if there was no EU-wide system to do so. Another example is the DACs. The DAC Automatic Exchange of Information system relies on the CCN, which is required for Member States to be able to fulfil the requirements of the DACs. Without the CCN system, it would be challenging for tax authorities to securely and automatically exchange information.

The IT systems were developed and implemented with Fiscalis 2020 funding (and earlier iterations of Fiscalis) and bring significant added value for Member States. Without Fiscalis, there would be less information sharing and knowledge transfer, and thus less harmonisation of rules and approaches. In that sense, Fiscalis can be considered a key instrument for the implementation of EU tax law.

Fiscalis will likely continue to be a relevant tool to enhance taxation systems at EU level and should continue to be funded

Fiscalis remains a relevant instrument to fund the development, implementation and operation of IT systems in the field of taxation. The operational Fiscalis IT systems are quite diverse, but all aim at achieving Fiscalis objectives and the case studies produced evidence that the IT structure and systems evaluated contributed to achieving these objectives. They enabled Member States to communicate and coordinate (CCN, DACs, eForms, EMCS, TIC); reduced administrative burdens and enhanced economic operators’ tax compliance (MOSS, VIES, VAT Return); helped to detect and combat fraud (TNA); or simply provided information to feed into any or all of the above (TEDB, TIN, VoW).

The collected evidence indicated that joint actions fostered tax authorities’ cooperation and provided space to exchange ideas among counterparts if a specific need was identified. They created networks for participating tax authorities, enabling further coordination and knowledge sharing. Based on the findings, seminars, workshops and project groups, bringing Member States together, were considered to have been useful because they foster mutual learning and harmonisation in the implementation of legislation. TNA is a good example of an IT system, which was developed and governed as a result of the cooperation of dedicated tax officials through a series of project groups. MLCs were also useful for administrative cooperation, particularly in fighting fraud. Other actions with a narrower scope, e.g. working visits and PAEOs, allowed participants to focus on sharing good practices on specific IT system activities, thus fostering mutual learning.

The evidence suggests that expert teams could play a decisive role in developing new IT systems, though they are relatively new and relevant for a limited number of EIS. TNA and the Excise Duty Calculator are key examples of systems that might not have existed if Fiscalis 2020 did not provide funding for expert teams. In light of capacity constraints, and the significant time investment for these expert teams, Fiscalis funding is crucial, especially because Fiscalis funds not only costs directly associated with expert team meetings, but also 50 % of staff costs of participants (100% now with the new programme). The expert teams saved tax authorities time and money, and increased quality of IT solutions, stemming from working together in a more synchronised, coordinated and collaborative way. In addition, as participation is voluntary, expert teams are an example of true engagement on the side of authorities. In the end, all Member States can benefit from the outcomes of expert teams, even if they were not involved. Considering the resource investment required, Member State participation in expert teams shows their commitment and solidarity to improving the taxation system for the EU as a whole.

The resources invested into Fiscalis 2020 and its components were worthwhile considering the achieved results

EIS traditionally take up the largest share of Fiscalis funding (around 75 % each year), because they are expensive to develop and operate. However, they can help Member States collect additional tax revenue, which offsets the investments. In the case of MOSS, the VAT revenue collected through the system grew at a much faster rate than the overall VAT revenue growth (+87 % for MOSS compared to +17 % for VAT overall between 2015-2019). This is due to the general increase in cross-border B2C supplies of telecommunications, broadcasting and electronic services, and all four consulted tax authorities reported that the system significantly reduced burdens for them in collecting VAT revenue for telecommunications, broadcasting and electronic services compared to a hypothetical situation in which the system had not existed. For the DAC1/DAC2, an estimated additional EUR 1.6 million in tax revenue per consulted Member State during 2020 alone 41 . It is important to note that most of the costs borne by the Fiscalis programme in the field of IT consist mainly in interconnecting 27 heterogeneous national IT systems via exchanges of messages instead of developing central solutions based on data sharing. This architectural decision follows the agreed legal basis of the corresponding Taxation European Information Systems. However, it can lead to higher costs at EU level while limiting the capability of data processing with a possible impact on the cost / benefits ratio.

