1.

Opinion No 1/2013: OLAF's Preliminary draft budget for 2014

2.

Opinion No 2/2013: Establishing an internal OLAF procedure for complaints

3.

Opinion No 1/2014: OLAF Investigation Policy Priorities

4.

Opinion No 2/2014: Case selection in OLAF

5.

SC's observations on investigation procedures in OLAF

6.

Paper on Reinforcing procedural safeguards in OLAF

7.

SC's 2012 Recommendations

8.

Mission, competences and objectives of the SC: Mid-term strategy

9.

Code of Conduct of the Members of the SC and Explanatory memorandum on safeguards of impartiality and risks of conflict of interest in the exercise of the monitoring functions

10.

Working arrangements with OLAF

1.

OLAF has been given far-reaching powers of investigation in order to efficiently carry out its mission to protect the financial interests of the EU. The exercise of these powers is very likely to touch upon the fundamental rights of persons concerned by investigations (1), and therefore their respect by OLAF is essential. It contributes — as the SC has already emphasised (2) — not only to ensuring the effectiveness and efficiency of OLAF's investigative activities, but also to consolidating its reputation, credibility and ultimately its independence.

2.

OLAF is obliged to ensure the protection of fundamental rights by safeguarding the procedural guarantees as listed by the EU Charter of Fundamental Rights (3). In addition, the EU legislator decided to enhance OLAF's accountability for its investigative activities by reforming its legal framework (4) which resulted in the adoption of Regulation (EU, Euratom) No 883/2013 (5) (hereinafter, the Regulation).

3.

The SC welcomes the changes brought by the Regulation but would like, nevertheless, to point out that the mechanisms put in place to enforce the procedural guarantees need further improvement. The Regulation provides for a legality check (6) and a monitoring mechanism (7), but does not establish a mechanism for dealing with individual complaints which would be, in the SC's view, indispensable for ensuring effective protection of fundamental rights.

4.

In the SC's opinion, persons involved in OLAF investigations do not have sufficient and immediate remedies to redress potential violations of their rights and the recent reform of the legal framework has not solved that problem. Therefore, the SC takes the view that the Director-General of OLAF ((hereinafter, the DG) should swiftly establish a transparent and stable internal procedure for dealing with individual complaints.

5.

Persons considering that, in the course of an investigation, OLAF breached their fundamental rights and wishing to complain, have at their disposal various means of judicial and non-judicial review which, however, may be insufficient, for the reasons stated below.

6.

Firstly, they may seek the direct judicial review by an EU Court (action for annulment, action for damages and request for interim relief) or indirectjudicial review by EU and national courts (preliminary ruling procedure). (8) However, it should be noted that these actions are governed by rather strict rules on admissibility (9) and must fulfil a number of conditions to trigger a right to reparation (10). Even when an action is successful, ‘Judicial protection is obtained long after the contested investigative act and the act as such is not affected by the EU court's ruling; only compensation for the prejudice created by that act is obtained’ (11).

7.

Secondly, persons wishing to complain may address themselves to the European Data Protection Supervisor (EDPS), who is specifically entrusted (12) with the competence of hearing and investigating complaints lodged by persons whose personal data are processed by OLAF. However, the material scope of his review is limited to the protection of personal data and privacy.

8.

Thirdly, the European Ombudsman (EO) is competent to investigate complaints regarding maladministration (13) against OLAF, namely failure to act in accordance with the law, to respect the principles of good administration, or breaches of fundamental rights. Nevertheless, it is the EO's view — which the SC fully shares — that an institution ‘in frequent contact with people who may have reason to complain should provide for a first line complaints mechanism allowing for their problems to be addressed and solved rapidly, before, in the event that resolution is not successful, having to turn to other redress mechanisms, such as Ombudsmen and courts’ (14). The SC considers that this statement also applies to OLAF.

9.

Fourthly, the SC itself — as the OLAF's supervisory body — is quite often the addressee of individual complaints which it takes into consideration as a valuable source of information concerning OLAF's investigations and thus triggering the SC's monitoring activity. The SC's role, as emphasised by a ruling of the EU General Court (15) and confirmed by Regulation (EU, Euratom) No 883/2013, is to monitor developments concerning the application of procedural guarantees and the duration of investigations (16) which cannot be carried out without examination of individual cases. However, the SC was not established as a complaints body and its supervision takes place only ex post, since the SC cannot intervene in ongoing investigations. Furthermore, the SC regrets the lack of appropriate monitoring tools at its disposal and the absence of a specific reporting obligation for the DG to inform the SC of complaints addressed to him and, in particular, on the way they were dealt with.

10.

In addition to the external mechanisms, the SC notes that an internal mechanism for dealing with individual complaints is not completely absent from the legal framework governing OLAF's investigations. Any official or other servant of the EU who is the subject of an internal investigation has the right to submit to the DG a complaint against an act adversely affecting him in connection with investigations by OLAF (17). However, according to the existing EU case-law, OLAF's investigative acts do not constitute acts adversely affecting the persons concerned within the meaning of Article 90a of the Staff Regulations and therefore are not challengeable acts (18).

11.

Moreover, the SC would point out that OLAF staff who are aware of possible illegal activities or serious violations of professional duties that might trigger potential breaches of fundamental rights and procedural guarantees within an investigation have the additional possibility to report wrongdoing within OLAF to the President of the SC, on the basis of Article 22a of the Staff Regulations and of the agreement between the DG and the SC (19).

12.

The SC notes that the intention of the EU legislator was to strengthen the protection of procedural guarantees applicable in OLAF's investigations (a) by inserting in the Regulation detailed provisions concerning the procedural guarantees and (b) by enforcing them by means of an enhanced review. As a result, Article 9 of the Regulation contains now explicit procedural guarantees and thus defines and clarifies the contents of those foreseen in the Charter. On the other hand, although the EU institutions involved in the legislative process all agreed on the need for a review of potential violation of fundamental rights and put forward various proposals aimed at establishing both an ex ante legality check (relating, inter alia, to the respect of procedural guarantees of persons under investigation, during all stages of the investigation) and an ex post review procedure for handling individual complaints (20), it seems that no consensus was finally reached, since the Regulation does not explicitly put in place any such procedure.

13.

The SC would point out that in the past, via its internal rules, OLAF put in place a review procedure open to all persons concerned in internal and external investigations in which procedural guarantees were allegedly not respected in a manner possibly having an impact on the conclusions of the investigation (21). This procedure evolved under the influence of interinstitutional discussions on reforming the OLAF legal framework. Since the Commission's proposal of 17 March 2011 contained provisions on a review procedure (22), the Instructions to Staff on Investigation Procedures (ISIP) (23) adopted by the OLAF DG as of 1 February 2012 anticipated the forthcoming legislative changes with the setting up of a procedure for dealing with complaints concerning an alleged failure to respect procedural guarantees (24). However, that Commission's proposal, which was subject to negotiations during the legislative process, was ultimately not maintained. Therefore, when the ISIP were replaced by the new Guidelines on Investigation Procedures (GIP), which entered into force on the same day as the new OLAF Regulation (1 October 2013), this internal procedure disappeared.

14.

As a result, neither the Regulation nor OLAF's internal rules foresee any formal procedure for dealing with individual complaints. Therefore, the objective of improving the mechanisms for redressing potential breaches by OLAF of procedural guarantees of persons under investigation has not been achieved. The SC would point out that the recent Communication from the Commission on Improving OLAF's governance and reinforcing procedural safeguards in investigations  (25) supports the SC's view and its repeated recommendations (26) concerning the introduction of transparent and stable procedures for an internal legality check and for an independent review of complaints. This Communication also confirms the need for further improvements to the Regulation, by establishing of the office of a ‘Controller of procedural safeguards’.

15.

