Ongoing doctoral research

Some of ATA's members are currently preparing a PhD. Below is a list of the doctoral research being carried out at the moment - click on a title to view a summary of the research. Unless indicated ontherwise, the research is conducted in English.

Design of a flexible valuation tool for companies

This research started in 2020 and is sponsored by the company HLB Dodémont - Van Impe & Co cvba. Supervisor: prof. dr. Eddy Laveren.

With every purchase or sale of a company it is necessary that the buyer and the seller can make a good estimate of the value of the company. Moreover, one of the main causes of problems and conflicts in acquisitions and business successions is the lack of a reliable estimate of the company's market value. Each valuation is to some extent a subjective valuation that incorporates a certain vision of the future and the reaction of the competitors.The digital valuation tool allows you to obtain an idea of ​​the market value of a company on the basis of accounting data and supplemented with a number of crucial parameters regarding the future evolution of the company's financial condition. Valuation parameters are to a large extent entered by the users. The users are given the opportunity to design multiple scenarios and to calculate their impact on the market value of the company according to the usual common valuation methods. When calculating the value of a company, attention must also be paid to the tax aspects of the valuation.

Transfer pricing framework in a digitalized economy

Doctoral research by Mariya Otto and supervised by Anne Van de Vijver en Bruno Peeters. Started on 10 March 2020.

Our international profit allocation rules, as refined by the BEPS-project, are based on the paradigm of ‘value creation’, i.e. the allocation of income (and thus taxation) should be aligned with the economic activity that generates that income. It is, however, this notion of ‘value creation’ that has in particular been challenged in light of the digitalized economy. More in particular, the question arises whether value creation can be (partially) attributed to the market jurisdiction(s), i.e. the customer and/or user jurisdiction.

One of the key peculiarities of the digitalized economy lies in the fact that digitalized businesses can generate profit in a market jurisdiction without any physical presence or without the need to deploy significant people functions. Even if there would be a (limited) physical presence in the market jurisdiction, the current profit attribution rules depend on the functional analysis (i.e. the functions performed, assets used and risks assumed) and more specifically on the significant people functions in case of permanent establishment. The digitalized economy is, however, characterized by the decreased need for local personnel to perform certain functions. Against this background, if one takes the position that the market jurisdiction(s) should be included in the ‘value creation’-concept, the current methodology of transfer pricing analysis and profit allocation cannot be applied.

In line with the reasoning of the OECD that treating ‘digital’ as separate from more traditional businesses for tax purposes would be difficult, if not impossible, as the whole economy is digitalizing, I will explore reform options than can be applied to all types of businesses, whether or not highly digitalized.

New methods and applications for counterfactual explanations

Doctoral research by Tom Vermeire and supervised by prof. dr. David Martens. Started in January 2020 and funded by AXA Joint Research Initiative.

Advances in artificial intelligence are spurred mainly by deep learning (artificial neural networks) and the availability of massive image, textual and behavioral data. This has led to great predictive accuracies, with positive economic and societal implications, but also to very complex models of which the decision logic is difficult to understand. This project, funded by AXA Joint Research Initiative, attempts to deal with this comprehensibility issue by explaining the predictions made by black box models that are built on such massive datasets. 

The taxation of artificial intelligence

Doctoral research by Kimberly Van Sande and supervised by prof. dr. Anne Van de Vijver and prof. dr. Bruno Peeters. Started on 13 December 2019.

The world has changed considerably in the last 100 years, especially in the last decades as a result of the digital evolution. The digital evolution seems to be accelerating and has a wide range of tax implications that thwart both direct and indirect taxation. Many countries appear to be struggling with the speed of digital evolution and its tax implications. Tax systems in general are still based on the economic reality of a century ago, which means that digital revenues risk escaping an adequate qualification in civil law, intellectual property law and tax law. One of the central questions is whether international, European and national income tax rules remain 'fit for their purpose' in the age of digital evolution, where new and often immaterial profit generators are being put forward. Physical components seem to be losing their relevance and current tax law appears insufficient to capture income resulting from the digital revolution.  

In this context, legal doctrine worries about the specific challenges that artificial intelligence poses to tax law (e.g. robots in the labour market, automatic cars, chatbots). How will income resulting from artificial intelligence be qualified and allocated under current civil, intellectual property and tax law? Is there a need to modernise traditional tax concepts? Which tax concepts can be developed to contribute to fair taxation? Should artificial intelligence be considered as a separate taxpayer, and if so, what should a tax on artificial intelligence look like?

