The European Personnel Selection Office: Time to adapt the selection process to changing recruitment needs

Executive summary

I

Every year, the EU institutions recruit around 1 000 new permanent staff members for long-term careers, from more than 50 000 applicants. These new staff members are selected from the successful candidates of open competitions run by EPSO, the EU civil service’s personnel selection office, in cooperation with the EU institutions.

II

EPSO is responsible for attracting and selecting suitable candidates in sufficient numbers to enable the institutions to meet their recruitment needs. EPSO runs two main types of competition: large competitions for entry-level position for candidates with generalist profiles, such as public administrators, lawyers, economists, translators and secretarial assistants, and smaller competitions for candidates with more specific profiles and experience. The selection process has three phases: planning the competitions for the coming year, assessing the candidates in each competition through a series of tests, and publishing the final lists of successful candidates of each competition.

III

EPSO’s competition-based selection process has been running in its current form since 2012. The language regime under which competitions operated has been challenged in court, and the EU institutions have begun to question the effectiveness of the selection procedure. In this context, EPSO’s management board launched an interinstitutional reflection group on the selection process in the second half of 2019. Our report provides further analysis, conclusions and recommendations of relevance for a possible significant update of the selection process.

IV

The objective of our audit was to assess whether the EPSO selection process has enabled the EU institutions to meet their recruitment needs for all types of staff. In particular, we examined whether the EPSO selection process:

1. was planned adequately to take account of EU institutions’ recruitment needs;
2. was accompanied by effective communications to attract sufficient candidates;
3. produced successful candidates suited to the jobs offered by the institutions;
4. was run in a timely manner;
5. was run in a way that minimised costs.

V

The audit focused on EPSO’s activities in 2012-2018. We reviewed EPSO’s management information; interviewed staff from EPSO and the human resources services in some EU institutions; reviewed a sample of open competitions to analyse their costs and duration; and carried out a survey of managers in some EU institutions.

VI

We concluded that the large competitions run by EPSO broadly enabled the institutions to meet their recruitment needs for candidates with generalist profiles during the period. However, the process proved less efficient and effective at meeting the institutions’ smaller but growing need to recruit specialists.

VII

We found that the selection process is broadly effective for large-scale competitions, for the following reasons: the planning is suited to these competitions; EPSO manages to attract a high number of candidates and promotes the ‘EU Careers’ brand mainly in Brussels; despite a number of weaknesses in the selection process, the overall quality of the successful candidates is high; in years yielding the highest number of successful candidates, the cost of competitions is low; and holding regular competitions for candidates with generalist profiles has helped to ensure an adequate supply of potential recruits.

VIII

We also found that the selection process is not adapted to small-scale, targeted competitions, which are those most suited to the current recruitment needs of the EU institutions: the planning of these competitions is not reliable, and takes place too early to be accurate; EPSO does not consistently target its communications towards attracting suitable candidates to specialist competitions; candidates with strong specialist profiles risk being eliminated early in the selection process; the cost of competitions is higher than the institutions’ own alternative selection procedures when only a small number of specialists are needed; and specialist competitions are less suitable to fill urgent recruitment needs than the institutions’ own procedures.

IX

Based on our observations, we make recommendations for:

• strengthening key aspects of the selection process, in particular measuring the institutions’ satisfaction, addressing the continuous issues of the language regime, and the coordination between EPSO and the institutions;
• introducing a new selection framework for specialist competitions;
• improving EPSO’s capacity to adapt to a fast-changing recruitment environment, by introducing a mechanism for regularly reviewing its selection process.

Introduction

01

Every year, the EU institutions recruit around 1 000 new permanent staff members (officials) for long-term careers, from more than 50 000 applicants. These new staff members are selected from the successful candidates (“laureates”) of open competitions. The quality of the output of the selection process determines the quality of recruited officials, who often spend their whole careers in the EU institutions.

02

Since 2003, these competitions have been organised by the European Personnel Selection Office (EPSO), in cooperation with the institutions. EPSO's role is to serve the EU institutions by providing high quality, efficient and effective selection procedures that enable them to recruit the right people, for the right jobs, at the right time¹. EPSO is responsible for attracting – through appropriate communications work – and selecting suitable candidates in sufficient numbers to enable the institutions to meet their long-term recruitment needs. These new staff members can be broken down into three categories: generalists (public administrators, lawyers, economists etc. as well as secretarial assistants), linguists (translators) and specialists (IT experts, scientists, etc.).

03

Generalist competitions, aiming at recruiting generalists and linguists, attract candidates in their tens of thousands, and some languages can attract up to 2 000 applicants for translator position; these competitions can have reserve lists of up to 200 laureates; and are interinstitutional in nature. In this report, we refer to them as large competitions. Specialist competitions (around 42 % of the laureates identified between 2012 and 2018) are more targeted, less often interinstitutional, and the number of applicants per field is more limited.

04

EPSO is an interinstitutional office providing staff selection services to all EU institutions. At the same time, it is administratively attached to the European Commission. EPSO’s management board is its highest decision-making body. Its members are the director of EPSO and a representative at senior management level from each institution. EPSO’s management board agrees by qualified majority on the principles of the selection policy (which are applied in the selection process), the management of lists of laureates (“reserve lists”, see Annex III and Glossary), and unanimously approves EPSO’s work programme, including the scheduling of and timetable for competitions. The work and decisions of the management board are prepared by the EPSO working group, where all the institutions also have a representative, at a lower level.

05

The overall budget of EPSO was €23.6 million in 2019. EPSO’s budget has remained stable over the years. Around half of EPSO’s budget (€12.8 million in 2019) is spent on its staff (around 125 people, most of them permanent officials). EPSO’s activities are presented in Figure 1.

06

The EPSO selection process has been running in its current form since 2012². It has been challenged in court (cases C-566/10, T-124/13, T-275/13, T-353/14 and C-621/16) for its language regime, and the EU institutions’ dissatisfaction about its selection process has led EPSO’s management board to launch an interinstitutional reflection group on the selection process in the second half of 2019.

07

The selection process has three phases: planning, assessing candidates, and publication of reserve lists. This process starts with the assessment of the institutions’ recruitment needs for the next three years. EPSO gathers these estimates to propose a detailed schedule, with a proposed timeline of publication for competitions, for the coming year. It is approved by the management board.

08

Recruitment needs for generalists and linguists (see Annex I and Annex II) are expressed as total numbers of new staff members required. The institutions are also invited to provide further details such as the languages which will be requested, or, if they intend to recruit specialists, the specialisms which will be required (e.g. security, scientific research) along with an estimate of the number of staff needed.

09

Once a competition has been approved, EPSO and the institutions concerned draft the notice of competition. This notice sets out the legal framework of the competition: the intended number of laureates, the minimum conditions which candidates must satisfy in order to be eligible, including the required level of professional experience when applicable, the tests which will be used during the competition, and the pass marks which will be required for each of them.

10

The institutions also designate the selection board members, who are responsible for validating the content of the tests and assessing candidates throughout the competition (as provided for in the EU Staff Regulations).

11

The tests vary from one competition to another, but the stages within each competition are usually similar (see Annex III).

• Pre-selection: competitions for generalists, and other competitions when they attract more than 1 000 applicants, begin with a pre-selection stage composed of computer-based psychometric tests (multiple-choice questions on verbal, numerical and abstract reasoning). For specialists, this can sometimes be replaced by a CV sift.
• Admission: the application files of the best-scoring candidates of the pre-selection stage are then reviewed by the selection board.
• Talent screener: for some competitions, the admission stage can be followed by a review of candidates’ CVs, with marks given depending on candidates’ qualifications and experience.
• Intermediate test: this stage is not compulsory for all competitions, and its use depends on the institutions concerned. It can be administered on its own, or as part of the next stage.
• Assessment centre: the best-scoring candidates from the previous stages then proceed to the assessment centre, where the selection board assesses them against a set of predefined skills and competencies. For candidates that did not sit the pre-selection psychometric tests (specialist competitions, or when the number of applicants is low), these are done at the assessment centre.

12

The names of the candidates who perform best at the assessment centre are put on a reserve list. The recruiting services in the institutions can then access their CVs and contact the laureates they think are the most suitable to meet their recruitment needs. EPSO monitors the use of the reserve lists by the institutions. If many laureates stay on a list without being recruited for many years, EPSO proposes to the management board that the list should be closed.