With its longstanding experience in monitoring OLAF's investigative function and thereby compliance by OLAF with fundamental rights and procedural guarantees, the SC is therefore concerned by the lack of a transparent procedure for dealing with individual complaints and takes the view that, at least during the transitional period, until further amendments to the Regulation are adopted, the current gap should be closed by the DG re-establishing an internal procedure for complaint in OLAF.

16.

In OLAF's view, the notion of ‘complaints’ covers (i) complaints concerning procedural guarantees in the context of investigative actions, (ii) complaints against a reply by OLAF to a request for access to personal data and (iii) complaints from whistle-blowers (concerning either OLAF's obligation to inform them on the action taken following their fraud reporting or the protection of their identity) (27). According to information provided by OLAF, in the last three years there were 25 complaints concerning OLAF's investigation activity, of which 13 were addressed to the EO (28), 8 were addressed to the EDPS (29) and 4 were requests for review addressed to OLAF (30). OLAF is also aware of other complaints, addressed to other bodies (e.g. to the SC), but it does not have any systemic overview, register or statistics related to them.

17.

Taking into account the number of cases opened by OLAF (718 in 2012 (31)) and the fact that complaints against OLAF are addressed, in particular, to the EDPS, the EO and the SC (32), the small number of complaints addressed to OLAF itself may lead to the conclusion that the absence of a transparent and independent procedure for handling complaints in OLAF effectively deters potential complainants from exercising their rights.

18.

The SC notes, from its monitoring experience, that in the previous years the handling of complaints appeared to be carried out on a case-by-case basis rather than through a stable and consistent procedure. The SC shares the EO's view that a case-by-case approach is not ‘the best way to ensure the efficient and transparent handling of complaints’ (33) and that without a proper complaint procedure ‘compliance [with fundamental rights] cannot ultimately be effective’ (34). Such a procedure is a key element of OLAF's accountability and legitimacy, because protecting fundamental rights ‘is not only to the benefit of persons under investigation but also of OLAF, whose legitimacy can only increase when fundamental rights are (seen to be) effectively protected’ (35).

19.

The SC considers that, from the OLAF perspective, a complaint procedure would help (i) the detection by OLAF of any breach of its own legal obligations, thus allowing for a prompt reaction, (ii) monitoring by OLAF of compliance with fundamental rights and procedural guarantees by its own staff, (iii) avoiding the risk of discriminatory treatment of persons concerned, (iv) settling possible disputes before they aggravate (36).

20.

In the SC's opinion, Article 41 of the Charter of Fundamental Rights (37) (right to good administration) would offer a sufficient legal basis for establishing an internal procedure for dealing with individual complaints. Complainants are entitled to expect from OLAF that their ‘affairs’ are ‘handled’ — according to the principles defined by the Charter.

21.

The SC therefore believes that the DG should put in place a transparent and efficient internal procedure for dealing with all the types of complaints related to OLAF investigative activities, including not only complaints alleging potential breaches of fundamental rights in the course of an OLAF investigation, but also those concerning duration of investigations, legality of OLAF's acts, breaches of confidentiality, refusals to provide information. The following practical recommendations are without prejudice to the forthcoming legislative proposals (38) or to the SC's position as expressed in its paper on Reinforcing procedural safeguards in OLAF — in view of the monitoring experience of the Supervisory Committee  (39) — they should be regarded as transitional measures until the current legal framework is improved.

22.

The SC would firstly point out that it is for the DG to establish and define the exact scope and content of rules on handling of complaints in OLAF. In so doing, the DG should take into consideration a number of requirements triggered by the right to good administration. Recommendations by the EO and the Code of Good Administrative Behaviour should be also taken into account. Without establishing an exhaustive list, the SC would point to the following requirements:

23.

Secondly, with regard to the person entrusted with the handling of complaints, he should have the appropriate competencies, qualifications and experience. Such person should have well defined tasks and benefit from access to appropriate resources. He should have access to all relevant case-related information and OLAF staff should fully cooperate with him.

24.

Regarding particularly the requirement of independence, the SC would stress that any effective internal complaint procedure, which would necessarily imply the handling of a complaint by an OLAF (senior) official, requires that this person act, when carrying out his review duties, in full independence of the DG. This is a key element in ensuring external credibility and public trust. The SC acknowledges the difficulty of implementing such a requirement, since this person should report to the DG on his findings while the latter remains the appointing authority deciding on the career progress of all OLAF officials (46).

25.

The SC would indicate three possible solutions to that problem.

26.

Finally, the SC considers that information on how to address complaints to OLAF, together with a description of the procedure for handling them, should be published on OLAF's website, for the sake of transparency and legal certainty. Moreover, the SC should be regularly informed of the complaints received by OLAF and the way they have been handled, this allowing it to properly perform its role of monitoring developments concerning the application of procedural guarantees and the duration of investigations.

(1)

Recommends that the Director-General set up an internal procedure for dealing with individual complaints concerning OLAF investigations;

(2)

Invites the Director-General to consult the details of the procedure with the SC, before its adoption;

(3)

Recommends that the Director-General publish the procedure on OLAF's website after its adoption;

(4)

Expects the Director-General to report regularly to the SC on complaints received by OLAF and the way they have been handled.

I.

I.1.

I.2.

I.3.

I.4.

I.5.

II.

II.1.

II.2.

II.3.

II.4.

III.

1.

The Investigation Policy Priorities (IPPs) were established by the European AntiFraud Office (OLAF) for the first time within the framework of the Management Plan 2012 as an element of the reorganisation of the Office, aiming at improving the selection procedure. ‘The IPPs [were] used by OLAF to decide on the opening of investigations. These priorities [came] into play after the competence of OLAF and the necessary level of suspicions [had] been established’  (1).

2.

On 1 October 2013, Regulation (EU, Euratom) No 883/2013 (2) entered into force. It obliges the Director-General of OLAF (OLAF DG) to determine the IPPs each year and also to forward them, prior to their publication, to the Supervisory Committee (SC) of OLAF (3). In the light of its role of reinforcing OLAF's independence and of its duties to assist the OLAF DG in discharging his responsibilities (4), the SC decided to thoroughly examine the IPPs for 2012, 2013 and 2014.

3.

There were five IPPs established in 2012:

4.

The special policy objectives were related to sectors and geographical areas decided by the OLAF DG on the basis of concerns and priorities expressed by the European Institutions and of OLAF's own risk analyses. For 2012 the special policy objectives included:

5.

The priority related to the financial impact included financial indicators, e.g. EUR 500 000 in the European Social Fund and Cohesion Fund and € 1 million in the European Regional Development Fund (estimated misappropriation or wrongful retention of funds), over € 1 million (estimated illegal diminution of resources) in the customs sector (6).

6.

The IPPs have been updated for 2013 and they include four criteria:

7.

The special criteria for 2013 are related to sectors only and no longer to geographical areas. The adoption of those criteria is linked to documents issued by stakeholders (7). There are three prioritised sectors:

Finally, the special criteria include also financial indicators which are equivalent, in content, to the financial impact priority of 2012 (9).

8.

Regulation (EU, Euratom) No 883/2013 establishes certain principles to be taken into account by the OLAF DG, alongside the IPPs, when deciding whether or not to open an investigation (10).

9.

Those general principles include:

10.

On 4 December 2013 the OLAF DG transmitted to the SC the draft IPPs for 2014 which are to be adopted by the end of January 2014. The draft priorities for investigation in 2014 include:

11.

The draft priorities are related to the areas of activities of OLAF Directorate B only (not Directorate A). They are based on 9 key documents issued by stakeholders and they do not contain any reference to financial indicators.

12.