This study examines the collective term of artificial intelligence, how artificial intelligence would be qualified under current civil and intellectual property law, and how income resulting from artificial intelligence would be qualified and imputed under current tax law. Finally, this research tests the tax treatment of artificial intelligence under current law and in the future, against comparative tax law and philosophical views in law (among others, Adam Smith, Bentham Hanneke Du Preez, Nozick and Rawls).

Fraud detection with behaviour data

Doctoral research by Dieter Brughmans and supervised by prof. dr. David Martens. Started in November 2019.

The globalization, digitalization and multi-or bilateral trade agreements have given rise to an explosion in the volume, variety and complexity of data currently available to tax administrations for analysis. Can these new data sources be adopted and tailored data mining techniques be developed as to improve the accuracy and transparency of the fraud detection system?

The data sources that are available for tax fraud are dauntingly large and various. These heterogeneous data sources have rarely been used for tax fraud detection and their integration poses challenges. In this research the potential data sources and suitable data mining techniques are listed, based on the availability of data from tax administrations. Their potential will be assessed in terms of accuracy, comprehensibility and computational requirements for tax fraud detection.

These large datasets require the data mining techniques to be scalable. This is possible with techniques such as clustering of data instances, intelligent feature selection/grouping or the use of fast learners that are easily parallizable or linear in the datasize. The impact of using such speedup mechanisms on time and accuracy will be assessed.

A crucial aspect in AI is the ability to explain the predictions made by the (often black box) models that are built on such massive datasets. The suitability of existing explanation procedures will be examined. An automatically generated, data-driven and comprehensible answer can form the starting point of an investigation and is crucial regarding acceptance of the data mining model by tax experts. Techniques to be assessed include the LIME and our own EDC (patented) method in these settings (Martens and Provost, 2014).

Tax audits on big data: exploring the legitimacy and limits in light of the prohibition of fishing expeditions

Doctoral research by Liesa Keunen and supervised by prof. dr. Eva Lievens, prof. dr. Bruno Peeters and prof. dr. Sylvie De Raedt. Started in October 2019.

The research project “Tax audits on big data: exploring the legitimacy and limits in light of the prohibition of fishing expeditions” is an FWO project and runs from 2019 to 2023. The main objective is to explore the legitimacy and limits of big data audits for tax purposes, induced from the principle of the prohibition of fishing expeditions in tax matters.

In our increasingly digitised society, tax administrations have started to explore the advantages of data that is shared online or is gathered by companies, such as energy suppliers, providers of telecommunication and payment services. The availability of these so-called “big data” could make the global fight against tax fraud more efficient.

These tax audits on big data have an enormous impact on numerous people: such audits involve a lot of intrinsically private information that becomes available to tax authorities (1) and much of the information concerns individuals that are not involved in any fraudulent action (2).

Against this background, several fundamental legal questions arise with regard to:

  • the prohibition of fishing expeditions. Exploiting big data for tax purposes seems incompatible with this principle, according to which tax authorities are not allowed to search (“fish”) for information, the existence of which is uncertain. This principle seems generally accepted by tax legal scholarship, as well as by legislators and judges. Yet, the prohibition of fishing expeditions is not always interpreted in the same manner. Moreover, there is no explicit legal ground for this principle although it seems so generally accepted.
     
  • the extent to which the rights to the protection of private life and of personal data can be restricted for taxation purposes. Namely, the gathering of information by the tax authorities, especially on the basis of big data audits, can constitute an interference in the private life in the sense of article 8 ECHR. Article 8 ECHR requires that interferences comply with certain conditions, including necessity and proportionality. These conditions might raise important questions in relation to big data audits for tax purposes. Next to that, the gathering of information by the tax authorities mostly implies the processing of personal data. The requirements of legitimate processing of personal data, laid down in the GDPR, include respect for the principle of fair and lawful processing, data minimisation and purpose limitation, all of which are relevant in this context.

Therefore, the purpose of this research project is to explore the different meanings of the fishing principle to identify whether there is a(n implicit) legal basis and, ultimately, to assess if and under which conditions (keeping in mind the right to privacy and the right to data protection) tax audits on big data are legitimate.

BEPS-Related Technical Assistance in the International Political Economy of Tax Avoidance. A Process-Tracing Analysis to Agency and Discipline in Transfer-Pricing Auditing

Doctoral research by Cassandra Vet and supervised by prof. dr. Danny Cassimon and prof. dr. Anne Van de Vijver (co-supervisor).  Starts in June 2019.