13

Figure 2 shows all the stakeholders playing a role in the selection and recruitment processes.

Audit scope and approach

14

We first audited EPSO in 2009 to assess how well it had fulfilled its mandate in the years following its creation. Ten years later, we decided to conduct a new audit of EPSO. In this report, we address EPSO’s response to the observations presented in 2009, and how it has adapted to the rapidly changing environment in which the EU and its institutions evolve (e.g. staff cuts, digitalisation).

15

The objective of our audit was to assess whether the EPSO selection process has enabled the EU institutions to meet their recruitment needs for all types of staff. In particular, we examined whether the EPSO selection process:

1. was planned to take account of EU institutions’ recruitment needs;
2. was accompanied by effective communication actions to attract sufficient candidates;
3. produced successful candidates suited to the jobs offered by the institutions;
4. was run in a timely manner;
5. was run in a way that minimised costs.³

16

In the course of this audit, we did not assess:

• the quality of test content designed by EPSO;
• the institutions’ internal policies and procedures for recruitment;
• the institutions’ training programmes for new recruits.

17

The audit focused on EPSO’s activities in 2012-2018. Our report provides analysis, conclusions and recommendations of relevance for EPSO’s 2020-2024 strategic plan. We based our findings on the following evidence:

• a documentary review of key documents related to EPSO’s activities;
• audit interviews with EPSO’s and other EU institutions’ staff;
• a review of a sample of open competitions organised between 2012 and 2018;
• a survey of middle managers in EU institutions.

18

The sample of competitions was selected to cover EPSO’s range of activities, and includes all of the types of competitions organised by EPSO during the 2012-2018 period. The aim of the managers’ survey was to gather the managers’ feedback on staff members recruited from EPSO’s open competitions and from other selection channels, to compare the merits of the different selection processes. The survey was sent to 471 managers, with an overall response rate of 58 %. More details on the audit methodology are presented in Annex IV.

Observations

19

To examine whether EPSO’s selection process was identifying the right people, for the right jobs, at the right time, in a way that minimised costs:

• we reviewed all the stages of the selection process, from the planning of competitions to the publication of reserve lists;
• we compared the effectiveness and efficiency of the selection process for both large and specialist competitions.

Large competitions run by EPSO broadly deliver the right people for the right jobs on sound financial terms

20

EPSO was set up mainly to organise large-scale competitions for generalists (for both the administrators – AD – and assistants – AST and AST/SC – function group, see Annex I) and linguists to provide the EU institutions with a sufficient number of recruits to meet their recurring recruitment needs. We would expect EPSO, if it is to carry out this task efficiently, to:

1. use the institutions’ estimates of their need for new staff members with these profiles to plan its competitions schedule;
2. implement a communication strategy likely to attract candidates suited to these profiles;
3. use appropriate testing methods;
4. ensure it runs competitions in a way that minimises costs;
5. ensure that reserve lists are published in time to meet the institutions’ recruitment needs.

The planning of competitions is well-suited to large-scale competitions

21

The objective of EPSO’s planning process is to gather the recruitment needs of the institutions, and, based on them, put forward a calendar of competitions to be organised the following year to ensure that those recruitment needs are met.

22

The current form of the EPSO planning process was introduced in the EPSO Development Plan (EDP), which was adopted in 2008⁴. The EDP introduced annual competitions cycles for three staff profiles: generalist administrators (mainly designed to recruit junior staff, and targeted at young university graduates with little or no professional experience), assistants (AST and AST/SC function groups), and linguists.

23

These annual cycles correspond to recurring interinstitutional recruitment needs. Each cycle is planned to start at a specific time in the year: for example, registration for generalist administrators starts in September, but registration for assistants starts in December. Specialist competitions (for staff profiles such as lawyer-linguists) and ad-hoc competitions (for specific profiles such as management positions) are organised in the time-slots between the three main cycles. Figure 3 illustrates the annual planning cycle.

24

To organise the competition cycles, EPSO asks the participating institutions how many generalists (administrators and assistants), and linguists they estimate that they will need to recruit in the coming year. It also asks the institutions to indicate what type of specialist staff members they will seek to recruit.

25

The EPSO working group analyses the needs expressed by the institutions and on this basis, it determines which competitions will be run in the coming year. EPSO does not organise all competitions requested by the institutions (annual cycles, specialist, and ad-hoc competitions) every year. The working group gives priority to the annual cycles, and once these have been planned, to other interinstitutional competitions depending on the number of laureates requested, and on the number of laureates still available on reserve lists for similar profiles.

26

Generalist profiles represented 37 % of all requests for laureates, over the 2012-2018 period, with linguists an additional 20 %. Despite a decrease over time, as illustrated in Figure 4, generalist and linguist profiles still make up a large part of the institutions’ requests for laureates, which is reflected in their importance in the planning process. Nevertheless, they no longer make up the majority of new staff members sought. The demand for translators especially has fallen in recent years, as a result of decisions to recruit fewer permanent staff members for translator positions, increased outsourcing, and technological progress (machine translation).

27

After they have told EPSO how many laureates they estimate they will need, the institutions tend to revise those estimations, usually downwards, when the competitions are being organised. On average, we found that the institutions’ estimates are relatively accurate for the profiles sought in the annual cycles: the required number of generalists was overestimated by 11 %, and linguists by 18 %. The projected need for specialists was less accurate: required numbers were overestimated by 33 %.

28

In our view, the planning process is well suited to the organisation of the annual cycles of competitions for generalist administrators, assistants and linguists.

• It is designed mainly around the gathering of the needs for these profiles.
• The level of precision in the estimates of recruitment needs is adequate for these competitions. Since the needs are recurring, the estimates do not need to be very precise.
• Based on these needs, EPSO is able to put forward a calendar that ensures that laureates are available on a regular basis.

EPSO attracts a high number of candidates and promotes the ‘EU Careers’ brand mainly in Brussels

29

Each competition has a profile for the applicants being sought, based on the minimum requirements of the Staff Regulations in terms of education level and professional experience, and the needs of the requesting institution (see Annex I and Annex II). Some competitions are organised specifically for one institution, while others are focused on one particular location. Communication work to identify and attract the right applicants is EPSO’s responsibility. However, EPSO also has to take into account external factors, such as the overall attractiveness of the EU institutions in some countries, when designing its communication actions.

30

The “EU Careers” brand was created in 2010, partly as a way to improve the attractiveness of the EU as a graduate employer in Member States (see Special Report 15/2019). The EU Careers brand relies on a network of ambassadors from the institutions (staff members participating in EU Careers events, called staff ambassadors) and from universities (students in charge of disseminating EU Careers communications in their university, called student ambassadors).

31

The EU Career brand is also supported by university career services, and by contact points in Member States. EPSO also organises targeted events for “focus countries”, i.e. countries whose nationals are underrepresented in competitions or among EU staff. Some of these events include helping public authorities from these countries set up strategies to promote the EU institutions as employers and advertise competitions. The impact of these actions on the geographical balance of applicants to competitions is not yet visible.

32

The content produced for EU Careers is focused on the EU as an employer, not on specific competitions. It includes, for instance, information for potential candidates with disabilities or special needs to ensure that they are aware of the possibilities offered by the institutions and the accommodations available to them when they take part in competitions.

33

With the creation of the EU Careers brand, EPSO increased the number of career and job-fair events it organised each year, from 52 in 2012 to 109 in 2018. Most of EPSO’s communication budget was also shifted to EU Careers, and staff and student ambassadors’ events, as illustrated in Figure 5. Most (69 % on average) of these events take place in Brussels, where there is already a certain awareness of the employment possibilities offered by the EU due to the presence of the institutions and various interest groups (local and national representations, private consulting and lobbying companies, etc.). 37 % of the respondents to EPSO assessment centre surveys declared a direct link to the institutions: they were either working within the institutions, or working in the public sector in roles related to the EU⁵.

34

Events outside Brussels are mostly organised by the Member States, sometimes with some support from EPSO. Similarly, the list of universities covered by the student ambassadors programme is drawn up by EPSO, but it is based on suggestions from Member States, with social sciences, law and economics faculties represented heavily. Irrespective of the nationality of applicants, Belgium and Luxembourg remain the countries where most respondents to EPSO’s assessment centre surveys start the competition process, with two-thirds of them sitting the pre-selection tests there.

