Skip to main content
European Anti-Fraud Office
Report fraud
News article3 March 2017European Anti-Fraud Office

Speaking points of Giovanni Kessler, OLAF Director General at the Stakeholder Conference on the Evaluation of the OLAF Regulation



PDF version (204.67 KB)

“The OLAF model after 1999: where are we?”


We are at this Conference to gain insights on the application of the Regulation 883/2014 – and I hope that we will have fruitful discussions on our work and experience with the "OLAF model" of that Regulation.

The OLAF model of 1999

  • When I speak about the OLAF model today, it is only a variation of the one of 1999.
  • OLAF was set up back in 1999 as an ambitious project. It was created in a very short time and was granted a unique administrative investigative mandate. It was designed to be the only EU body tasked with investigating fraud affecting the financial interests of the EU, whilst also becoming the EU institutions’ watchdog that would look into any serious misconduct by Members and staff of these institutions.
  • From its creation, the legislators considered than the independence of this new body was to be paramount. This is why OLAF was granted full independence in carrying out its investigative mandate. The Office and its Director-General neither seek nor take instructions from any government, institution, body or agency, and the Director-General can bring action against the Commission before the Court of Justice if he believes a measure taken by the Commission calls his independence into question.
  • At the same time, since the beginning, OLAF had a dual mandate. It not only investigates independently, but also carries out tasks as a traditional Commission service, currently under the guidance of the Commissioner for Budget and Human Resources, particularly in preparing legislative and regulatory initiatives and designing methods to prevent and combat fraud.
  • OLAF therefore acts both as an independent investigative service and as a traditional Commission service for its non-operational responsibilities. This can lead to challenges to its independence but is also a big opportunity. It means that OLAF can profit from its investigative experience to support its policy function.
  • Beyond this dual nature, I need to say from the outset that the OLAF genesis of 1999 already contained an “original paradox” – OLAF was to be given this ambitious investigative mandate, but would only be endowed with limited investigative powers. The expectations were high and we were to act almost like a police, but with the legal framework and tools that fall short of those of many administrative bodies with comparable mandates. Furthermore, our investigative work, however good, would end in recommendations and it would be for the recipients of these recommendations to decide on enforcement measures, if any.
  • What was a first step in the establishment of a new EU Office back then has therefore been maintained to date. We are still today, some twenty years later and faced with a completely different landscape, operating on the basis of the same model of 1999.

The changes brought about in 2013 with Regulation 883

  • Regulation 883/2013 did not really shake this model to its grounds, even though it was the result of nine years of inter-institutional negotiations. This is not to say that it did not make any improvements.
  • For instance, the Regulation strengthened the procedural rights of the persons concerned by our investigations and those of witnesses. It enshrined the right to avoid self-incrimination, the right to be heard and the right to be assisted by a person of your own choice. OLAF's procedural guarantees thus became very close to those applicable in criminal proceedings, although OLAF’s investigations are only administrative ones.
  • Regulation 883 also introduced an obligation to carry out legality checks on OLAF’s investigative measures. Through these checks, a new, dedicated unit in OLAF was tasked to verify that the procedural guarantees and fundamental rights of persons concerned are respected.
  • Strengthening procedural rights undoubtedly had a positive effect. Statistics confirm that, despite the large number of investigations carried out by OLAF every year, despite the sensitive nature of OLAF’s activities and despite the fact that OLAF deals with cases involving the integrity and reputation of persons, there are very few complaints related to procedural guarantees filed against us. However, they also comprise some dangers: by giving the procedural rights already in the administrative investigation (e.g. right to be heard), you may endanger the subsequent criminal investigation.
  • Various provisions were intended to enhance OLAF's investigative activity. All Member States designated anti-fraud coordination services (AFCOS) to facilitate cooperation and exchange of information in the context of investigations and to work together with OLAF to prevent fraud. Provisions also included possibilities for a strengthened information exchange with Union institutions and bodies on possible investigations and investigations in progress, including to allow the adoption of precautionary measures.
  • Last but not least, Regulation 883 improved OLAF's governance by clarifying the mandate of the Supervisory Committee and by adding modalities for OLAF to inform and consult the Committee.
  • So overall, Regulation 883 was certainly a welcome development, providing more clarity both for us as investigators and for those whom we investigate.
  • However, it was never meant as a major reform of OLAF's investigative function at the time.
  • This means that today we truly need to reflect about the future of OLAF. And beyond that, the EU as a whole has to consider whether the current OLAF model is still good enough to cater for the overall protection of the EU's financial interests.