How do state agents contest the structures of a world economy governed through closed communities of expertise? Whereas financial operators work with the complexity of the corporate tax regime to avoid tax obligations, a community of transnational tax experts seem inept to create the institutional change needed to halt the avoidance game (Picciotto, 2013:9). Theories of change, or rather theories about the lack of change, focus mainly on ruling elites associated with the OECD and their expert power, as well as on the distributive conflicts hindering political coordination. Consequently, the majority of agents engaging with the everyday challenges of international tax avoidance remain marginalized. Yet, the corporate tax regime mainly relies on soft power instruments and therefore the willingness of local agents to articulate these norms. Therefore, the aim of this project is to zoom in at local power struggles in developing countries between state agents and representatives of the transnational expert community to analyze how the regime is articulated and contested in the efforts to slow down corporate tax avoidance and how non-dominant players open alternative routes of change in the Global Political Economy of corporate taxation.

Currently, international organizations as well as national governments bolster their efforts to assist developing countries in minimizing the impact of international corporate tax avoidance and integrate developing countries within the international corporate tax regime. These efforts coincide with the global ambition of the BEPS-Project, a project that began when the G20 endorsed the OECD in 2013 to design the content for a multilateral project that would update the principles of corporate taxation and curb tax avoidance. Despite the fact that this project led to a breakthrough by increasing transparency provisions and provides a multilateral tool to update the bilateral framework of Double Taxation Treaties (DTT) ‘s, the project puts forward the same kind of technocratic patch-up solutions that previously failed to bring the principles of taxation in accordance the financialized world economy. Further, a transaction-based logic of auditing is kept in place, a system that is especially cost-intensive and insecure for developing countries. However, the technical assistance projects remain focused on the implementation of these transfer-pricing audits and explicitly serve the purpose of streamlining the responses of developing countries with the BEPS-framework. As a result, the comparative case-study of the impact of technical assistance on the local responses to corporate tax avoidance is an ideal case to investigate the everyday practices of anti-avoidance governance in a local-global setting.

The Use of Taxes for Regulatory Purposes and the Protection of the Cultural Heritage

Joint PhD between the University of Antwerp and University of Salerno

Phd candidate: Andrea Mucciariello and supervised by  prof. dr. B. Peeters, prof. dr. P. Pistone and prof. dr. S.J.C. Hemels (Erasmus University Rotterdam). Started on 28 March 2019.

The thesis investigates a classic theme of tax studies such as the regulatory purposes of taxation, choosing in particular to focus the investigation on a specific field of study like the protection of historical and artistic assets. The thesis work carried out in Belgium and in Italy, in particular favoring the comparative methodology, wants to offer a model for the protection of historical and artistic heritage through tax law instruments compatible with European Union law and based on the valorisation of best practices.

Tax doctrine which has questioned the purpose of taxation has traditionally held that it serves to find the resources that guarantee the functioning of the State. It has been recognized, however, that this function is not the only one possible and that taxation offers the possibility of reaching other goals as well. The theoretical reconstruction initially presented as strongly different the purpose of attracting resources from those of the achievement of other objectives, while later on it was understood that in reality the boundary between the purposes appears less clear and much more nuanced than initially envisaged. The thesis therefore moves from the succinctly described theoretical framework and addresses the issue of regulatory purposes, that is those that consist in guiding citizens to take certain behaviors judged positive by the Legislator or to avoid behaviors that the Legislator himself considers negative. A first question that the thesis deals with is the existence or otherwise of limits to the use of tax law with regulatory purposes. The answer that the thesis offers is therefore that specific limits to regulatory intervention should be identified both in national constitutions and in European law.

Linking the regulatory tax investigation to the protection of cultural heritage, the work addresses the significant problems of the definition of artistic assets in the tax field. The artistic heritage suffers from the considerable problem of finding the resources necessary to ensure its proper protection and enhancement, and the thesis deepens the possibility of offering support to solve the problem through tax law instruments. The thesis strives to jointly deal with the tax issues of cultural heritage (by way of non-exhaustive example, the taxation of historic buildings, tax credits, etc.) which present the difficulty of being very different from each other and to build a model of tax instruments to defend the artistic heritage compatible with European Union law.

Cooperative tax compliance for companies – Key Success Indicators

Doctoral research by Wouter Dister and supervised by prof. dr. Anne Van de Vijver and prof. dr. Miet Vanderhallen.  Started on 28 March 2019.