35

There are indications that the EU Careers brand has not been successful in attracting a sufficient number of graduates or young professionals to an EU career. Applicants to AD5 competitions (the most junior grade for graduate administrators) often have professional experience, and the proportion of candidates under the age of 35 is decreasing. This is particularly striking for some focus countries: for one of them, in 2018, half of the applicants to the AD5 competitions were older than 35. This was also visible in our survey of managers about their recent recruitment experience, which indicated that 22 % of recently recruited administrator officials were younger than 35. 62 % were over 40, which makes it likely that they already had at least 10 years of professional experience when they joined the institutions, more than what is usually required for the grades of the competitions.

36

The recruitment of experienced people to junior positions can be detrimental to the age composition of the workforce and to career management. Some institutions have started their own “junior professionals” programmes to address the lack of young graduates on reserve lists.

37

Most of EPSO’s communication work is thus centred on measures likely to attract candidates for the annual cycles of competitions, and particularly the generalist administrators cycle. This communication work follows the organisation of competitions in annual cycles (see paragraph 23), and targets the same people, in particular for the generalist administrators cycle, i.e. university graduates with limited or no professional experience.

38

In our view, such an approach is coherent, but it is limited in its geographical and socio-economic reach. Events are organised in places where people are already likely to be aware of the opportunities for employment in the EU institutions, and the graduate profiles targeted are limited (mostly to social sciences and economics graduates). Furthermore, it does not succeed in attracting a sufficient number of graduates or young professionals to an EU career.

Despite weaknesses in the selection process, the overall quality of the successful candidates is high

39

The EPSO selection process (see Annex III for details) is very competitive. Some competitions in our sample use computer-based pre-selection tests as a first step. In competitions where such tests are used, an average of 16.5 % of applicants give up before the start of the competition and do not sit those pre-selection tests. These computer-based psychometric tests are used because they are considered by many recruitment professionals to be one of the most accurate predictors of an employee’s future performance. However, in EPSO competitions, these tests are not only used to assess applicants’ suitability for the job, but also, in effect, to eliminate applicants to make the competition process more manageable.

40

The average success rate (laureates as a percentage of the number of candidates) for those who take at least the first test is 2 %. Within our sample, this figure varies between 27 % for a competition for lawyer-linguists with 26 candidates for 7 laureates, and 0.5 % for a translator competition (184 candidates for 1 laureate). Figure 6 shows how the number of candidates is reduced at each stage of the process for a generalist competition where computer-based tests were used.

41

This process is successful in identifying laureates whose skills and competencies are recognised by managers. In the survey of managers we conducted, 88 % declared they were satisfied or very satisfied with the quality of the EPSO laureates they had recruited.

42

According to the managers in our survey, candidates recruited through EPSO were not of a noticeably higher calibre than staff they had recruited through other channels (contract and temporary staff). They told us that the EPSO laureates they had recruited did not exhibit noticeably greater levels of the core skills and abilities necessary for a career in the institutions (see Figure 7), except, to a limited extent, for leadership skills.

43

EPSO does not have a formal mechanism in place to measure the institutions’ satisfaction with the services it offers. Discussions on the suitability of laureates take place informally during working group or management board meetings. Interinstitutional reflection groups can be put in place to deal with acute problems, such as competitions being cancelled in court, or dissatisfaction with some aspects of the selection. However, EPSO has put in place no procedure to address identified problems before they become serious enough to justify the creation of such a reflection group.

44

The language regime of the competitions has been impacting their effectiveness since the 2010s. Court cases have resulted in the cancellation of competitions and the suspension of EPSO’s activities for a large part of 2016. Before 2016, EPSO ran generalist competitions in three languages: English, French and German. Following legal challenges from Member States in the early 2010s, the Court of Justice ruled that this practice was discriminatory and unjustified.

45

In September 2016, following the work of an interinstitutional reflection group, EPSO changed to a two-step process for deciding the languages used in generalist competitions. Applicants are first invited to declare all EU languages they have knowledge of, along with their level of competence in these languages. Once they have all done so, the five most frequently declared languages are used for the competition. However, some parts of the application form, such as the talent screener tab, can only be filled in English, French and German. Complaints can also only be submitted in a set number of languages, which does not necessarily include every candidate’s mother tongue. The current regime has not been overturned in court, but these limitations leave it open to further legal challenges. This brings a continued uncertainty about EPSO’s activities. The EPSO management board is aware of the situation, but by the end of 2019 it had not agreed on a new language regime that would provide legal certainty to the organisation of competitions.

46

The selection process also relies heavily on the staff members nominated by the EU institutions to sit on selection boards. Their role is to check the admissibility of candidates and to evaluate them throughout the selection process. EPSO is responsible for their training and offers administrative support during the competition, but it is up to the institutions to assess the board members’ motivation and commitment to the task, and to ensure that they are given enough time to carry out their duties.

47

The feedback we gathered during the audit – from EPSO, from chairs of selection boards, and from the EU institutions – suggests that it is a challenge for all parties to find competent staff to sit on selection boards, and to ensure that the running of a competition does not conflict with the work programme of the institutions. The majority of these organisational issues are the result of insufficient cooperation between EPSO and the institutions.

48

Box 1 illustrates the consequences of the division of responsibilities between EPSO and the institutions on the working of the selection board.

49

In our view, the selection process run by EPSO is broadly effective in identifying good-quality laureates. It is highly selective, and it is particularly appropriate for generalist profiles with a high number of candidates. The process is nevertheless beset by weaknesses. EPSO does not measure the institutions’ satisfaction with the services it provides; no satisfactory long-term solution has been found to the uncertainties of the language regime; and competitions are dependent on high-quality selection board members being available, which is not always the case due to the division of responsibilities between EPSO and the institutions.

In years yielding the highest number of successful candidates, the cost of competitions is low

50

EPSO has fixed costs (staff and buildings expenditure) of around €15 million per year, no matter how many competitions it is organising. As a result, the cost of EPSO competitions is dependent on its volume of activity: the more laureates are placed on a reserve list in a given year, the lower the proportion of fixed costs in the cost per laureate.

51

As EPSO does not report on the cost of competitions, we conducted our own review. We analysed the overall cost per laureate for 15 competitions (or groups of competitions with the same selection board) which took place between 2012 and 2018. We took into account all the related costs, whether these were paid by EPSO (salaries of EPSO staff, payments to providers, etc.) or by the EU institutions (salaries of their staff members involved in the selection process). We allocated the costs to three categories: overheads (such as buildings and utilities); indirect costs (linked to EPSO’s competition activity, but not to a specific competition, such as the salaries of EPSO staff); and direct costs (linked to a specific competition, such as meetings and assessment centres).

52

The results are presented in Table 1 below. On average, across all profiles, the cost per laureate is around €24 000. Broken down by function group, it is €25 000 for an AD laureate, €21 900 for an AST laureate and €15 300 for an AST-SC laureate (there is only one such competition in our sample).

Table 1 – Cost per laureate of an EPSO competition

Reference	Profile	Tested candidates	Laureates	Cost per tested candidate	Cost per laureate
EPSO/AST-SC/06/17	Secretaries/Clerks AST/SC1 and SC2	4 121	359	€1 334	€15 318
EPSO/AD/287/14	Translators AD5	237	34	€2 597	€18 105
EPSO/AD/289/14	Lawyer-linguists AD7	364	14	€782	€20 324
EPSO/AST/130/14Field 4: Project management	Assistants in the buildings sector AST3	1 054	18	€351	€20 567
EPSO/AD/318/15(Options 1 and 2)	Translators AD5	762	25	€694	€21 158
EPSO/AD/249/13Field 1: Macroeconomics	Administrators AD7	1 068	41	€835	€21 748
EPSO/AD/347/17	Experts in media and digital communication AD6	1 860	63	€750	€22 138
EPSO/AST/144/17 (6 languages)	Linguistic assistants AST1	2 010	35	€403	€23 162
EPSO/AD/233/12	Translators AD5	1 751	70	€940	€23 525
EPSO/AD/331/16 (all fields)	ICT experts AD7	1 629	191	€2 808	€23 947
EPSO/AD/301/15	Administrators AD5	20 985	159	€188	€24 802
EPSO/AD/354/17-LV	Lawyer-linguists AD7	26	7	€6 900	€25 630
EPSO/AD/236 to 239/12	Conference interpreters AD5 and AD7	275	30	€3 064	€28 089
EPSO/AD/332/16 EPSO/AD/334/16 EPSO/AD/336/16	Lawyer-linguists AD7	1 170	26	€663	€29 814
EPSO/AD/256 to 259/13	Conference interpreters AD5 and AD7	501	13	€1 036	€39 936

Source: ECA based on EPSO data.