OLAF post 883: what we have and what we need

  • So, ladies and gentlemen, we are at a crossroads. OLAF post-Regulation 883 is still operating with the investigative tools made available to it in 1999, but is now held accountable to controls that are close to those of police-like bodies.
  • This metaphorically amounts to being a bicycle with the breaks of a sports car. At the time of 1999, the model was innovative, and we are still using it. Our good old bicycle from 1999 can still beat traffic and reach places that sports cars would not be able to. We have done a lot for the protection of the financial interests and for the accountability of the institutions. We should be proud to be the European independent investigative body that we are. The bike as worked well but it has it limits.
  • Since 1999, the world has changed and the European Union has changed. Europe has grown organically welcoming new Member States, and our internal market has liberalised in numerous ways, with the further removal of barriers to the free movement of goods and persons. Today’s economy is fuelled by products moving freely across former borders, by businesses operating across the EU and by financial flows moving instantaneously and seamlessly.
  • The EU, institutionally, has also changed after adoption of the Lisbon Treaty, which brought important changes to our framework and ability to act in the area of law enforcement, including regarding criminal law.
  • At the same time, fraudsters have changed. They have adapted to this new context. We see this not only in an increasing number of transnational fraud cases, and also in their complexity. Fraudsters today have also professionalised and are well-versed in the practices of justice or sanction forum shopping. There are websites and law firms specialised in finding all loopholes. So called "OLAF-help desks" offer 24h services in an attempt to hinder our investigations.

Clearly the world has moved on since 1999 and the challenges facing law enforcement authorities have risen. Against this background, we need to reflect carefully whether our bike from 1999 is still up to the challenge.

[1. Improved powers and tools]

  • Our main investigative tool, the on-the spot check, dates from the mid-90s. It is referred to in different ways in different pieces of legislation which, in turn, refer to conditions of national law for the conduct of on-the-spot checks. The fact that the relevant provisions are not consolidated in a single text (ideally Regulation 883) affects the legal certainty on what OLAF can and cannot do. Furthermore, OLAF's ability to investigate varies, sometimes greatly, from one Member State to another: in some Member States OLAF can carry out on-the-spot checks – in others our investigators return after having been refused to act by the national authorities.
  • We have no access to financial flows and bank accounts. This means that, while the European Court of Justice repeatedly clarified that VAT is part of the EU financial interests, and while the Court of Auditors has found that fighting VAT fraud exclusively at national level is insufficient and recommended granting OLAF clear competences and tools to investigate intra-Community VAT fraud, OLAF is currently unable to do so. Even the main investigative power of OLAF to conduct on-the-spot checks and inspections does not apply to VAT, which is not collected directly on behalf of the Union.
  • We have no clear legal basis to make full use of technological progress in the collection of evidence, nor do we even have easy and direct access to all databases managed at EU institutions' level. Moreover, we have no access to records of telephones and to data traffic, nor the ability to seal premises of economic operators, although these would help our investigators work more efficiently.

[2. Enforceability of powers]

  • The issue is not only about clearer and better suited powers for anti-fraud investigations. Sometimes, even if our tools are clear, we cannot enforce our powers. We cannot compel persons to testify and we sometimes cannot even enter premises of economic operators without prior authorisation, or without the consent of the owner because. So even if legally have the right to do so, we have no means to enforce that right.
  • OLAF inspections of European Institutions in the case of internal investigations are also an issue. OLAF's powers for internal investigations depend on the internal decision issued by the institution concerned. So OLAF's effective means to investigate are not uniform.

[3. Controls tailored to the nature of OLAF's activity]

  • The same OLAF acts can today be supervised by a plethora of oversight bodies, such as the European Court of Auditors, the European Ombudsman, the EDPS, national courts, the European Court of Justice, the Internal Audit Service and our own Supervisory Committee. OLAF, of course, also has to report on its activities to the Parliament and occasionally in Council. Some of these controls are contained in Regulation 883 and specific to OLAF; others apply to OLAF as to any other Commission service or, more generally, to any Union institution or body.
  • This in itself is not a problem as the oversight bodies in principle have different mandates, but in practice this is not always the case. It sometimes leads to the unintended consequence that the same issue is brought to the attention of several of these controllers, simultaneously or sequentially. Or that controls not designed to oversee an investigative function are applied to OLAF's investigations in addition to its specific controls.
  • By the very nature of its activity, controls on OLAF must be in place, and they should be proportionate to the powers and tools that the Office is given to investigate and prosecute. However, I see a need to ensure that control functions are complementary and do not overlap. One oversight body should not become a sort of appellate jurisdiction for another with a different mandate. The controls should be tailored to OLAF, which has a very unique nature among Commission services and EU offices.