Traditionally, enforcement of tax compliance is based on deterrence. Taxpayers are compliant because of the fear of detection and punishment. However, findings in the field of behavioural sciences show that also mutual trust between the taxpayer and the tax authorities increases tax compliance. Accordingly, over the last decade the OECD and the EU have promoted a shift from the deterrence-based approach to a tax compliance model based on cooperation and trust. Cooperative tax compliance programs are designed to establish such a relationship. They are intended to deliver various benefits for tax authorities and companies: enhanced relationship, improved tax risk assessment, reduction of administrative costs, improved confidence in the tax system. Since cooperative compliance induces transparency and voluntary compliance, it is de facto also an instrument against aggressive tax planning structures. A growing number of countries have introduced cooperative compliance programs. In some countries these programs have been evaluated, mainly through surveys. Belgium has not yet introduced cooperative compliance. In 2019 a pilot drafted by the Belgian tax authorities, will be introduced.

The project will study this pilot. The research hypothesis is that a broad range of key indicators contribute to the success (or failure) of such program. The research aims to develop a framework revealing success indicators and legislative requirements to install a legal breeding ground for future initiatives. The combination of various research methods will contribute to a rich picture of key success indicators.

First, a literature review will inventory the theoretical substructure of cooperative compliance as well as insights of empirical studies in which cooperative compliance has been examined and tested. Second, a process and impact evaluation will be carried out. In the process evaluation, also important indicators for success, such as working relationship and mutual trust, will be accounted for to understand possible effects found in the impact evaluation. The impact evaluation will follow a quasi-experimental design, involving the companies that took part in the project as well as a control group consisting of similar companies that will be selected using paired matching. Third, the survey will examine the reasons to whether or not participate in the pilot on cooperative compliance. Finally, based upon the results a framework will be developed. This framework will be twofold, revealing success indicators and legislative requirements to install a legal breeding ground for future initiatives.

The research plan is innovative not only because it combines various research methods. It is also unique since the Belgian tax authorities have agreed to facilitate the research and to give access to information. Such access is seldom achieved. The research will be supervised by Prof. dr. Van de Vijver (tax aspects) and a second promotor, Prof. dr. Vanderhallen (methodological aspects). Prof. dr. Van de Vijver and prof. dr. Vanderhallen already collaborated regarding the topic of tax compliance and are both affiliated to Antwerp Tax Academy (interfaculty institute for tax science).

Critical analysis of the (distribution of the) burden of proof in tax law, taking into account the legal protection of the taxpayer

Doctoral research by Inès Rivière and supervised by prof. dr. Nicole Plets. Started on 12 March 2018.

A general legal provision with regard to the distribution of the burden of proof between the taxpayer and tax administration does not exist in the Belgian tax legislation. This is somewhat surprising, as a legal dispute can give rise to a complex distribution and interpretation of the burden of proof. The case law is therefore of great importance.

In practice, it is assumed that the tax administration bears the initial burden of proof with respect to the taxpayer's acquired income. However, this principle seems to become deteriorated, as the taxpayer increasingly takes over this initial burden of proof. In this context one can think of the heavy burden of proof resting on the taxpayer in the case of the regularization of income or a payment to a tax haven.

This research initially aims at a critical analysis of the distribution and interpretation of the burden of proof in the domain of tax law. Since the burden of proof in tax law entails a great deal of uncertainty, the question arises whether a legal provision that provides a specific allocation key regarding the distribution of the burden of proof in tax law can meet this uncertainty. The research also focuses on the sustainability and desirability of such distribution in view of the ever broader agreements and obligations regarding the data exchanges and the ongoing digitalization within tax law. Due to the international exchange of information between states and the many reporting obligations that taxpayers and third parties must meet, the tax administration can no longer be considered as “uninformed”.

Because of this evolution, the legal protection of the taxpayer will form the recurring theme throughout this research. If a tax administration collects more and more knowledge through the exchange of information and digitalization, this can be at odds with the protection of the taxpayer's private life. It is also questionable whether the taxpayer has sufficient possibilities to oppose the tax administration.

Ability to pay taxes: The future of the ability-to-pay principle in fiscalibus

Doctoral research by Karl Pauwels and supervised by prof. dr. Anne Van de Vijver. Started on 12 March 2018.

Today, a decline in the importance of the (fiscal) ability-to-pay principle can be observed. This decline has two causes. Firstly, the Belgian legislator has introduced taxes as a means to steer people’s behaviour. This excessive desire to use taxes for non-financial purposes has resulted in a deviation from the initial intention of the legislator, which was to levy taxes in accordance with a citizens’ ability to pay. Secondly, different state reforms where local regions gained more fiscal autonomy, have had a negative impact on the functioning of the ability-to-pay principle. Due to the fragmentation of tax competences, it is not easy to levy taxes in a coherent way according to the taxpayer’s ability to pay.