53

EPSO was set up at a time when the institutions needed to recruit a large number of people every year due to the 2004-2007 enlargements. Its cost structure reflects that situation. The competitions with the lowest cost per laureate in our sample are those for which the final number of laureates is quite high (EPSO/AST-SC/06/17) or which took place in years of high activity such as 2014 (1 200 laureates identified that year) and 2015 (1 636 laureates).

54

Another important factor in the cost of competitions is the ratio between tested candidates and laureates. Pre-selection tests have a cost of between €48 and €92 per tested candidate depending on the competition; when used, they can therefore increase the final cost per laureate substantially. Box 2 presents examples of this impact.

55

In our view, the cost structure is designed to minimise costs for large-scale competitions. It reflects EPSO’s initial set-up when it was expected to manage competitions with a large number of applicants and to identify a high number of laureates every year.

Regular generalist competitions helped to ensure an adequate supply of potential recruits

56

Along with the annual cycle, the EDP introduced targets of 9 to 10 months for the duration of competitions. Their duration improved markedly when the EDP came into force: before 2010, the average length of a competition was 18 months, and this came down to 13 months for the 2012-2018 period. Figure 8 shows the current timeline of the selection and recruitment process, with the fastest (14 months) and slowest (more than 4 years) durations between planning and recruitment.

57

The overall average duration of an AD5 (entry grade) generalist competition, at 13 months, is a little longer than the procedures run by other international organisations to select recruits for their young professionals programmes (between 10 and 12 months). The duration of competitions is reported on in EPSO’s annual activity reports, but no actions are presented to remedy the fact that competitions consistently take longer than the EDP targets to be completed.

58

The annual cycles of competitions are designed to ensure that the supply of new recruits is constant, since these are profiles for which recruitment takes place all year round. However, timeliness (publishing the list “at the right time”) was not factored in when designing the annual cycles in the EDP, except to the extent that reserve lists were not planned to be published in July/August when activity is low. The expected date of publication of reserve lists (May/June for administrators, September for assistants and December for linguists) does not correspond to any particular recruitment pattern in the institutions. Specialist and ad-hoc competitions are organised when time slots become available (see paragraph 23) and the publication of their reserve lists is on an “as soon as possible” basis rather than being set for a precise date.

59

In our view, holding regular large competitions has generally helped to ensure an adequate supply of potential generalist recruits. This system, however, takes no account of the timing of the institutions’ recruitment needs. EPSO competitions are still longer than the targets set in the EDP.

EPSO’s selection process for specialist profiles is not suited to the current recruitment needs of the EU institutions

60

The recruitment context in the institutions has changed since the EDP was drawn up and implemented. There have been staff cuts, and some strategic decisions to replace officials with temporary staff for some jobs. Since 2012, most of the competitions organised by EPSO have had reserve lists with fewer than 20 laureates (see Table 2).

Table 2 – EPSO competitions by size of reserve list, 2012-2018

Number of laureates on the reserve list	2012	2013	2014	2015	2016	2017	2018	2012-2018 average
1 to 10	52 %	36 %	15 %	61 %	33 %	59 %	10 %	44 %
11 to 20	14 %	18 %	23 %	18 %	41 %	11 %	30 %	21 %
21 to 40	14 %	23 %	27 %	8 %	14 %	14 %	20 %	16 %
41 to 60	0 %	8 %	12 %	2 %	7 %	0 %	20 %	5 %
61 to 80	10 %	8 %	4 %	2 %	5 %	7 %	10 %	6 %
81 to 100	0 %	3 %	0 %	0 %	0 %	2 %	0 %	1 %
101 to 150	7 %	5 %	12 %	4 %	0 %	5 %	0 %	5 %
151 to 200	0 %	0 %	4 %	2 %	0 %	0 %	10 %	1 %
More than 200	3 %	0 %	4 %	2 %	0 %	2 %	0 %	2 %

Source: ECA based on EPSO data.

61

The rapidly changing environment in which the EU and its institutions are now operating (affected by staff cuts, requiring new skills to implement digitalisation, etc.) is not the one EPSO was designed for. Since 2008, there has been an increased demand to meet urgent or crisis-driven recruitment needs which call for more targeted profiles. To carry out its mission in this new environment, in parallel to its work on large-scale competitions we would have expected EPSO to:

1. adequately gather the institutions’ estimates of staff needs for specialist, targeted profiles to prepare a reliable timetable for small-scale competitions;
2. target its communications to reach hard-to-recruit specialists;
3. use testing methods appropriate for specialists;
4. ensure that the cost of these competitions is kept low;
5. ensure that reserve lists are published in time to meet the institutions’ recruitment needs.

The planning of specialist competitions is not reliable and takes place too early to be precise

62

Since 2012, the institutions’ needs for specialist profiles has risen (see Figure 4 and paragraph 26). This increased need for specialists is also visible in our survey of managers: 57 % of the managers who took part agreed with the statement “reserve lists for specialists are exhausted too quickly” (only 35 % agreed with the same statement for generalists). 76 % of the managers who gave an answer about whether they preferred recruits from specialist or generalist reserve lists chose specialist ones.

63

Information about recruitment needs for specialist profiles is based on each institution telling EPSO during the planning exercise what profiles it is looking for. The competitions for specialists are accommodated as much as possible in the slots left free by the organisation of the annual cycles.

64

Specialist competitions are small-scale competitions, usually with small reserve lists (fewer than 30 laureates). As a result, they are not always prioritised (see paragraph 25) and are sometimes postponed or cancelled. This in turn leads the institutions to find other, more flexible ways of filling their recruitment needs, such as organising their own selection procedures for temporary staff.

65

When comparing requests for specialists, as stated in the institutions’ initial planning estimates, with the actual numbers of laureates sought in the notices of competition, we found that the needs for specialist laureates are over-estimated at the planning stage (see paragraph 27) compared with the number of laureates sought when drawing up the notice of competition.

66

The duration of the planning exercise contributes to this over-estimation, since institutions’ priorities change between the moment EPSO requests estimates of recruitment needs (May), the time the management board approves the timetable (November), and the moment competition starts (some time in the following year). As a consequence, the timetable of competitions is constantly being revised and adjusted, as is the number of laureates requested.

67

In our view, the planning process is not adapted to the current needs of the institutions. It gives too much priority to large-scale generalist competitions at a time when the institutions are requesting smaller competitions. In addition, the timetable for the planning process, which starts up to a year before the competition takes place, does not allow specialist needs to be estimated precisely, leading to over-estimations and constant adjustment. Its reliability and practicality for specialist competitions is thus limited.

EPSO does not consistently target its communication actions to attract good candidates to specialist competitions

68

EPSO can only provide the institutions with suitable potential recruits if it manages to attract appropriate candidates to apply to competitions. We found that applicants do not always match the profiles sought. Some candidates do not genuinely wish to be employed by the institution seeking recruits, but rather by another EU institution. The latter problem is particularly problematic for some specialist profiles (e.g. IT) in locations outside Brussels. The only tool currently used to help candidate assess their own suitability for the competition is a non-mandatory short multiple-choice test that they can take before applying. It is designed to help potential candidates assess whether they are suited to a job in the EU institutions, but not to a particular competition.

69

A further problem arises when competitions have too few applicants. The EDP sets a target ratio of applicants to laureates (38 candidates for 1 laureate for administrators, 27 for 1 for assistants, and 40 for 1 for linguists). We used these ratios to assess the effectiveness of the communication actions in attracting candidates. In the competitions in our sample, this target was not always reached, no matter how much was spent on communication expenditure for the competition. Other competitions with little or no communication expenditure widely exceeded their targets (see Table 3). There is no clear link between communication expenditure and the number of applicants.