On 11 February 2013 the responsible unit in the Directorate-General for Agriculture and Rural Development (DG AGRI) formally alerted OLAF via a registered note (11) that the financial indicator for one of the priorities — rural development — adopted by the OLAF DG, apparently without having consulted DG AGRI and defined as over EUR 250 000, would exclude 99,83 % of cases, since only 1 036 beneficiaries, out of 563 196, received subsidies above that threshold. DG AGRI expressed the view that it might be worthwhile considering a revision of the threshold of EUR 250 000, in order to increase the impact of OLAF's activities in the rural development sector. In May 2013 the Director-General of DG AGRI complained (12) to the OLAF DG that the above mentioned note of 11 February 2013 had never been answered. OLAF DG replied in July 2013, ‘taking note’ of the proposed ‘adaptation of the financial thresholds’, stressing that they are just one of the indicators and that when cases are dismissed on the basis of the IPPs, the relevant information is ‘systematically’ transferred to the competent authorities (13).

13.

With regard to a number of cases dismissed by OLAF and transferred to DG AGRI for appropriate action, DG AGRI continued to express its disagreement. In particular, DG AGRI pointed out that its competence lies in carrying out system audits; it therefore cannot undertake verifications regarding individual cases (e.g. possibly irregular payments to individual beneficiaries) that would amount to investigative activities for which it is not competent. As a consequence, DG AGRI decided not to follow-up such cases forwarded by OLAF. (14) In reply, DG AGRI was referred to OLAF's ‘Guidance note for treatment of dismissed cases’ (15) listing actions that can be taken in such circumstances.

14.

The SC's inquiries have shown that the problem of insufficient consultation with the stakeholders concerns also other ‘spending’ DGs. In particular the issue of criteria for establishing financial indicators and the issue of the follow-up of cases of fraud or irregularity which are dismissed by OLAF seem to be addressed unilaterally by OLAF, without taking effectively into account opinions of the stakeholders.

15.

The IPPs for 2012 and 2013 contained, as priority criteria to be taken into consideration by OLAF to decide on the opening of investigations, three general principles, namely proportionality, need for efficient use of investigative resources and subsidiarity. The SC has pointed out in technical meetings with OLAF that proportionality and subsidiarity belong to the general principles of the EU law and as such must be always applied by all the EU bodies in accordance with the established EU case-law. As a result, those principles cannot constitute discretionary priorities of the Director-General of OLAF.

16.

The SC notes that Regulation (EU, Euratom) No 883/2013 has established a clear distinction between the three general principles which always need to be taken into account by the Director-General and the priorities which he himself is to determine each year within the context of the annual management plan (16). The principles established in the Regulation correspond largely to the first three IPPs for 2012 and 2013. Therefore, since 1 October 2013, only the special annual criteria and the financial indicators determined by the Director-General can be considered as the IPPs for OLAF in the sense of the Regulation.

17.

18.

Moreover, the SC would draw attention to the fact that the general principles established in the Regulation are binding on the Director-General for all his decisions on opening or not an investigation. Since these decisions are based on the opinions of the Investigation Selection and Review Unit, the SC believes that it is important to provide the responsible staff in that unit with guidelines on the application of those principles.

19.

The IPPs for 2012 and 2013 listed a certain number of financial indicators to be taken into account in the selection process, together with other priority criteria. Already in its 2012 Annual Activity Report, the SC indicated that some of the thresholds were excessively high. At that time, the SC expressed its concern that potential fraud or irregularities likely to have a lower, but still significant financial impact, may not be taken into consideration, despite clear evidence (17). In particular, the SC highlighted that ‘The setting by the Director-General of high financial thresholds, to be taken into consideration by OLAF prior to open an investigation in various sectors and the possibility for the Director-General not to open an investigation even if there is ‘sufficient suspicion’, without a parallel duty for the Director-General to inform the competent authorities of institutions when he decides not to open an investigation, risks resulting in areas of impunity’ (18).

20.

Another concern that the SC wishes to express relates to the fact that those financial indicators were established apparently without any consultation with the stakeholders. It would not be consistent with the European Commission's anti-fraud strategy, which promotes OLAF's proactive and reinforced role in assisting the Commission Services, by providing, inter alia, a methodology and guidance on the development and implementation of the sectoral anti-fraud strategies (19). The SC would point out that the departments responsible for managing and/or supervising EU funds should receive proper information concerning suspicions of fraud or irregularities detrimental to the EU financial interests. In the absence of such information, they cannot ensure adequate protection of the EU's financial interests. In addition, the SC considers that OLAF would benefit from a constructive dialogue with its stakeholders with regard to financial indicators, since they are primarily responsible for managing EU funds and, as such, hold the most reliable information on elements of expenditure potentially affected by fraud, irregularities or other illegal activities detrimental to the EU financial interests.

21.

22.

The SC would point out that, despite the financial indicators for 2012 and 2013 being arbitrary and dangerously overstated, it considers that some financial impact criteria — discussed with the stakeholders — are, however, relevant for the assessment of the seriousness of the risk involved and could be useful as an element of reference and as internal guidelines on the application of the proportionality principle. The financial criteria should take into account the input from the European Commission's Directorates-General and from the EU institutions or bodies involved in the financial lifecycle and which are the best placed for holding timely information on the management of EU funds and on potential risks of fraud or irregularities detrimental to the EU financial interests.

23.

The SC notes that the subsidiarity principle, combined with the assessment of the added value of OLAF's action, was one of the IPPs for 2012 and 2013. In Regulation (EU, Euratom) No 883/2013 the added value appears as a stand-alone principle concerning internal investigations (20), while the principle of subsidiarity is referred to (in the meaning of Article 5 of the Treaty on European Union) in the context of external investigations (21). However, they continue to be applied by the Investigation Selection and Review Unit as one selection criterion, regardless of the type of case (22). Since the two principles have been reinforced by the Regulation and defined in two different contexts, the SC considers that OLAF should clarify their application in the selection process.

24.

The SC also notes, from the feedback received from other Directorates-General of the Commission, as well as from its own analysis of opinions of the Investigation Selection and Review Unit, that the application of this principle by OLAF is a matter of concern and that the likelihood of a follow-up by another authority seems to be insufficiently taken into consideration. While reiterating its views with regard to OLAF's ‘de minimis’ policy, in particular the opinion that, in cases where minor wrongdoings/low impact cases can be dealt with satisfactorily by other services, OLAF may forward incoming information to other Commission services, rather than decide to open an OLAF investigation (23), the SC would point out that OLAF should be careful to verify whether or not the recipient authority has the necessary competence and powers to deal with the case, in order to be sure that appropriate follow-up is given to cases dismissed by another authority.

25.

The SC notes that the IPPs, which are determined within the context of OLAF's Annual Management Plan, change every year and there are significant differences between the IPPs for 2012, 2013 and 2014. In the SC's opinion, such changes should be justified by concrete needs and measurable indicators. However, the SC is not aware of any internal or external evaluation, feedback or follow-up of the IPPs for 2012 and 2013, and equally notes the absence of any action aimed at assessing the impact of the priorities prior to their adoption.

26.

The SC believes that the IPPs should have as a purpose accomplishing certain objectives and therefore their implementation should be regularly assessed by OLAF, in order to establish whether or to what extent those objectives have been achieved. Moreover, the adoption of any new IPPs should be based on an impact assessment, which should include concrete and measurable indicators, lessons learnt from the implementation of previous IPPs and results of internal audits. The draft IPPs which shall be forwarded to the SC every year should be accompanied by the assessment of the implementation of former IPPs and the impact assessment of new IPPs, together with background documents.

27.

In light of the above considerations, the Supervisory Committee:

PART I

PART II

(1)

(1.1)

(1.2)

(1.3)

(1.4)

(2)

(2.1)

(2.2)

(2.3)

(3)

(3.1)

(3.2)

(3.3)

1.

On 1 February 2012, OLAF's investigative procedures and internal organisation were significantly changed. New Instructions to Staff on Investigative Procedures (ISIP) were issued and a new Investigation Selection and Review Unit (ISRU) was created. This unit deals with both the selection and review of cases (1) and provides opinions on the basis of which the Director-General (DG) of OLAF takes decisions on opening or dismissing cases, on the main investigative activities, on the final report and on the recommendations. The ISRU is thus involved in the whole lifecycle of an investigation or coordination case and most of the decisions taken by the DG in the investigation area are based on its opinions. Therefore, the internal rules, the organisation, the competences, the efficiency and the quality of the work of the ISRU have a major and direct influence on the performance of the whole OLAF investigative function. The Supervisory Committee (SC) has therefore decided to pay particular attention to the execution by this unit of its role.