Joint actions are a relatively small financial investment around 15 % of the Fiscalis budget, compared to the EIS. As outlined above, the evidence suggests that they were very well received by participants. They enabled knowledge sharing, mutual learning and capacity building, which in turn can led to efficiency gains for authorities. The benefits clearly outweigh the costs for tax authorities because their participation is largely covered by the Fiscalis budget. This is not the case for non-EU countries, however. The interviews with non-EU participants highlighted the perception that their ability to participate is lacking compared to their financial contribution. They do not have access to most of the IT systems as they are designed to share data within the EU, and since many of the joint actions are linked to the IT systems, these are perceived as less relevant for non-EU countries. However, participation may increase if their awareness of activities that could be relevant for them to participate in is improved.

Common training activities made up only around 5 % of the Fiscalis budget and were used by more than 1 million people. As mentioned above, they not only benefited tax administrations, but also taxpayers. At an average cost of less than EUR 5 per person trained, paired with their perceived usefulness, training activities can be considered cost-effective although their actual impact cannot be estimated.

Fiscalis 2020 was coherent and consistent with EU priorities and it complements other funding programmes such as Customs 2020 and the Structural Reform Support Programme, though there is scope to improve synergies with Hercule III.

Fiscalis 2020 was aligned and coherent to the Commission overall political objectives, especially to the Europe 2020 strategy. It flexibly adapted to changing priorities and new legislation, fulfilling its role as a funding programme that supports the implementation of EU law on taxation.

Complementarities and synergies exist between the Fiscalis 2020 and the Customs 2020 programme. DG TAXUD manages both programmes with similar management structures, and the programmes share operational features that are co-funded by both programmes and available for customs and tax administrations. This promotes efficiency gains in managing resources. Complementarities between the programmes are likely to increase in importance in light of the recent adoption of the VAT eCommerce package. Nevertheless, a better discrimination of costs and benefits was seen as beneficial

Fiscalis is also complementary to the SRSP now TSI, which offered tailor-made expertise to all EU Member States to design and implement institutional, administrative and growth-enhancing reforms in different policy areas including revenue (tax and customs) administration. It complemented Fiscalis 2020 by deploying tailored technical support for the programmes and bringing expertise from external providers such as IMF, WB and OECD, private consultancy companies and public organisations. SRSP could assist a country to undertake more structural reforms to tackle problems discovered during Fiscalis joint actions.

Finally, Fiscalis is complementary to the Hercule III programme managed by OLAF, as they share certain objectives. The fight against fraud supported with Hercule equipment contributes to the fight against tax fraud and evasion. However, despite their complementarity, there seems to be limited active cooperation or actively sought synergies. Synergies could be further explored through aligning management and planning of the two programmes more closely.

1.Lead DG, Decide Planning/Work Programme reference.

The final (PLAN/2020/8201) and mid-term (PLAN/2017/933) evaluations are non-major initiatives organised in accordance with Article 17(1)(2) of Regulation (EU) No 1286/2013 of 11 December 2013 establishing a multiannual action programme Fiscalis 2020 for the period 2014 to 2020 to improve the operation of taxation systems in the European Union.

The lead Directorate General for the evaluation was the Directorate General for Taxation and Customs Union (DG TAXUD).

2.Organisation and timing.

The evaluation was informed principally by Final Fiscalis 2020 study conducted by an external contractor Ramboll. An inter-service steering group with the following Commission services who expressed their wish to participate in the group supported the study. However, elements of the mid-term evaluation were incorporated into this report also as was considered relevant and appropriate.

Table 12: Commission services that formed the final study’s ISG

Notes: x 2 indicates the service was also an ISG member for the mid-term study

The inter-service steering group for the final evaluation met four times on 18 June and 17 November in 2020 and 21 January and 8 July in 2021. The last meeting was a validation for Fiscalis 2020 programme during which the external consultants presented their findings, conclusions and recommendations. These were further on discussed with the group, their validity and credibility confirmed. The mid-term study was also supported by an inter-service steering group that met fives time over the life and finalisation of the project. The mid-term evaluation report was published 42 as well as an executive summary of the mid-term evaluation study 43 .