Table 3 – Communication expenses and achievement of the target number of applicants

Competition	Profile	Achievement of the EDP target	Expenses per candidate	Expenses per laureate
AD/354/17	Specialist Linguist (lawyer-linguist LV)	9 %	€270	€1 003
AD/287/14	Linguist (translators SV)	23 %	€7	€67
AD/331/16	Specialist Administrator (ICT)	23 %	€4	€37
AD/249/13	Specialist Administrator (macroeconomics)	69 %	€13	€330
AD/233/12	Linguist (translators HR)	78 %	€3	€98
AD/318/15	Linguist (translators PL)	95 %	€14	€521
AD/256-259/13	Specialist Linguist (conference interpreters)	96 %	€1	€39
AD/347/17	Specialist Administrator (communication)	98 %	€4	€154
AD/289/14	Specialist Linguist (lawyer-linguist FR)	99 %	€3	€112
AD/332-334-336	Specialist Linguist (lawyer-linguist ES, MT and IT)	161 %	€4	€271
AST/144/17	Specialist Linguist (AST)	200 %	€2	€145
AST/130/14	Specialist Assistant (project managers)	220 %	€0*	€0

* no competition-specific communication expenses incurred.

Source: ECA based on EPSO data.

70

In our view, EPSO has not adequately publicised its competitions targeting specialist staff, despite the growing need for such competitions. It is both ineffective and costly when applicants do not match the profiles sought or refuse the job offers from the institutions seeking recruits. This issue is more problematic when reserve lists are small.

Candidates with strong specialist profiles risk being eliminated early in the selection process

71

The EDP sets out a standard format for competitions (see Annex III). This set-up means that specialist job-related knowledge is tested at the intermediate test (this is often the case for linguist profiles), or even as late as the assessment phase, through a job-specific interview. If the selection criteria (field of study or work experience) are not defined adequately, then suitable candidates might be eliminated at the pre-selection stage.

72

For very targeted profiles, it is more attractive for the institutions to organise their own selection procedures, because doing so gives them full flexibility to use the sort of tests they consider necessary. For 19 out of the 22 of such procedures that we examined, the institutions started with a CV sift to allow only suitable candidates to proceed to the testing stage. This is a good way of eliminating unsuitable candidates upfront.

73

The assessment centre was introduced in the EDP as a way to shift the focus of competitions from knowledge to skills. Its purpose is to assess candidates against a set of competencies and skills which the institutions consider important in their own staff.

74

However, as illustrated in Figure 9, our survey shows that when managers are asked to agree on whether staff in general possess these skills, they do not see much difference between the skills and abilities of laureates of EPSO competitions and those of temporary staff recruited under the institutions’ own selection procedures (see also paragraph 42 on managers’ assessment of their own latest recruits). In those procedures, the institutions tend to focus more on technical tests (e.g. drafting and translation) and skip the “skills” part corresponding to the EPSO assessment centre.

75

Except for language skills and the ability to work in a multicultural environment, managers do not consider that new staff members recruited through EPSO are noticeably more skilled than members of staff recruited through other channels. They consider, in general, the relevant professional experience of staff recruited through other selection procedures to be slightly better, which suggests that these procedures are more suited to the recruitment of specialist, targeted profiles.

76

In our view, the format of competitions and the testing methods set out in the EDP are not fit for competitions for specialists and other small-scale competitions.

• Checking the experience of candidates and testing their job-related knowledge after the pre-selection phase increases the risks both of eliminating good candidates too early in the process and of being left with too few competent candidates at the assessment centre stage.
• The assessment centre does not bring any noticeable advantages, and recruits from the institutions’ own selection procedures are considered by managers to have more relevant professional experience, which is what is needed for specialist profiles.

When they target only a few specialists, EPSO competitions are more expensive than the institutions’ own selection procedures

77

Direct costs (see paragraph 51) account for between 20 % and 65 % of the overall cost of the competitions in our sample. When analysing the direct costs of each phase of competition, the most expensive phase is usually the assessment centre, as shown in Figure 10. The longer the assessment centre phase, or the larger the selection board, the more expensive the competition is.

78

In our sample, the average cost of the assessment centre phase was €4 038 per laureate, but there were large variations. The assessment centre phase for competition AST-SC/06/17 was rather fast and the number of laureates was rather large, so the cost per laureate was only €843. For the other assistant-level competitions, it was €2 661. For administrators, it varied between €2 587 per laureate for the AD5 generalist competition (AD/301/15) to more than €10 000 per laureate for conference interpreters (average of two competitions).

79

The assessment centre is expensive for specialist profiles for different reasons: specialist profiles require more experienced selection board members with higher grades; it can also take longer, especially if different specialisms have been grouped together to reach a critical mass for the competition.

80

Expected success rate at the assessment centre is around 30 %. For some profiles, however, the pass rate at the assessment centre is quite high. For one competition in our sample (AD/354/17), it was as high as 88 %. This can happen when there are too few applicants to start with, or when a large proportion of the candidates are eliminated in the intermediate/competency tests before the assessment centre. In such cases, the assessment centre, is not a cost-effective way to test candidates.

81

We also compared the direct costs per laureate of EPSO competitions with the direct costs of the institutions’ own selection procedures (see details in Annex v) and calculated an average depending on the size of the reserve lists. The results are presented in Table 4.

Table 4 – Direct cost per laureate, by size of reserve list, EPSO and EU institutions

Number of laureates on the reserve list	1-20	21-40	41-60	61-80	>100
EPSO	€14 165	€9 600	€7 866	€8 177	€6 828
Institutions	€7 444	€4 882	N/A	N/A	N/A

Source: ECA based on data from EPSO, the European Parliament, the Commission and the Court of Justice.

82

Based on the competitions in our sample, EPSO is a cheaper option for large competitions, but for smaller ones, the institutions have much lower direct costs, up to 50 % less. This can be explained by the smaller number of candidates files to handle (on average, in our samples, for competitions with fewer than 20 laureates, EPSO had 533 candidates, the institutions 66) which helps bring the costs down. Institution-run small-scale competitions also tend to have fewer testing stages.

83

In our view, EPSO’s current competition format is not effective in keeping costs low for the small-scale specialist competitions that the institutions are now requesting. The institutions’ own procedures have lower costs for this type of competition, often because they have fewer stages, target a smaller population of candidates, and are more tailored to the requirements of the institution concerned. The assessment centre in particular increases costs with no clear benefit, especially since the pass rate is very high, for small competitions.

EPSO procedures are less suitable to fill urgent recruitment needs for specialists than the institutions’ own procedures

84

Each of the stages of a competition (see paragraph 11) takes time to be carried out, with additional “waiting time” for candidates between stages to allow for the harmonisation of marks, or to accommodate requests for review or complaints.

85

In our sample, five competitions included all possible stages. Figure 11 shows the duration of each of these stages. It is worth noting that the shortest competition in this series of five, AD/301/15, is the generalist competition, with the largest reserve list (159 laureates). The others all had reserve lists with fewer than 35 laureates.

86

As Figure 8 shows, a considerable amount of time elapses between a need being expressed by an institution and the associated competition being launched. Furthermore, even if the planning guidelines mention flexibility for “ad-hoc needs”, it is not possible to launch a competition within a couple of months. The selection process is thus not suited to the task of meeting urgent recruitment needs.

87

As a consequence, and also to avoid the interinstitutional negotiation phase of the planning process, institutions sometimes decide to launch their own recruitment procedures, opting in that case to recruit temporary staff members rather than officials. We examined 22 such procedures (see Annex VI for details) to assess their duration.

88

Institutions manage to complete procedures with four stages of selection (CV sift, interviews, written test and sometimes group discussions) within eight months of the publication of the call for expression of interest (compared to an average of 13 months for an EPSO competition). This makes these procedures more flexible and thus more suited to the task of filling urgent recruitment needs than specialist competitions. Institutions also have more flexibility in the timing of their own procedures, compared to the organisation of specialist competitions which must be negotiated between institutions, and which are highly dependent on the availability of selection board members from different institutions (see paragraphs 25 and 47).

89

In our view, the set number of stages in EPSO’s competitions makes the entire process long, not flexible enough, and encourages the institutions to organise their own procedures when they are looking to fill a specialist role quickly. This format is not appropriate for meeting urgent recruitment needs.

Conclusions and recommendations

90

Overall, EPSO is successful at doing what it was designed to do: produce a large number of laureates whose skills are broad enough for a long and varied career in the institutions. We found that the current organisation of EPSO’s selection process is driven by the ideas put forward in the EDP back in 2008: organising competitions in annual cycles for profiles for which recruitment needs are expected to be recurrent.