2.

The SC's review has focused on the selection function of the ISRU — and, in particular, on the selection process (2). The SC's analysis takes into account the legal provisions and instructions to staff in force at the time when the analysed opinions were delivered, namely Regulation (EC) No 1073/1999 (3) and the ISIP. However, the SC's recommendations are made in the light of the current Regulation (EU, Euratom) No 883/2013 (4) and the new Guidelines on Investigative Procedures (GIP), both of which entered into force on 1 October 2013.

3.

On 1 February 2012, the assessment of the incoming information of possible investigative interest to OLAF was transferred from the investigation units to the ISRU. A centralised decision-making system was created, consisting of two levels: the decisional level, represented by the DG, and the advisory level, represented by the ISRU. Such centralisation may help to improve the efficiency and consistency of the selection of cases, provided that the ISRU carries out its function in an effective, competent and transparent manner, according to clear principles and drawing on the expertise of investigative, analytical and legal units. The SC's review has as its objective to verify whether those requirements have been satisfied.

4.

The SC selected a sample of opinions on the opening or dismissal of cases: it consisted of a statistical sample of opinions from all the cases created between 1 September 2012 and 31 March 2013 plus all opinions of May 2013 and plus further 20 opinions selected by the DG. According to the criteria submitted by the SC, OLAF identified 314 opinions (5) of which 1 contained erroneous numbering and 20 were submitted in duplicate (appearing twice on OLAF's lists). Therefore the total number of opinions eligible to be examined by the SC came to 293, of which 218 opinions (74,4 %) recommended the dismissal of cases and the remaining 75 opinions (25,6 %) recommended the opening of investigation or coordination cases (figure 1). All the recommendations have been followed by the DG.

5.

The SC's review is based on:

6.

The details of the analysis and of the findings are presented in Part II of this opinion. In Part I the SC presents its conclusions and recommendations made to the DG.

7.

The SC's examination of allocation of resources to and within the ISRU revealed that, generally, there are experienced selectors in the unit, covering a wide range of specialisations. However, each selector deals with different sectors, including those where they have less expertise, which, in particular with regard to selectors who have no investigative experience, may have affected the quality of the assessment carried out. The limited number of the training courses completed by the selectors does not appear to compensate for the lack of previous experience. The SC found also that legal knowledge was not sufficiently demonstrated in many opinions and that language expertise was sometimes missing (see paragraphs 31-35 and 38).

8.

The SC is of the opinion that the fact that selectors are required to assess information in domains with which they are not always familiar may increase the risk of losing time when having to switch between different sectors and of performing incomplete assessments of incoming information. Selectors need to be experienced and properly trained to perform selection tasks. Selectors should also concentrate their efforts on tasks and sectors in which they are most qualified.

9.

The SC equally noted that it was neither possible nor productive to strictly adhere to the mandatory 2-month period foreseen in the ISIP for completing the selection of cases. Therefore, the SC welcomes the fact that the new Guidelines on Investigation Procedures eliminated this overly rigid time limit, which allows for a better balance to be struck between OLAF's objective to improve the efficiency and effectiveness of its investigation and coordination cases (by speeding up the period dedicated to the assessment of incoming information) and the need to avoid excessive time pressure (which might be damaging to the quality of the assessment) (see paragraphs 39-40).

10.

Lastly, the SC noted that the adoption of clear instructions for dealing with whistle-blowers, as well as technical improvement of the Fraud Notification System (FNS) would also be desirable (see paragraphs 41-44).

11.

The SC also noted that the reviewers have been entrusted not only with review tasks, but also with the selection of cases. The SC would point out that, the tasks of the selectors being fundamentally different from those of the reviewers, combining their responsibilities is questionable. The SC is therefore of the opinion that the attribution of cases for selection to the reviewers, in order to compensate for the insufficiency of the language expertise within the unit and resulting in an increase of the reviewers' workload, should remain a temporary and exceptional solution (see paragraphs 36-37).

12.

The SC found that the assessment of OLAF's competence to act was insufficiently substantiated and in general little consideration was given to the relevant legal instruments (see paragraphs 47-49).

13.

The SC notes that there are no clear instructions with regard to the indicators to be used by the selectors to evaluate whether or not the information is sufficient to justify the opening of an investigation or coordination case. As a result, evaluation of this selection criterion was not always properly carried out, while at the same time a variety of practices and a certain degree of inconsistency was noted in the ISRU's opinions. The SC concludes that OLAF's approach to assessing this selection criterion needs to be further developed and clarified. To that nd, OLAF could take note of concrete indicators used by selectors in some of the opinions (see paragraphs 50-54).

14.

The SC notes that the Investigation Policy Priorities (IPPs) established by the DG indicate very clearly a number of concrete and measurable indicators that shall be used in the selection process. However, the SC's review of opinions revealed that these indicators are not systematically and rigorously applied, in particular with regard to ‘proportionality’, ‘efficient use of resources’ and ‘special policy objectives/criteria’.

15.

With regard to assessment of proportionality of OLAF's action, the SC found that, despite instructions in the IPPs, the opinions frequently contain unsubstantiated statements and, in many of them, the proportionality test is either absent or incomplete. Moreover, the special policy objectives/criteria are not systematically used (see paragraphs 56-57 and 64-65).

16.

The SC also found that the evaluation of the criterion ‘efficient use of resources’ was not properly carried out. The SC believes that several factors may explain that. On the one hand, the evaluation of the four indicators mentioned in the IPPs is hardly achievable without close contact with the investigation units. While the selectors may indeed check in the CMS on the workload of the investigation units (in terms of number of investigators and investigations per unit), the management of these units is better placed to appreciate whether this workload permits swift and continuous investigative activities or could slow down priority investigations. On the other hand, neither the IPPs nor other guidelines on selection (8) establish a threshold above which the ratio of cases per investigator would be considered to be excessive workload. As a consequence, in the absence of clear instructions as to the way in which the size of the workload should be evaluated and of a systemic dialogue between the ISRU and the investigation units, the evaluation of the efficient use of investigative resources would appear to depend to a great extent on a personal approach of each selector. The variety of approaches and sometimes inconsistencies noted by the SC in the opinions confirm this conclusion (see paragraphs 58-60).

17.

The SC's review of opinions revealed that, in the few cases where an estimate of the size of the workload of investigators was made, it was not a determining factor for the dismissal of cases. The SC would support the approach that the workload of investigation units should not in itself be sufficient justification for the dismissal of a case. The lack of sufficient resources in investigation units should be tackled by other means, especially via management tools (e.g. temporary reallocation of staff).

18.

The SC found that the reasons outlined as to why an OLAF action would not bring any added value to the control activities carried out by other EU or national bodies, which are better placed to act, are generally well explained. Some reasons are recurring, so it would be more efficient for OLAF to identify the main reasons and establish for the selectors a pre-determined list of possible situations where another authority is considered better placed to act (see paragraph 61).

19.

The SC noted with concern that the likelihood of a follow-up by another EU or national authority seems to be insufficiently taken into consideration by OLAF. When information on dismissed cases is forwarded to national or EU authorities, it is of the utmost importance that OLAF follows up the action taken by them, in order to be able to react in an appropriate manner when a case is not effectively dealt with by these authorities (see paragraphs 62-63).

20.