Furthermore, several sociological and technological evolutions have been observed that may have an impact on the ability-to-pay principle.

Therefore, the current status of the ability-to-pay principle in our tax system is rather precarious. Case law of the highest courts in Belgium is relatively scarce and sometimes contradictory, which further fuels the uncertainty with regards to the impact of the ability-to-pay principle.

To address these issues, the ability-to-pay principle will be analysed in a changed legal and sociological context. Particular consideration will be given to digital and technological progress.

The study will analyse the value, the field of application and the scope of this principle. Moreover, it will discuss how the ability-to-pay principle can be put into practice to ensure harmony with other principles on which our tax system is founded (i.a. the ‘polluter pays’-principle, the ‘user pays’-principle and the principle of neutrality), however without subverting the effectiveness of the system itself.

In the end, the aim is to enhance fairness in our tax system. The base assumption for this study is that the ability-to-pay principle contributes to more justice. By exploring and demystifying this principle, the debate on ‘fairness in taxation’ will be more substance.

The function and Criteria of Intergenerational Inheritance Law"

Phd candidate: Linde Wuyts and supervised by prof. dr. Renate Barbaix and prof. dr. Frederik Swennen. Started on 16 November 2017.

Belgian intergenerational intestate inheritance law is parentage-based. The sole criterion for intestate inheritance rights is legal parentage. However, parentage neglects different family situations. Today the societal understanding of family relations is no longer limited to legal parentage relations. There has been an evolution to families of choice (e.g. blended families, homo-sexual couple with children). Despite this evolution, children who are only biologically or socially related to the deceased, do not inherit because of the requirement of legal parentage. Furthermore, because legal parentage is sufficient to inherit -neither financial need nor any criteria other (such as a good relationship) is required-, situations of financial need of heirs and the quality of the relationship are not taken into account for the entitlement to intestate inheritance rights.

Yet, the jurisprudence of the European Court of Human Rights and the Belgian Constitutional Court do take into account a broader variation of family situations and have evolved to a less parentage-based approach. This tendency should also be noted in foreign legal systems, such as the Netherlands where it is possible for

stepchildren to inherit equally to legal children. Also in other domains of Belgian law, legal parentage is not an

exclusive criterion to give entitlement to certain rights, for example in inheritance tax law. Although stepchildren and children whom the deceased has taken care of during his lifetime do not have intestate inheritance rights, they do enjoy the same tax rate as legal children in case they would be entitled to any inheritance rights e.g. through a last will. Moreover, specific favourable measures for the acquisition of the inheritance by disabled children in Flanders indicate financial care need as an additional criterion.

These evolving societal and legal perspectives give rise to questions about the contemporary function and

criteria of intergenerational inheritance rights. Thus, this research will verify whether legal parentage is necessary to inherit intestate, and whether it is sufficient to inherit. To answer this question the function of intergenerational inheritance law in the society will be determined.

Criminal Enforcement of Tax Obligations: a Critical Analysis from a Comparative Law Perspective

Doctoral research by Jef Van Eyndhoven and supervised by prof. dr. Bruno Peeters. Started in April 2014.

Belgian tax law is fundamentally ‘criminalized’.  Most tax codes contain a general penalty provision stipulating that every intentional infringement of the tax code constitutes a criminal offense.  This thesis examines whether and to what extent a criminal enforcement of tax obligations is sensible and justified, taking into account the principles and characteristics of criminal law.  The question is being raised whether criminal law, as a ‘last resort’ (‘ultimum remedium’), should be deployed for every tax infringement.  We also try to identify different degrees of tax fraud in order to examine whether certain degrees of fraud may justify a criminal enforcement and others do not.  Where no justification for the use of criminal law can be found, we also examine the suitability of alternative enforcement methods.

Furthermore, it can be noted that the Belgian legislator has provided for a dual enforcement of tax infringements.  Infringements that constitute a criminal tax offense can also be subjected to an administrative penalty.  The result is that tax investigations and criminal investigations, as well as administrative sanctions and criminal sanctions constantly collide.  The usefulness and viability of this dual system of penalties will also be examined.

The research relies significantly on a comparative study in which penalty systems in other countries as well as penalty systems in other Belgian branches of law, especially social law, will be investigated.