91

The planning process is centred around large competitions organised every year (paragraphs 21 to 28). Communications work carried out under the EU Careers brand, mainly in Brussels, has managed to attract a high number of candidates (paragraphs 29 to 38). Testing methods have ensured that the quality of the successful candidates is good, generally meeting managers’ expectations (paragraphs 39 to 42).

92

There are, however, a number of weaknesses in the selection process that impact its effectiveness. Psychometric tests are not only used to assess applicants’ suitability for the job, but also, in effect, to eliminate applicants to make the competition process more manageable (paragraphs 39 and 40). There is no formal mechanism for measuring the institutions’ satisfaction with the laureates provided (paragraph 43). The restriction in the number of languages that can be used for some parts of the selection procedure has led to continuing legal uncertainty (paragraphs 44 and 45). Insufficient coordination between EPSO and the institutions can delay competitions (paragraphs 46 and 47). Furthermore, EPSO does not monitor the cost of competitions and has not taken action to ensure that targets for the duration of competitions are met (paragraphs 51 and 57).

Recommendation 1 – Address identified weaknesses in the selection process

The management board should update the selection process to address the following identified weaknesses that impact its effectiveness:

• measure the institutions’ satisfaction with the selection process;
• review the language regime, in particular for the application forms and the complaints system, to ensure legal certainty for the candidates and the institutions;
• ensure a better coordination between EPSO and the institutions on the availability of selection board members;
• introduce monitoring arrangements to report on the cost of competitions;
• reduce the duration of competitions to the target of 10 months set out in the EDP.

Timeframe: end of 2021

93

We also found that EPSO’s organisational set-up keeps the cost of competitions low when the number of successful candidates per year is high (paragraphs 50 to 55). In addition, holding generalist competitions regularly has helped to ensure an adequate supply of potential recruits for the institutions (paragraphs 56 to 59).

94

However, we found that the current selection process is not well suited to the needs of the institutions in an environment where staff numbers have been reduced. Instead of large-scale recruitment exercises aimed at generalist profiles, the institutions are increasingly looking for more specialised staff who can become operational quickly. This implies a need to organise competitions with small reserve lists (fewer than 30 laureates).

95

The current selection process is not well adapted to this type of competition. The planning procedure is not reliable and takes place too early to be precise (paragraphs 62 to 67). EPSO does not consistently target its communications to attract suitable candidates to specialist competitions (paragraphs 68 to 70). Job-specific knowledge is tested late, with the risk that candidates with strong specialist profiles may be eliminated early in the selection process, resulting in too few suitable candidates reaching the assessment centre (paragraphs 71 to 76).

96

The institutions also incur lower costs when they organise their own selection procedures for specialist profiles, and put in place a more flexible selection process for these procedures (see paragraphs 72, and 77 to 83). Specialist competitions organised by EPSO are too slow to meet institutions’ urgent recruitment needs (paragraphs 84 to 89), compared to what the institutions can manage with their own, lighter procedures.

Recommendation 2 – Introduce a new selection framework for specialist competitions

The management board should introduce a new selection framework for specialists with separate planning arrangements.

These specialist competitions, and competitions with small reserve lists, should be carried out based on a precise agreement between the requesting institutions and EPSO, specifying at least:

• how job-related knowledge testing is to be carried out, and its relative weight compared to the skills tested at the assessment centre;
• the resources and staff that the institutions will put at EPSO’s disposal (including provisions specifying the qualifications and experience of the selection board members);
• the communication work planned to publicise the competition;
• the timeframe EPSO commits to respect for the competition;
• the estimated cost of the competition, the related monitoring and reporting arrangements, as well as the specific measures envisaged to minimise the costs.

Timeframe: first quarter of 2023

97

EPSO has not adapted its selection process to the changing environment in which the institutions have been operating since 2012. The EDP was a response to criticisms from the 2004-2008 period; it did not introduce sufficient flexibility to equip EPSO with the means to adapt its selection process to the requirements of the rapidly changing environment in which the EU and its institutions now have to operate (e.g. staff cuts, new skills linked to digitalisation) (paragraphs 62 to 67, 71 to 76, 84 to 89).

Recommendation 3 – Improve EPSO’s capacity to adapt to a fast-changing recruitment environment

The management board should introduce a regular review of the selection process to ensure its capacity to react to a fast-changing recruitment environment.

Timeframe: end of 2021

This Report was adopted by Chamber V, headed by Mr Tony Murphy, Member of the Court of Auditors, in Luxembourg on 29 September 2020.

Annexes

Annex I – Categories of staff employed in the European institutions

The institutions’ workforce is made up of different categories of staff. Within each category, there are different grades, reflecting increasing professional and educational requirements and levels of responsibility.

An official is any person who has been appointed, after passing a competition, to a permanent post at one of the institutions. Commissioners are not officials.

Officials belong to one of three function groups:

• Administrators (AD), in grades AD5 to AD15;
• Assistants (AST), in grades AST1 to AST11;
• Secretaries and clerks (AST/SC), in grades AST/SC1 to AST/SC6.

Temporary staff are engaged:

• on short-term contracts (maximum 6 years) to fill a permanent post at one of the institutions or in the European External Action Service;
• on a short-term contract or, more rarely, for an indefinite period to fill a temporary post at an institution or an agency;
• to assist a person holding an office (e.g. a Commissioner), in which case the length of their contract is linked to that of the office holder’s term of office.

The function groups for temporary staff are the same as for officials.

Contract staff are not assigned to an established post. They are divided into four function groups (GFI to GFIV), depending on the tasks they carry out: from GFI for manual tasks to GFIV for administrative tasks. Staff in GFI or those working at an agency or in a delegation, representation or office may be engaged for an indefinite period; the contracts of other staff cannot exceed 6 years.

Local staff are engaged in places outside the European Union according to local rules and practices. They are not assigned to established posts.

Annex II – Entry requirements for each staff category

01

The table below summarises the minimum qualifications and language skills required for officials and temporary staff, along with the sort of tasks carried out by each function group.

Officials and temporary staff

Qualifications and professional experience

AD

Completed university studies of at least 3 years (four for AD7) attested by a diploma. AD7 competitions usually require a minimum of 6 years of experience after the qualifying diploma. AD5 competitions do not require any experience and are more adapted to young graduates.

AST and AST/SC

Post-secondary education attested by a diploma; or secondary education attested by a diploma, and professional experience of at least 3 years.Competitions for AST/SC1 do not require professional experience, but those for AST3 or SC2 usually require at least 3 or 4 years’ additional professional experience directly related to the nature of the duties, gained after obtaining the qualification or professional experience required for access to the competition

Language requirement

For all function groups: thorough knowledge of one of the EU languages and a satisfactory knowledge of another.

Indicative tasks

AD

Administrators: generalist profiles such as lawyers, auditors, economists and translators, and more specialist profiles (lawyer-linguists, researchers, IT and communication experts).

AST

Generalist roles in administration, policy development and implementation, or more technical ones in finance, communication, research.

AST/SC

Clerical and secretarial tasks, office management

 