The SC would point out that, while the national authorities may indeed often be better placed to act, the situation is, however, different with regard to the EU institutions. When OLAF forwards information on serious suspicions of fraud to various DirectoratesGeneral of the Commission or to other EU institutions, bodies, offices or agencies for further action, OLAF must check whether they have, apart from a general competence to carry out system audits, the appropriate powers to undertake (possibly investigative) actions in individual cases. These checks are necessary especially in the EU staff sector, where OLAF has, in certain matters, sole competence, while in others it has shared competence with the EU institutions, bodies, offices and agencies, with some of which it has concluded agreements on the de minimis policy.

21.

The SC's overall assessment of the way in which selectors evaluated and applied the selection criteria revealed that the quality (in terms of completeness, clarity, consistency) of the motivation of opinions depends to a great extent on the individual approach and experience of selectors. To enhance the quality, OLAF should apply more rigorously the indicators it established for the assessment of these criteria and further develop and clarify some of them.

22.

The examination of the information flow during the selection process showed that in general OLAF is ready to provide information on dismissed cases to other EU or national authorities, while at the same time it would appear that sources of information are not systematically informed of OLAF's decisions upon completion of the selection process (see paragraphs 67-69 and 72).

23.

The SC finds it important for the persons or entities providing OLAF with initial information to be informed of the relevant action (not) taken by OLAF. Appropriate feedback encourages fraud reporting, and it is, at the same time, a strong transparency indicator. The SC shares the view of the European Ombudsman that providing the sources of information with reasons for decisions taken by OLAF as a follow-up to that information increases transparency and strengthens trust in OLAF's functioning (9).

24.

The SC also noticed an apparently low level of cooperation between the ISRU and the investigation units (or investigation support units) which receive very little information on dismissed cases. The SC would point out that, when a case is dismissed, the information held (sometimes exclusively) by the ISRU may still be of interest for the investigation units (e.g. to detect new fraud mechanisms), and they could also sometimes provide, thanks to their expertise, useful feedback for the selection unit and thus increase the efficiency of the selection process (see paragraphs 70-71).

25.

The SC noted that the conclusions of the opinions do not always clearly mention the actions that OLAF should take at the end of the selection process. When they are not completely omitted, some of these actions can only be deduced from other sections of the opinion (see paragraphs 67-69 and 72).

26.

The SC found instances where it had not been informed of obstructions that OLAF had encountered during the selection process. Moreover, OLAF did not inform the SC of cases dismissed where the opinions recommended that information should be transmitted to national judicial authorities. OLAF's obligation to inform the SC of ‘cases’ requiring information to be forwarded to the judicial authorities of a Member State (10) covers both cases opened as investigations and cases dismissed. Although such examples may be quite rare, the SC would highlight that such situations fall within its mandate and OLAF should thus provide the SC with appropriate information (see paragraphs 54 and 71).

27.

During technical meetings with OLAF in 2013, the SC presented some concerns arising from its examination of the ISRU's opinions and from the interviews conducted with OLAF staff, with regard to, inter alia, the time-frame for completing selections, the decreasing number of analysts within the ISRU and the division of tasks between selectors and reviewers. The SC notes that OLAF introduced some significant improvements during the period when this opinion was being drafted. The 2month time limit for selection was removed from the GIP. The internal structure of the ISRU was modified on 1 January 2014: the unit is now divided into 3 sectors (two sectors dedicated to the selection and one dedicated to the review), run by heads of sectors assisting the head of unit. The SC welcomes these improvements.

28.

Moreover, the SC wishes to underline that it is not its intention to substitute its own judgment on cases examined for that of OLAF, i.e. the SC is not judging whether individual cases should or should not have been opened. However, taking into consideration all of the above conclusions and recommendations and, in particular, the concerns with regard to the sufficiency of human and time resources allocated to the ISRU as well as with regard to compliance with the selection criteria, the SC believes that it would be of benefit for OLAF to carry out an internal evaluation of the activities of the ISRU, aimed at establishing, inter alia, the level of resources needed (number of staff and expertise), the strengths and weaknesses, the ‘error rate’ in evaluated cases (in particular those dismissed) and the relation between selectors and reviewers.

29.

The SC defined a number of indicators for the evaluation it carried out. To assess the efficiency of the selection function, the SC looked into the resources made available to the ISRU to carry out its tasks and the concrete results achieved in 2012 and 2013. The quality of the opinions was evaluated in the light of their conformity with the selection criteria established by the DG (11), while the transparency of the selection process was scrutinised by looking into the information flow throughout the selection process.

30.

Appropriate staff (number and expertise), sufficient time and clear instructions, appropriate technical and IT tools — those elements are critical for a successful selection process. Therefore, the SC has assessed the allocation of these resources to and within the ISRU.

31.

Number of staff — The ISRU had 16 selectors in 2012 and 2013 (12), who dealt with an ever increasing amount of incoming information (13). The average number of selections per selector was 74, meaning that an average of 3 working days was spent on one selection (14). An opinion was expressed in the interviews conducted by the SC that the staffing of the ISRU may be insufficient and that the heavy workload may affect the quality of the opinions. The SC does not have the necessary means and information to compare whether the performance of the ISRU is equivalent to that of the investigators who formerly carried out the assessment of the initial information prior to the OLAF reform. The SC believes that OLAF itself could and should perform such an analysis. (15)

32.

Technical and investigative expertise — In order to accomplish its mission, the ISRU needs to work within all the areas of OLAF's investigative competence. A wide range of expertise is covered by the staff within the unit, both through their education (economy, law, accounting/finances, biology, political sciences) and their professional experience (public finances inspectors, repression of corruption/fraud specialists, lawyers and magistrates, lecturers, auditors, customs officers, national police force, administrative assistants and secretaries). However, it appears from information provided by OLAF that some of the selectors do not have any investigative experience.

33.

The SC noted, from the examination of the opinions of the ISRU, that cases appear to be attributed to the selectors on the basis of their personal skills (competencies, language etc.). However, the distribution of cases per sector and per selector indicates that each selector provides opinions not only within their sectors of expertise. Statistical information extracted from the CMS showed that selectors without investigative experience dealt with a high number of cases — mostly dismissed — in very different sectors. Some of the opinions examined clearly reflected the lack of specific or investigative expertise of the selectors, who themselves sometimes acknowledged that they ‘are not specialist [in the matter analysed]’ (16).

34.

Legal expertise — In the opinions analysed, the SC notes the limited use of references to the relevant legal instruments (17). This is a clear indicator that the appropriate legal knowledge is either missing or not being sufficiently applied within the unit.

35.

Language expertise — The ISRU must be able to handle incoming information potentially drafted in all the 24 EU official languages, which is clearly impossible given the current number of selectors (18). As a result, the lack of language expertise has sometimes prevented the selectors from properly assessing the incoming information (19).

36.

The lack of appropriate language expertise has been supplanted by the attribution of cases for selection to the reviewers (20), based mainly on their language skills (21), leading to an increase in their workload, while they were already in charge of providing opinions to the DG on the legality of OLAF's key investigative activities (22), on the requests to extend the scope of cases (23) and on case closures (24).

37.

The SC believes that this may represent only a temporary solution of an exceptional nature, more especially considering that the workload of reviewers will likely continue to increase in the future given that the new GIP foresee that, in addition to the opinions provided under the ISIP, they will also need to advise the DG on interviews with witnesses (25) as well as to provide opinions on the requests to split and merge cases (26).

38.

Trainings — The SC would like to stress that the lack of previous experience or knowledge need not be an impediment to performing a good selection process, at least for ‘simple’ cases (e.g. cases clearly outside OLAF's competence, the so-called ‘prima facie’ non-cases or where allegations are not substantiated), as long as appropriate training has been provided. The SC was surprised to note that, on the one hand, the training courses followed by the selectors did not cover all the different sectors of OLAF's field of work while, on the other hand, the specialised courses were followed by only a very limited number of selectors (27).

39.