[4. Follow-up to recommendations]

  • OLAF issues recommendations and does not itself recover or prosecute. Without questioning this model, some improvements could increase the follow-up rate and the ability of Member States and EU institutions to act on OLAF findings.
  • The evidentiary value of our final case reports is not uniform across Member States and this is a crucial problem. Our reports have the same evidentiary value of the reports of national administrative authorities; as a result, OLAF reports can be used only in those Member States where such reports are admissible in criminal proceedings. For some Member States, art. 11(2) of Regulation 883 is not a sufficient legal basis to allow judicial authorities to use OLAF reports as evidence in trial. Therefore, after receiving the OLAF final report, some Member State prosecutors carry out all the investigative activities once again in order to acquire admissible evidence.
  • Regulation 883 does also not provide for a specific legal basis regulating the transmission of information on the results of OLAF investigations to third countries or to International Organisations. At the end of an investigation, OLAF can only send a letter to third country authorities, asking them to consider possible prosecution against exporters in their national territories whose actions are leading to fraud in Europe.

Way forward

  • With these limitations in mind, it is almost incredible what OLAF has managed to do. Hundreds of investigations and recommendations and billions recovered to the EU budget.
  • The skills of our investigators and our staff as a whole have helped OLAF develop into an effective, professional and respected service and I believe the European Institutions should be as proud of OLAF's achievements as I am. However, we've done the most with the powers we currently have.
  • So what I am calling for today ultimately is reform. Let us have effective investigations and adequate safeguards, and let us have the tools allowing us to perform the investigative acts necessary in the area of fraud, to generate evidence that can be relied on in the courts of the Member States.
  • Let us have access to bank accounts, or even the ability to ask national authorities to retrieve bank account information. Simply being able to link a particular person with one or multiple bank account numbers, without even going into the content of the transactions, would do wonders for the speed with which we would be able to examine cases.
  • Let us have clear authority to interview witnesses and to enter premises of economic operators, so we can be most effective in our inspections and searches.
  • This would provide more clarity for us, as investigators, but also for the persons concerned.
  • Given the challenges we're faced with, we need a strong, efficient and modern OLAF to protect the interests of the European Union.
  • The evaluation criteria also require us to look at issues of coherence and relevance of our legal framework: whether and how Regulation 883/2013 fits into the wider EU policies and current policy developments in the area of fraud, and how OLAF can contribute. Compared to other Commission services and EU agencies, OLAF's added value lies in both its investigative capacities and the expertise derived from them, and in its ability to cooperate and exchange information directly with third countries.
  • That makes OLAF uniquely qualified to tackle areas related not only to the European Union's financial interests, but to other concrete, direct interests of the European citizens, such as health or intellectual property rights. As Commissioner Oettinger pointed out, a vision for OLAF's future could include, for example, an increased involvement in tackling food fraud, which can endanger the health and safety of EU citizens and damage consumer confidence. Under our current legal framework, there are no investigative powers at EU level to deal with instances of fraud that affect the health and safety of consumers. Therefore, in the future, we could envision OLAF taking on such challenges, in view of its investigative and fraud expertise.
  • There is also scope for strengthening administrative cooperation for the protection of the Union's financial interests, in areas such as structural funds, where no such mechanisms exist. We are detecting an increasing trend in the number of transnational fraud cases and also in the financial damage associated with transnational fraud. This is an area of high visibility for citizens, where fraud can lead to significant reputational damage for the EU.
  • As you can see, there is so much we are doing, but so much more we could do in the future. Using the existing expertise and structure of OLAF, we could make best use of our existing resources to the benefit of European citizens.
  • We need a new step change beyond the modifications brought by Regulation 883. A way for OLAF to stay ahead of the game and, ideally, to even grow in a way that allows it to tackle broader fraud issues more effectively than ever before.
© EU

Stakeholder conference on the Evaluation of the OLAF Regulation, ©EU



Publication date
3 March 2017
European Anti-Fraud Office
News type
OLAF news article