3.Evidence used together with sources and any issues regarding its quality

The present evaluation SWD is based on work carried out by external consultants. The material informing this evaluation consisted of (1) programming documentation, (2) previous studies, reports and evaluations, and (3) the Performance Measurement Framework (PMF), as detailed in Table 13 , which also presents the intended purpose of use.

The evaluations gathered also primary data on the programme’s implementation, functioning, results and use directly from the best-placed stakeholders: the tax administrations of the participating countries and, albeit to a smaller extent, from economic operators. Evidence was collected through targeted consultations and cases studies, details of how these were collected are in Annex II 0. 

Finally, the Commission organised a public consultation on the Commission’s spending programmes and funds. This public consultation was launched within the context of the Commission’s proposals for the next generation of financial programmes for the post-2020 Multiannual Financial Framework (MFF). The Fiscalis 2020 programme was included in the public consultation on investment, research and innovation, SMEs and single market. The findings of the consultation were incorporated into the mid-term stakeholder consultation report. The public consultation part is also Annex V of this report.

Table 13: Sources of information used for the evaluation of Fiscalis 2020

2Use of external expertise

The Commission carried out two evaluation studies using external consultants procured through a framework contract. The Fiscalis 2020 final evaluation study, which is the main study informing this report was conducted Ramboll Management Consulting 44 . This was signed October 2020 for a period not exceeding 12 months. The mid-term evaluation study was commissioned to Oxford Research, Coffey, Economisti Associati and wedoIT, with a consortium led by Economisti Associati Srl 45  for 13 months.

Approach

The final study had an evaluation approach that consisted of assessing relevance, effectiveness, efficiency, coherence and EU added value of the Fiscalis 2020 programme. To ensure this assessment’s robustness, a structured evaluation approach was used, relying on an elaborated intervention logic, operationalised evaluation matrix, and a mix of data collection and analytical methods. The overall approach included the following four main tasks:

·Task 1: Structuring and scoping

·Task 2: Fieldwork

·Task 3: Analysis and judgement

·Task 4: Recommendations

Evaluation questions

In line with the Commission’s Better Regulation Guidelines 46 , the evaluation criteria of relevance, effectiveness, efficiency, coherence and EU added value were covered. They were addressed through seven key evaluation questions:

These questions were operationalised and associated indicators and judgement criteria were developed, to provide a framework for collecting data and making evaluative judgements (see Annex III).

Overall methodological framework

The methodological approach consisted of a theory-based evaluation, which had two streams: one at the level of Fiscalis 2020 as a programme, and the other at the level of case studies concerning the specific objectives of the programme. The evaluation made use of a contribution analysis to uncover the Fiscalis contribution to the observed results, including an assessment of why observed results occurred (or not) and the role played by the intervention compared to other internal and external factors.

The figure below is a visualisation of the approach, which is described in more detail below.

Figure 12: Methodological approach

Source: Own elaboration

Programme level evaluation

At the programme level, we evaluated Fiscalis 2020 achievements as a whole as compared to its intervention logic (see  Figure 1 ), which outlines what it intended to achieve and how. Although a theory-based evaluation approach cannot assess causality and attribute observed changes to actual activities or initiatives, it can nevertheless demonstrate programme contribution by clearly identifying logical linkages in what is being observed. These serve as the basis for making plausible judgements on the level and scale of Fiscalis 2020 achievements.

The programme-level evaluation relied on desk research, interviews with stakeholders and the case study results.

The evaluation used contribution analysis to draw conclusions on the programme’s contribution to observed outcomes. This enabled structuring the analysis around information gathered, which helped generate hypotheses. These hypotheses were on the links between Fiscalis 2020 funded activities, the outputs they delivered (IT systems, joint actions, etc.), and the results and long-term impacts these have on beneficiaries, economic operators and society as a whole. The hypotheses were tested against the data collected. Crucially, it assessed potential alternative explanations to strengthen the evaluation validity by providing a sound analysis of how and to what extent these alternative explanations helped or inhibited the programme’s ability to achieve its outcomes.