The percentage of selections which the selectors completed within the 2-month period foreseen in the ISIP (28) was 76 % (29). Some cases were dismissed (mainly on the grounds of the insufficiency of information), because it was not possible to acquire the necessary information within the fixed time limit imposed (30). At the same time, it appears that some cases which were previously dismissed have been opened later, on the basis of the information provided by the source after the fixed period (31). Moreover, it was ascertained during the interviews conducted by the SC that the assessment of the incoming information was affected by time pressure. These examples show that it was not only impossible, but sometimes also counterproductive to adhere strictly to the 2-month time limit, in the absence of a procedure, in the ISIP, allowing extending this period, in duly justified situations. The SC believes that a flexible approach is recommendable taking into account that sometimes the necessary supplementary information may emerge after the 2-month period or the source of information may take some time to reply to supplementary questions.

40.

The SC notes that Regulation (EU, Euratom) No 883/2013 does not impose a mandatory selection period, except for cases where a Member State concerned or an EU institution, body, office or agency requests that OLAF open an investigation and where a decision whether or not to open an investigation shall be taken within two months of receipt by OLAF of the request, otherwise OLAF shall be deemed to have decided not to open an investigation (32). The new GIP eliminated the overly rigid 2-month time limit, while at the same time maintaining strict deadlines for the Registry to register the incoming information and transmit it to the ISRU (33).

41.

Selectors carry out their tasks on the basis of instructions given by the DG, namely the ISIP (and, since 1 October 2013, the GIP) and of standardised work-forms (34). The SC has analysed these instructions and already provided the DG with a number of comments (35). The SC also makes some recommendations with regard to the work-form ‘Opinion on opening decision’ used by the selectors in Part I of this opinion.

42.

During the interviews carried out with OLAF staff it was stated, inter alia, that in all probability many cases are not reported to OLAF due to the lack of a clear procedure for dealing with whistle-blowers. The SC notes that the selectors are instructed to comply with general rules (36), but do not have specific instructions or any formal procedure for dealing with whistle-blowers (e.g. formal contact points, work forms, process for the selectors to contact the OLAF's Legal Advice Unit). As a result, different approaches have sometimes been noted amongst the selectors (37). While taking note of the increase over the past few years, of the number of cases where the source of information was a whistle-blower (figure 2), the SC believes that the adoption of clear and detailed guidelines on how to deal with them would be helpful for the selectors.

43.

Successful assessment of incoming information depends to a great extent on IT/technical tools available to the ISRU which is able to access various internal and external databases. The SC noted that, for the most part, the opinions make clear mention of the consultation of databases and of the results of the research carried out by the selectors.

44.

The IT tools made available to the public to report fraud are also important. One of them is the OLAF Fraud Notification System (FNS), a web-based information system that may be used to submit information to OLAF, and through which the selectors may communicate with the source of information (38). On several occasions, the opinions of the ISRU mention that the informants have tried to send documents through the FNS, but these documents have not been received because of the restrictions imposed by the FNS, which blocks the upload of overly large files or limits the number of characters that can be used, without, however, informing the sender.

45.

The selection process consists of a step-by-step application of three selection criteria: OLAF's competency to act, the sufficiency of information to open an investigation or coordination case and the Investigative Policy Priorities (IPPs) established by the DG (39). If the first selection criterion (or the first and second selection criteria) is not fulfilled, the case is dismissed and the information assessed no further (40). Figures 3 and 4 show the distribution of cases dismissed and opened by sector.

46.

The ISIP and the IPPs contain instructions on how to apply each of the selection criteria. The SC's review of the opinions was thus aimed at assessing their conformity with these instructions.

47.

The SC notes that the lack of OLAF's competence to act, being the reason given for the dismissal of 15 % of the cases (figure 5), was in general well explained (42). On the other hand, little consideration was given to the relevant legal instruments in those cases where OLAF was considered to be competent to act (43). Moreover, the assessment of OLAF's competency to act is not sufficiently substantiated: in many opinions there is a general and unsubstantiated reference to a potential impact on the EU financial interests or to the fact that EU funds were involved (44); in some opinions it is just stated ‘yes’ in the parts referring to the potential impact on the EU financial interests/potentially serious matter relating to the discharge of professional duties (45); other, very few, opinions only mention that ‘OLAF is competent’ without giving any further explanation (46).

48.

In general, the SC notes that the quality of the assessments varies according to selectors and their experience in specific sectors: opinions concerning customs, cigarettes and EU staff are comprehensive and generally contain a clear explanation as to the irregularity arising from the allegations, the way it impacts the EU budget and the amount at stake (when determined).

49.

The SC also notes that the work-form ‘Opinion on opening decision’ used by the selectors includes two different options for referring to the potential impact on the EU financial interests and to possibly serious matters relating to the discharge of professional duties: one in the ‘summary’ part of the work-form, where the selectors need to tick the relevant boxes, and one in the body of the opinion (part 3 of the work-form), where the selectors need to assess in concreto OLAF's competency to act. This double option might have created some confusion amongst the selectors, since some of them seem to assume that ticking the boxes in the ‘summary’ part exempts them from carrying out a concrete analysis of OLAF's competency to act in Part 3 of the opinion, which should not be the case.

50.

The insufficiency of information was the main reason for dismissing cases (figure 5). The SC therefore paid particular attention to the grounds on which the selectors based their conclusions.

51.

The SC identified opinions where the assessment of the sufficiency of the information (reliability of the source and credibility of the allegations) is unsubstantiated (48), incomplete (49) or inconsistent (50). Equally, little consideration was given to a third parameter emerging from the EU case-law (sufficiency of suspicions), which was rather rarely used, mainly because it was not mentioned in the ISIP or in the workform used by the selectors (51). Moreover, it seems to have various meanings for the selectors: most of the time it was considered that the information was insufficient due to lack of ‘clear indications of fraud’ or ‘absence of fraud elements’ or because there was ‘no concrete information as regards possible irregularities or fraud’ (52); in other cases the (in)sufficiency of suspicions was determined by the quality of the information at OLAF's disposal (53), sometimes the ‘seriousness of the allegations’ was also an element taken into consideration (54). Bearing in mind the existing difference between the ‘seriousness of the allegations’ and the ‘seriousness of suspicions’, the SC would point out that this latter parameter should not have been considered as optional, since the EU case-law has established the ‘sufficiently serious suspicion’ as a precondition for the opening of an investigation (55). The new Regulation (EU, Euratom) No 883/2013 has now explicitly incorporated this requirement (56) and this change is reflected in an amended work-form accompanying the new GIP. The SC welcomes this improvement.

52.

The SC believes that the deficiencies noted above could be explained, at least partially, by the fact that neither the ISIP nor any other internal document (57) provides the selectors with precise indicators for implementing this criterion, namely concrete situations when a source of information can be considered reliable (or not), when the allegations put forward are credible (or not) or when there are sufficient suspicions of fraud or irregularities (or not). Appreciation of these elements depends, therefore, very much on the personal approach of each selector, based on their investigative experience and specialised knowledge.

53.

The SC was, however, able to identify a number of concrete indicators used in some of the opinions. The reliability of the source was evaluated by using indicators such as (i) the verifiability of the source (anonymous or not), (ii) the type of the source (natural person; EU institution, body, office, agency or a Member State or third country authority; OLAF itself), (iii) the (in)direct knowledge of or connection with the matter reported to OLAF, (iv) the possibility for OLAF to contact the source to request further information (v) the degree of cooperation between the source and OLAF, (vi) the trustworthiness of the source, (vii) the motivation of the source and potential degree of subjectivism. As to the assessment of the credibility of the allegations, their appreciation was based on indicators such as (i) the quality of the initial information provided to OLAF (58), (ii) its verifiability (59), (iii) the context.

54.

On the other hand, the SC identified a number of objective reasons explaining the insufficiency of information gathered during the selection process, such as (i) the imprecision of the initial information, which did not allow further research; (ii) technical problems with the FNS preventing the reception of documents; (iii) time pressure of the 2-month period for selection, which did not allow in-depth searches for additional information; (iv) lack of response from stakeholders (EU, national or third countries authorities, sources of information) or even obstruction on their part (60). The SC regrets that, in the latter case, such obstruction was not reported to it.