Annex III – Stages of an EPSO open competition

Stage	Description
Self-assessment	Before applying, potential candidates are invited to respond to a non-mandatory questionnaire to learn more about what to expect from a career within the EU institutions, and the type and difficulty of tests they will face.
Online application	Candidates apply through their EPSO accounts (an electronic profile for managing application details). A separate application form has to be completed for each competition. Candidates also receive communications on the organisation of the competition from EPSO through their EPSO accounts.
Pre-selection tests	Pre-selection tests are a part of some competitions. Their format is set out in the notice of competition, but they mostly take the form of computer-based multiple-choice tests. Candidates sit the tests in test centres in the EU Member States and sometimes worldwide.
Admission phase/Talent screener and selection based on qualification	The selection board checks that the candidates that had the best scores in the pre-selection tests also fulfil the eligibility criteria set out in the notice of competition, and invites the best-performing applicants to the next phase. The number of candidates to be invited is indicated in the notice of competition. For specialists competitions, selection is usually based on qualifications. The selection board assesses the applications and selects the candidates whose qualifications best satisfy the criteria set out in the notice of competition. This selection is made solely on the basis of responses to specific questions in the ‘talent screener’ tab of the online application form in EPSO's IT tool.
Preliminary/intermediate tests	In some competitions, there is an intermediary phase where candidates have to sit further tests. These tests are usually delivered in test centres in the EU Member States, and can for example include an e-tray exercise or a translation test.
Assessment centre	If candidates’ scores in the previous stages are among the best and they meet the eligibility criteria set out in the notice of competition, they are invited to the assessment centre. Assessment centres are designed to evaluate the pre-defined competencies by observing participants’ behaviour. They consist of a number of different simulation exercises in a job-relevant context where behaviour is compared to the competency profile. Candidates may be tested on two types of competencies: field-specific competencies and general competencies. Field-specific competencies are the applied knowledge and skills needed to meet the immediate requirements of the specific job profile for that particular competition (e.g. competition lawyer, assistant auditor). General competencies are abilities which all officials need in order to have a successful career within the European institutions. Assessment centres are normally held in Brussels or Luxembourg (except the case study, which is held separately in the Member States). Typical assessment centre exercises are: —    case study; —    group exercise; —    oral presentation; —    structured interview; —    job-specific interview; —    e-tray exercise.Successful candidates’ diplomas and evidence of their professional experience are checked after the assessment centre, and their names are placed on the reserve list of laureates.
Reserve list	The reserve list contains the names of those candidates who performed best in the competition. The number of places available on the reserve list is stated in the notice of competition. The list is sent to the EU institutions, which can then recruit successful candidates from the list according to their needs.

Annex IV – Audit methodology

01

We gathered our evidence for this audit from the following sources:

• a review of a sample of competitions organised by EPSO, for which we analysed their costs and duration;
• desk reviews of EPSO’s management information such as: annual activity reports, management plans, minutes of the management board (2016-2018), minutes of the EPSO working group (2016-2018);
• audit questionnaires to EPSO, the European Parliament, the European Commission and the Court of Justice;
• audit interviews with EPSO, the European Parliament, the European Commission, the Court of Justice, and EPSO’s service provider for the assessment centre phase;
• survey of 471 middle managers in the European Parliament, the European Commission and the Court of Justice;
• analysis of EPSO’s Twitter communications in 2018-2019;
• questionnaires to and interviews with chairs and alternate chairs of EPSO competitions launched in 2015-2016.

Table 5 – Sample of competitions examined

Reference	Year	Profile	Candidates	Tested candidates	Laureates
EPSO/AST-SC/06/17	2017	Secretaries/Clerks AST/SC1 and SC2	5 573	4 121	359
EPSO/AD/287/14	2014	Translators AD5	306	237	34
EPSO/AD/289/14	2014	Lawyer-linguists AD7	552	364	14
EPSO/AST/130/14Field 4: Project management	2014	Assistants in the buildings sector AST3		1 054	18
EPSO/AD/318/15Options 1 and 2	2015	Translators AD5	948	762	25
EPSO/AD/249/13Field 1: Macroeconomics	2013	Administrators AD7		1 068	41
EPSO/AD/347/17	2017	Experts in media and digital communication AD6	2 327	1 860	63
EPSO/AST/144/17 (6 languages)	2017	Linguistic assistants AST1	2 803	2 010	35
EPSO/AD/233/12	2012	Translators AD5	2 178	1 751	70
EPSO/AD/331/16(all fields)	2016	ICT experts AD7		1 629	191
EPSO/AD/301/15	2015	Administrators AD5	31 400	20 985	159
EPSO/AD/354/17-LV	2017	Lawyer-linguists AD7		26	7
EPSO/AD/236/12 EPSO/AD/237/12 EPSO/AD/238/12 EPSO/AD/239/12	2012	Conference interpreters AD5 and AD7	333	275	30
EPSO/AD/332/16 EPSO/AD/334/16 EPSO/AD/336/16	2016	Lawyer-linguists AD7	1 677	1 170	26
EPSO/AD/256/13 EPSO/AD/257/13 EPSO/AD/258/13 EPSO/AD/259/13	2013	Conference interpreters AD5 and AD7		501	13

Source: ECA based on EPSO data.

02

To estimate the costs of EPSO’s competitions and selection procedures, we established a calculation methodology in cooperation with EPSO. The main points of the methodology were:

• split the costs in three categories: overheads (such as buildings and utilities); indirect costs (linked to EPSO’s competition and selection activity, but not to a specific competition); and direct costs (linked to a specific competition, such as meetings, assessment centres);
• for the share of overheads relating to competitions, and the indirect costs, we divided them by the number of laureates in that year and then added them (pro rata) to the direct costs of the individual competitions in our sample;
• the direct costs were either taken at face value (for invoices from providers or mission expenses) or calculated by us.

03

To compare part of EPSO’s direct costs with the cost of the institutions’ own selection procedures we proceeded in the following manner.

• We sent to each of the three institutions a questionnaire to identify the number and type (internal competitions or recruitment of temporary agents via calls for expression of interest) of internal selection procedures carried out between 2015 and 2018.
• From these procedures, we selected the most recent ones concerning the selection of temporary staff, for which the institutions had not requested any support from EPSO. We asked the institutions to provide us with the direct costs when they had them (invoices, travel expenses) or with evidence allowing us to calculate them (e.g. number of meetings necessary for the running of the selection, their duration and the grade of the attendees).
• Based on this, we calculated the cost of the institutions’ selection procedures.

Annex V – Direct costs per laureate, EPSO competitions and institutions’ own selection procedures

Reference	Year	Candidates	Laureates	Direct cost per laureate
Institution procedure 12	2017	8	1	€751
Institution procedure 7	2017	7	1	€991
Institution procedure 22	2018	21	2	€1 027
Institution procedure 14	2018	37	7	€1 166
Institution procedure 11	2017	35	10	€1 274
Institution procedure 13	2018	74	1	€1 784
Institution procedure 20	2018	23	1	€2 355
EPSO/AST-SC/06/17	2017	5 573	359	€2 753
Institution procedure 17	2018	8	1	€3 774
Institution procedure 6	2017	148	12	€3 999
Institution procedure 16	2018	228	32	€4 808
Institution procedure 10	2017	59	10	€4 831
Institution procedure 19	2018	227	23	€4 956
EPSO/AD/331/16 (all fields)	2016	1 629	191	€5 648
Institution procedure 15	2018	117	12	€5 698
EPSO/AD/287/14	2014	306	34	€6 467
EPSO/AD/233/12	2012	2 178	70	€6 820
Institution procedure 18	2018	223	14	€7 837
EPSO/AD/249/13 – Field 1: Macroeconomics	2013	1 068	41	€7 866
EPSO/AD/318/15 – Options 1 and 2	2015	948	25	€8 375
EPSO/AD/289/14	2014	552	14	€8 681
EPSO/AST/130/14 (AST3) - Field 4: Project management	2014	1 054	18	€8 949
Institution procedure 2	2015	156	6	€9 216
Institution procedure 1	2015	120	5	€9 463
EPSO/AD/347/17	2017	2 327	63	€9 534
EPSO/AST/144/17 (AST1) (6 languages)	2017	2 803	35	€10 516
Institution procedure 4	2016	37	3	€11 174
EPSO/AD/332/16 EPSO/AD/334/16 EPSO/AD/336/16	2016	1 677	26	€11 263
EPSO/AD/236/12 EPSO/AD/237/12 EPSO/AD/238/12 EPSO/AD/239/12	2012	333	30	€11 377
Institution procedure 3	2016	13	3	€11 990
EPSO/AD/301/15	2015	31 400	159	€12 082
Institution procedure 21	2018	79	5	€12 359
EPSO/AD/354/17-LV	2017	26	7	€12 979
Institution procedure 8	2017	n/a	3	€13 630
Institution procedure 9	2017	72	3	€13 734
EPSO/AD/256/13 EPSO/AD/257/13 EPSO/AD/258/13 EPSO/AD/259/13	2013	501	13	€26 052
Institution procedure 5	2016	15	1	€31 832

Source: ECA based on data from EPSO, the European Parliament, the Commission and the Court of Justice.