The programme was split into manageable pieces for which individual theory of change models were developed, underlying assumptions hypothesised, and results measured. This was done through five case studies, each concerning one Fiscalis-funded IT tax system.

Data collection and analysis

Desk research

Programme-level

As the main aim of this study was to assess how well Fiscalis 2020 performed during the 2014-2020 period, the evaluation’s starting point was to assess its implementation, processes and results using the Performance Management Framework. The study relied on the Annual Progress Reports and raw data provided by DG TAXUD. On this basis, a comprehensive database of activities and associated outputs (e.g. number of joint actions organised, number of IT system users, level of awareness of the activities, etc.) was developed. This enabled comparing results over time and across activities.

Case studies

Programme-level data, previous studies and IT-specific documents (e.g. vision documents, business cases) were used to develop the theory of change for each IT system case study. DG TAXUD stakeholders verified these in dedicated meetings.

Stakeholder consultations

Programme-level

Consultations with key stakeholders aimed to quantify and explore existing evidence made largely available through the Performance Management Framework and provided additional qualitative evidence to answer evaluation questions. Notably tax authorities, who are the main target audience of Fiscalis 2020, were consulted. Commission officials as the implementers were also consulted. A number of videoconference meetings with DG TAXUD’s operational and IT unit staff were held during the inception phase. They provided a deeper understanding of the programme and its components, as well as data availability and case study selection insights. Additional interviews with Commission staff in DG TAXUD and DG NEAR took place during the fieldwork phase to further fill gaps and validate findings. An overview of the stakeholder activities is presented in Figure 12  below 47 .

As the mid-term evaluation was finalised in late 2018, to avoid consultation fatigue, Fiscalis coordinator interviews were only conducted with Member States that were not in the case studies. There were 14 Fiscalis coordinator interviews asking high-level questions about the functioning and usefulness of Fiscalis 2020. This number was considered appropriate, as national coordinators would likely not have a clear picture of programme details but a general view of their key challenges and how and to what extent Fiscalis 2020 helped address them. The Commission’s Better Regulation Guidelines require public consultations in evaluations however; as the public know, little about Fiscalis 2020 there was none.

Case studies

The stakeholder consultations were adapted to each case study’s IT system. The case studies included in-depth interviews with national tax officials in five Member States and DG TAXUD staff working with the systems, as well as other Commission staff (e.g. DG GROW in the context of TEDB), as necessary (see Table 14 ). For the TNA IT system, a survey was developed to reach a wider group of TNA users, who were also invited to provide additional information in follow-up interviews. For MOSS, SEED and TEDB, 9 (representatives of) economic operators were interviewed on the IT systems’ usefulness, effectiveness and efficiency.

Table 14 Overview of consultation activities

Data analysis

Programme-level

The programme-data analysis consisted primarily of descriptive statistics for the Fiscalis 2020 quantitative performance data, qualitative interviews coding analysis 48 and cost-effectiveness analysis to the extent possible with the available cost data. The analysis followed the evaluation matrix to structure and triangulate results from all the information collected.

Case studies

The theories of change were developed based on desk research and initial discussions with DG TAXUD. Relevant sources of information were analysed and used in the contribution analyses. Finally, stakeholder consultation and desk research information were assembled, analysed, using contribution analyses and written up, as case study reports. The data collected was mapped, coded and analysed to assess if and to what extent they confirmed or discounted hypotheses or assumptions of its underlying theory of change. The analysis was finalised by tracing the extent to which causal pathways from Fiscalis 2020 funded activities produced the intended outputs, results and long-term impacts.

Case study selection

To select five case study topics, we mapped out Fiscalis-funded IT systems’ key characteristics such as functionality, users, objective, cost and baseline situation. This was informed by DG TAXUD documentation and previous reports and evaluations.