55.

Each of the investigation policy priorities established by the DG — proportionality, efficient use of investigative resources, subsidiarity/added value and special policy objectives/criteria for 2012/2013 — can be assessed on the basis of the concrete and measurable indicators which are clearly indicated in the IPPs. The SC's review is aimed at assessing the way in which the selectors applied those indicators. (62)

56.

In general terms, the proportionality principle requires that there be a reasonable relationship between an objective to be achieved and the means used to achieve it. Applied to OLAF's decisions, these require an appropriate equilibrium between, on the one hand, the investigative means to be deployed by OLAF if a decision to open an investigation or coordination case is taken and, on the other, the results that are expected. This approach is well reflected in the IPPs, which provide the selectors with clear indications as to the elements that need to be weighed against each other, resulting in a ‘fair return for [OLAF's] efforts’.

57.

The SC notes however that there are a limited number of cases where the indicators for proportionality, as mentioned in the IPPs (likelihood of recovery and of prosecution, possible deterrent value) are clearly indicated and where a balance is struck between the number and type/complexity of the investigative activities needed to be carried out and the expected amount to be recovered/the likelihood of action to be taken by a competent authority, or between the relatively low financial impact of the alleged irregularities and the workload of the investigative unit (63). However, the indicators contained in the IPPs were not rigorously applied in some cases: investigations were opened on the basis of only the two first selection criteria (64); the proportionality test consisted of unsubstantiated statements (65) or in remarks with regard to a different policy criteria (66); reference to a financial cost/benefit ratio of an investigation to be carried out was incomplete (67).

58.

The SC notes that the evaluation of the workload of the investigation units is often missing (68), unsubstantiated (69), inconsistent (70), or makes reference to indicators such as proportionality or the added value of OLAF's action (71). When verification in concreto of the workload of the investigation unit is carried out, the selectors mention the number of cases per unit and the number of investigators per unit, without making any estimate as to the size of the workload of the investigation unit (72). In the few cases where such an estimate was made, the heavy workload of the investigators was a reason for dismissing them, but always used in conjunction with subsidiarity reasons (73).

59.

The number of opinions making reference to the possibility to carry out activities continuously and without slowing-down on-going or priority investigations is also very limited (74), while the availability of expertise within a specific unit seems to be better evaluated (75).

60.

Few opinions make explicit reference to an internal consultation as to the availability of the human resources and/or expertise within the investigation units (76). From the interviews conducted by the SC it is also clear that the investigation units are only rarely consulted regarding the resources situation.

61.

The SC notes that the criterion of subsidiarity/added value represents the main reason used for advising the dismissal of a case on the grounds outlined in the IPPs, taken alone or combined with other reasons (77). Apart from a small number of opinions giving it no consideration (78) or unsubstantiated (79), the reasons outlined as to why an OLAF action would not bring any added value in terms of recovery, prosecution or deterrence to the control activities carried out by other EU or national bodies, which are better placed to act, are generally well explained (80).

62.

The SC received information, from other Directorates-General of the Commission, pointing out that OLAF is dismissing cases, while at the same time forwarding them for action to those Directorates without, however, checking whether they had the necessary competence and powers to act. The SC notes with concern that the Directorates-General of the Commission, which have been considered by OLAF to be better placed to act and thus receive from OLAF information on dismissed cases, are not required to report back to OLAF on actions taken (81).

63.

Moreover, the SC notes that cases concerning Members of the European Parliament (EP), where OLAF has clearly competence to act (82) were dismissed on the grounds of subsidiarity, although it was stated that in similar cases the EP did not report back to OLAF, despite an explicit request, on the action taken with regard to the information sent by OLAF (83). The SC would draw attention to the Practical Arrangements recently agreed between the EP and OLAF (84) which do not include such situations on the list of activities which usually/probably do not lead to serious situations requiring OLAF investigations and where OLAF is presumed not to intend to open an investigation or where there is a high probability that OLAF will not open an investigation (85).

64.

According to the IPPs, OLAF will focus its investigative activity on sectors considered to be a priority and also on those cases where financial indicators determine the seriousness of fraud (86). When it comes to the prioritised sectors, the SC noted that the reference to them was quite rare (87). Similarly, the financial indicators were not always used: in some cases there is no reference at all, even when the financial impact is known (88). When they are referred to, mainly in sectors such as EU staff or structural funds, they are used either to dismiss cases on the grounds that the financial impact was non-existent or too low, taken alone or in conjunction with other IPPs, or to recommend the opening of an investigation or coordination case. The SC noted that the financial indicators were not always a determining factor when proposing to dismiss or open a case: when the financial indicators correspond to the IPPs, the case may, however, be dismissed on subsidiarity/added value grounds (89) or, conversely, when a case does not fall within the special policy objectives it may however be opened as an investigation if the other selection criteria are fulfilled (90).

65.

As a general remark, the SC would point out that the financial indicators should not be used as thresholds for justifying the dismissal or opening of cases, but rather as an indicator for estimating or measuring the seriousness of the fraud risk involved (proportionality test) (91).

66.

A centralised system where the number of OLAF staff aware of specific incoming information is very limited (92) may lead, in the SC's opinion, to a lack of transparency and of accountability in the decision-making process, especially with regard to the cases dismissed. It should be balanced by a procedure for providing the appropriate information to the relevant EU or national authorities with which OLAF shares competence in the antifraud fight, to the investigation directorates and also to the sources providing OLAF with the initial information. The SC therefore examined the transparency of the selection process in the light of these three information-sharing requirements.

67.

Whenever an EU or national authority is better placed to deal with a case, OLAF should transfer it there. Such follow-up is necessary for every case where a sufficient suspicion of fraud has been established, but where OLAF decides not to open an investigation. Otherwise, not only would areas of impunity be created, but the transparency and coherence of the selection process would also be compromised.

68.

Opinions advising dismissal of a case on the grounds of subsidiarity/added value usually indicate that information should be forwarded to the authority considered better placed to act (93). However, it is not always clearly stated in the ‘conclusions’ part of the opinion, but can sometimes only be deduced from other sections.

69.

Another important aspect that OLAF needs to take into account when transferring information to a competent authority is the necessity to protect the identity of the source, in particular of whistle-blowers. The SC notes the very limited number of opinions recommending non-disclosure of the identity of the source to those authorities, sometimes upon request by the source itself (94).

70.

During the interviews with OLAF staff, the lack of transfer of information on dismissed cases to the investigation units was described as inhibiting the exchange of knowledge and experience. On the other hand, the SC was informed by OLAF notes (95) that there was a regular contact between the ISRU and the investigation units during the selection phase, principally in order to check the availability of investigative resources in a specific unit, or connections between new incoming information and already existing cases.

71.

However, the opinions reviewed by the SC do not reflect regular contacts. Few opinions make explicit reference to an internal consultation (96) or to the forwarding of information on a dismissed case of possible interest for an on-going investigation to the investigator in charge (97). This lack of consultation is even more evident in a few other cases where the selectors faced difficulties in carrying out searches or assessing information drafted in a language with which they were not familiar and where the investigation units (possessing the relevant language expertise) were never consulted (98), as well as in instances where investigation/coordination cases similar to or connected with the information being assessed were identified (99). That situation could raise problems in cases where it is recommended to send the information under assessment to a competent national judicial authority but not to the investigation unit dealing with the connected case, or to the SC (100).

72.

The SC notes that the opinions of the ISRU do not systematically propose informing the source of information of OLAF's decisions (101). While this is understandable with regard to decisions to open investigation/coordination cases (for confidentiality reasons) (102), it may be questionable with regard to dismissed cases.

14.6: on-the spot checks,

15.2: digital forensic operations within inspections or on-the-spot checks,

16.2 and 16.5: interviews,

18.1 and 18.2: the opportunity to provide comments,

19.10: the final report and proposed recommendations.