Annex VI – Duration of the institutions’ selection procedures

Institutions’ selection procedures for temporary staff

	Year	Stages	Stage 1	Stage 2	Stage 3	Stage 4	Stage 5	Candidates	Laureates	Selection notice	Reserve list	Duration (months)
1	2015	5	CV sift	Drafting test in language 1 (job-specific competency)	Drafting test in language 2 (case study)	Interview	Group discussion	120	5	30.1.2015	21.12.2015	10.8
2	2015	4	CV sift	Drafting test in language 2	Interview	Group discussion		156	6	8.9.2015	1.8.2016	10.9
3	2016	5	CV sift	Drafting test in language 1 (job-specific competency)	Drafting test in language 2 (case study)	Interview	Group discussion	13	3	21.4.2016	18.10.2016	6.0
4	2016	4	CV sift	Drafting test in language 2	Interview	Group discussion		37	3	24.5.2016	14.6.2017	12.9
5	2016	4	CV sift	Drafting test in language 2	Interview	Group discussion		15	1	26.8.2016	14.2.2017	5.7
6	2017	2	CV sift	Interview				148	12	1.2.2017	16.6.2017	4.5
7	2017	3	CV sift	Interview	Real-life testing of skills			7	1	13.3.2017	13.7.2017	4.1
8	2017	4	CV sift	Drafting test in language 2	Interview	Group discussion		12	3	22.6.2017	6.3.2018	8.6
9	2017	4	CV sift	Drafting test in language 2	Interview	Group discussion		72	3	28.7.2017	26.4.2018	9.1
10	2017	2	CV sift	Interview				59	10	11.9.2017	31.8.2018	11.8
11	2017	2	Translation test	Interview				35	10	22.9.2017	11.6.2018	8.7
12	2017	3	CV sift	Interview	Real-life testing of skills			8	1	24.5.2017	22.9.2017	3.9
13	2018	2	CV sift	Interview				74	1	17.1.2018	8.6.2018	4.7
14	2018	2	Translation test	Interview				37	7	3.4.2018	30.5.2018	1.9
15	2018	3	CV sift	Physical tests (job-specific)	Interview			117	12	25.4.2018	9.1.2019	8.6
16	2018	3	CV sift	Interview	Drafting test			228	32	9.7.2018	25.6.2019	11.7
17	2018	2	Translation test	Interview				8	1	24.9.2018	26.10.2018	1.1
18	2018	3	CV sift	Interview	Drafting test			223	14	9.11.2018	23.7.2019	8.5
19	2018	3	CV sift	Interview	Drafting test			227	23	9.11.2018	1.8.2019	8.8
20	2018	2	CV sift	Interview				23	1	30.11.2018	28.2.2019	3.0
21	2018	2	CV sift	Interview				79	5	21.1.2019	4.7.2019	5.5
22	2018	3	CV sift	Interview	Real-life testing of skills			21	2	16.4.2018	2.7.2018	2.6

Source: ECA based on European Parliament, European Commission and Court of Justice data.

Acronyms and abbreviations

CBT: computer based test (usually psychometric)

EDP: EPSO Development Plan

EPSO: European Personnel Selection Office

Glossary

Laureate: Successful candidate of an EPSO open competition.

Office: A department of the European Commission, headed by a Director. Offices were created to perform specific tasks, often of an administrative nature (e.g. Paymaster’s Office, Offices for Buildings and Infrastructure in Brussels and Luxembourg). Unlike Directorates-General, Offices do not deal with policy development.

Reserve list: A list of all successful candidates (laureates) of an EPSO open competition. It is published in the Official Journal of the EU.

Selection board member: Official from an EU institution nominated to sit on the selection board assessing the candidates of an EPSO open competition. Selection board members are called “permanent” when they are seconded by their institution to EPSO for a longer period of time, instead of being nominated for one competition.

Audit team

The ECA’s special reports set out the results of its audits of EU policies and programmes, or of management-related topics from specific budgetary areas. The ECA selects and designs these audit tasks to be of maximum impact by considering the risks to performance or compliance, the level of income or spending involved, forthcoming developments and political and public interest.

This performance audit was carried out by Audit Chamber V Financing and administration of the EU, headed by ECA Member Tony Murphy. The audit was led by ECA Member Annemie Turtelboom, supported by Florence Fornaroli, Head of Private Office and Celil Ishik, Private Office Attaché; Bertrand Albugues, Principal Manager; Marion Kilhoffer, Head of Task; Daria Bochnar, Panagiota Liapi and Jesús Nieto Muñoz, Auditors. Richard Moore provided linguistic support.

Endnotes

¹ See Decision 2002/621/EC of 25 July 2002, OJ L 197 of 26.7.2002, and https://epso.europa.eu/.

² See http://europa.eu/epso/doc/edp_2012_final_version.pdf.

³ See EPSO’s mission statement: “EPSO's role is to serve the EU institutions by providing high quality, efficient and effective selection procedures that enable them to recruit the right person, for the right job, at the right time.”

⁴ EPSO Development Programme: roadmap for implementation, 10/09/2008. See also EPSO Development Programme, Final report, 2012 https://europa.eu/epso/doc/edp_2012_final_version.pdf.

⁵ EPSO candidates satisfaction surveys done after the assessment centre, 2014-2018.

Timeline

Event	Date
Adoption of Audit Planning Memorandum (APM) / Start of audit	3.9.2019
Official sending of draft report to the European Commission (or other auditee)	9.7.2020
Adoption of the final report after the adversarial procedure	29.9.2020
Official replies of the European Commission received in all languages	26.10.2020

Contact

EUROPEAN COURT OF AUDITORS
12, rue Alcide De Gasperi
1615 Luxembourg
LUXEMBOURG

Tel. +352 4398-1
Enquiries: eca.europa.eu/en/Pages/ContactForm.aspx
Website: eca.europa.eu
Twitter: @EUAuditors

More information on the European Union is available on the internet (http://europa.eu).

Luxembourg: Publications Office of the European Union, 2020

PDF	ISBN 978-92-847-5334-5	ISSN 1977-5679	doi:10.2865/63410	QJ-AB-20-021-EN-N
HTML	ISBN 978-92-847-5310-9	ISSN 1977-5679	doi:10.2865/645011	QJ-AB-20-021-EN-Q

COPYRIGHT

© European Union, 2020.

The reuse policy of the European Court of Auditors (ECA) is implemented by the Decision of the European Court of Auditors No 6-2019 on the open data policy and the reuse of documents.

Unless otherwise indicated (e.g. in individual copyright notices), the ECA’s content owned by the EU is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence. This means that reuse is allowed, provided appropriate credit is given and changes are indicated. The reuser must not distort the original meaning or message of the documents. The ECA shall not be liable for any consequences of reuse.

You are required to clear additional rights if a specific content depicts identifiable private individuals, f. ex. as on pictures of the ECA’s staff or includes third-party works. Where permission is obtained, such permission shall cancel the above-mentioned general permission and shall clearly indicate any restrictions on use.

To use or reproduce content that is not owned by the EU, you may need to seek permission directly from the copyright holders:

Figures 1, 6 and 8: Icons © 2010-2020 Freepik Company S.L. All rights reserved.

Software or documents covered by industrial property rights, such as patents, trade marks, registered designs, logos and names, are excluded from the ECA’s reuse policy and are not licensed to you.

The European Union’s family of institutional Web Sites, within the europa.eu domain, provides links to third-party sites. Since the ECA does not control them, you are encouraged to review their privacy and copyright policies.

Use of European Court of Auditors’ logo

The European Court of Auditors logo must not be used without the European Court of Auditors’ prior consent.

GETTING IN TOUCH WITH THE EU

In person
All over the European Union there are hundreds of Europe Direct Information Centres. You can find the address of the centre nearest you at: https://europa.eu/european-union/contact_en

On the phone or by e-mail
Europe Direct is a service that your questions about the European Union. You can contact this service

• by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),
• at the following standard number: +32 22999696 or
• by electronic mail via: https://europa.eu/european-union/contact_en

FINDING INFORMATION ABOUT THE EU

Online
Information about the European Union in all the official languages of the EU is available on the Europa website at: https://europa.eu/european-union/index_en

EU Publications
You can download or order free and priced EU publications at: https://op.europa.eu/en/web/general-publications/publications. Multiple copies of free publications may be obtained by contacting Europe Direct or your local information centre (see https://europa.eu/european-union/contact_en)

EU law and related documents
For access to legal information from the EU, including all EU law since 1952 in all the official language versions, go to EUR-Lex at: http://eur-lex.europa.eu/homepage.html?locale=en

Open data from the EU
The EU Open Data Portal (http://data.europa.eu/euodp/en/data) provides access to datasets from the EU. Data can be downloaded and reused for free, both for commercial and non-commercial purposes.