Based on exploratory discussions with DG TAXUD programming and IT units, we selected five case study topics. This took into account considerations such as objectives and targeted users; development costs; and a mix of ‘older’ and recently implemented systems. The final selection consisted of the following EIS:

Table 15: Case study selection – description of the IT systems

Triangulation

Drawing on evidence from the case studies and the programme evaluation, answers to the evaluation questions were formulated. By means of triangulation of the information collected through desk research, interviews and case studies, the study aimed to overcome the weakness or intrinsic biases and problems that come from single method/observer/theory studies. Throughout the triangulation process, we referred to the evaluation question matrix for the analysis and judgement of each evaluation question, to ensure transparent and robust results.

Conclusions and lessons learned

The findings informed conclusions and lessons learned about the achievements and impact of different Fiscalis 2020 elements. The Fiscalis 2020 mid-term evaluation recommendations were also assessed, as presented in section 3.2.

Limits to the evidence base and robustness of findings

The evidence base is considered robust, and the generated findings valid and consistent. Nevertheless, there are inherent challenges in evaluating a programme such as Fiscalis, which should be highlighted and taken into account when reading the report.

First, Fiscalis 2020 is the continuation of several initiatives to support and increase administrative cooperation between Member States’ tax authorities, dating back to 1993. To some extent, Fiscalis cooperation has become “engrained” in EU tax authorities and its counterfactual (i.e. situation without Fiscalis) is challenging to construct. This is particularly true for IT systems in place a long time, such as the CCN/CSI, VIES and EMCS/SEED, where respondents have difficulty imagining implementing EU legislation without EIS support. This is an evaluation challenge but can also be seen as a testimonial of Fiscalis’ relevance and EU added value.

Secondly, quantifying Fiscalis 2020 costs and benefits proved difficult. Some EIS costs are shared with the Customs 2020 programme and costs per policy field or IT system could not be disentangled and attributed with certainty. The evaluation tried to work with estimations with DG TAXUD input. Benefits of fraud detected, or tax revenue generated proved difficult to obtain and monetise. This is because it is difficult to link such benefits to Fiscalis 2020 and there is a lack of Member State level information in part due to their reluctance to share sensitive information on combatting fraud and tax evasion.

Finally, the evaluation struggled to reach certain stakeholders. Respondents found it difficult to prioritise information requests, which is not surprising, as Fiscalis has two evaluations per cycle, requiring feedback from stakeholders every two to three years and potentially leading to consultation fatigue. Efforts to reach economic operators to assess Fiscalis 2020 simplifying effects were moderately successful, but still showed added value where it was possible (in particular for MOSS).

Stakeholder consultation activities – methodology and approach

This report presents the stakeholder consultation activities performed within the context of the “Study on the final evaluation of the Fiscalis 2020 programme”. The purpose of this report is to present the stakeholder groups consulted as part of the study and the consultation methods, topics covered, as well as the associated limitations.

The stakeholder consultation activities conducted as part of this study our outlined in Table 16  below.

Table 16 : Overview of consultation activities

Interviews

Programme level interviews

During the inception phase of the study, two exploratory interviews with the Fiscalis team within DG TAXUD were held. The objective of these interviews was to gain a deeper understanding of the programme and its components, as well as data availability and case study selection insights. An exploratory interview with DG NEAR staff was also conducted. At a later stage, exploratory interviews with operational and IT unit staff working directly with the EIS (European Information systems) analysed were conducted (see  Table 17 ).

Moreover, 11 semi-structured targeted interviews were carried out with Fiscalis programme coordinators in Member States that were not interviewed in the context of the case studies. Three Fiscalis programme coordinators in third countries (RS, ME, TK) were also interviewed.

Overall, 17 semi-structured interviews were conducted with relevant stakeholders in the EU institutions, as well as with Fiscalis national coordinators both in EU Member States and third countries. The table below presents a full overview of the stakeholders, which were interviewed as part of the programme-level evaluation. The data gathered through the interviews was coded, analysed and categorised into themes after which, triangulation with additional data sources, provided insights into the evaluation questions.

Table 17: Stakeholders consulted as part of the program-level interviews