1.

The European Antifraud Office (OLAF) and its Supervisory Committee (SC) have been established by Commission Decision of 28 April 1999 (1). The scope of supervision has been reinforced and specified by Regulation No 883/2013 (2) of 11 September 2013. The SC is the guarantor of OLAF's independence and it supervises the investigatory function of OLAF through regular monitoring aimed at ensuring the proper conduct of investigations. The SC focuses in particular on examining whether fundamental rights and procedural guarantees are respected in OLAF's investigations and whether the cases are dealt with efficiently, effectively, in due time and according to the relevant rules and legal provisions.

2.

In its Annual Activity Report 2012 (in particular in Section 2 of Annex III), the SC recommended, on the basis of its monitoring experience and in view of the identified shortcomings, a number of actions aiming at reinforcing procedural safeguards in OLAF's investigations. The SC recommended in particular introduction of transparent and stable procedures for the internal legality check and for independent review of complaints. The SC recommended also clarification of OLAF's powers in different types of administrative investigations and insisted on providing the SC with effective tools for monitoring the respect of procedural guarantees and fundamental rights by OLAF.

3.

That need to strengthen the procedural safeguards and legality checks seems to be a common conclusion of the European Parliament, the Council, the Commission, the Court of Auditors and the SC as expressed in their exchanges of opinions on OLAF investigative activities.

4.

Such strengthening should be considered in view of the Commission's proposal for the establishment of the European Public Prosecutor's Office (EPPO). Creation of the EPPO would be a change of paradigm by transferring cases of possible criminal fraud from administrative investigations in OLAF to criminal investigations conducted by European prosecutors. Such change is to result in a substantial reinforcement of the procedural guarantees for persons concerned by investigations.

5.

Even with the establishment of the EPPO, OLAF would still have a significant role to play in the protection of the EU against offences and irregularities affecting its financial interests. There is a high degree of uncertainty when it comes to the geographical coverage of the EPPO which most probably will be established by enhanced cooperation of some Member States — their number remains unknown, but almost surely not all of them are going to participate. Therefore investigations of similar nature might be in future conducted in parallel by the EPPO and OLAF, depending on a Member State, which would require strengthening the procedural safeguards in OLAF's investigations, so that they could match the foreseen EPPO standards.

6.

Therefore, the SC welcomes with satisfaction the Commission's Communication on Improving OLAF's governance and reinforcing procedural safeguards in investigations  (3). The SC is currently analysing possible consequences of particular solutions proposed there and is looking forward to discussing them with the Institutions and with OLAF. As the first reflection, the SC considers the substance of the proposals as positive, providing for instruments potentially enabling to improve the current level of safeguards. At the same time, some structural solutions, as proposed in the Communication, should be reconsidered with particular regard to the independence of OLAF and of the SC, to ensure avoiding conflicts of competences with respect to the supervision of OLAF as well as duplication of work and inefficient allocation of resources.

7.

At this stage, the SC is considering specifically the issues of the institutional framework for two new offices proposed by the Commission and the scope of their competences.

8.

The Communication advocates establishment of a new office of a ‘Controller of procedural safeguard’ who would ‘monitor compliance with the procedural guarantees applicable to OLAF investigations and of prompt handling of investigations to avoid undue delay’. It is very hard to see how the Controller could avoid having his work overlapping with the work of the SC which would ‘monitor systemic developments regarding respect of procedural rights and reasonable deadlines for handling cases’.

9.

Even if the differentiation between the tasks of the Controller and of the SC could be possible argued on a theoretical level, the practical experience of the SC shows that review of individual cases is an indispensable element of systemic monitoring of OLAF. Therefore, the daily work — in the separate secretariats of both bodies — would be very similar and consist in examination of individual case files in view of respect of procedural rights and duration of investigation.

10.

In its systemic analyses based on samples of individual cases, the SC could be examining, coincidently and even unknowingly, the same case as would be examined by the Controller in the framework of his individual review. It could lead not only to redundant duplication of work, but also to issuing diverging or even conflicting recommendations to the Director-General of OLAF.

11.

Furthermore, the Commission proposes to establish an office of a judicial reviewer who would authorise OLAF's intrusive investigative measures concerning Members of EU Institutions. As a result, also his work (as the SC's and Controller's work) would concern examination of the respect of procedural/fundamental rights (although ex ante and not ex post).

12.

The reviewer's competences, his punctual interventions and his placement within or next to the Commission could raise serious concerns as regards OLAF's independence. In view of his placement, his competence to advise on investigative measures against Members of the Commission but also of other Institutions could affect the interinstitutional balance.

13.

Also as regards the Controller, who would be similarly placed and separated from the SC, there could be concerns with regard to his independence and to the cost effectiveness of him having a separate secretariat doing the job currently done by the SC's Secretariat.

14.

Creation of two additional offices controlling OLAF but separate from the Supervisory Committee of OLAF would lead to multiplication of independent supervising structures, probably resulting in confusion or conflict of competences and duplication of work. That could ultimately decrease the efficiency of the supervision of OLAF.

15.

The SC is considering other options which would retain the useful instruments proposed by the Commission, but which would at the same time incorporate them into a comprehensive and effective supervision structure.

16.

To achieve the important aim of reinforcing the procedural safeguards in OLAF set forth by the Communication, the Controller should form a part of a wider supervisory committee. His independence would be strengthened and he could benefit from the expertise of the SC and its Secretariat having a long experience in examining the respect of procedural guarantees in OLAF. The whole joint structure would be stronger, more efficient and would produce an effect of synergy. A common secretariat would ensure consistency, economy of scale and allow avoiding duplication of work.

17.

As regards the judicial reviewer, he could be, theoretically, also attached to a wider supervisory committee to benefit from the knowledge and independent resources of the already established structure. On the other hand, the judicial reviewer being attached to the SC (or to the Commission) would implicate the SC (or the Commission, respectively) in the decision-making process in OLAF which could jeopardise OLAF's independence.

18.

Therefore, it seems more logical to place the judicial review back in a dedicated judicial unit within OLAF which would consist, as it used to be the case, of national magistrates. They could, without putting OLAF's independence in danger (as it could be the case with an external body), provide the Director-General, rather expeditiously thanks to their presence within OLAF, with independent and formal legal advice, ex ante, on all intrusive investigative measures and other actions requiring compliance with specific national provisions. The SC would continue to monitor, ex post, the judicial recommendations to the Director-General and his ensuing decisions, with a view of ensuring OLAF's independence, on the one hand, and the procedural rights of persons concerned, on the other.

19.

In the light of its monitoring experience, the SC supports the Commission's proposal to reinforce the procedural safeguards in OLAF investigations. It is particularly recommendable having regard to the abolition in the reformed OLAF of the SC's prior examination of the respect of fundamental right and procedural guarantees before an OLAF case is sent to national judicial authorities — which was considered by the Court of Justice as a crucial safeguard for persons concerned. The improved supervisory structure must be functional and efficient. The SC is looking forward to the forthcoming exchange of opinions between the Institutions under the new OLAF Regulation which could focus on working out optimal and broadly supported legislative solutions.

1.

1.1.

1.2.

1.3.

1.4.

2.

3.

3.1.

3.2.

4.

1.

In the light of the mission and tasks entrusted to it by Regulation (EU, Euratom) No 883/2013 (1), the Supervisory Committee (SC) of the European Anti-fraud Office (OLAF) hereby defines its strategic objectives and priorities for the first half of its five years mandate (as of 23 January 2012).

2.

The mission of the SC, as outlined by Regulation (EU, Euratom) No 883/2013, is to reinforce OLAF's independence in the proper exercise of the competences conferred upon it (2). To accomplish this mission, the SC was entrusted with a threefold role:

3.

The core tasks of the SC are also defined by the Regulation from this triple perspective.