Brussels, 26.4.2022

COM(2022) 180 final

REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND THE COUNCIL

on the application of Annex XI to the Staff Regulations and Article 66a thereof


REPORT FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT AND THE COUNCIL

on the application of Annex XI to the Staff Regulations and Article 66a thereof

1.Introduction

Article 15(2) of Annex XI to the Staff Regulations (SR) as last amended in 2013 1 provides that, before 31 March 2022, the Commission shall submit a report to the European Parliament and the Council having regard to the survey conducted under Article 2 of the same Annex and assessing whether the evolution of purchasing power of remuneration and pensions of Union officials is in accordance with the changes in the purchasing power of salaries in national civil services in central governments. On the basis of this report, if appropriate, the Commission shall submit a proposal to amend Annex XI as well as Article 66a of the SR, on the basis of Article 336 of the Treaty on the Functioning of the European Union.

This report implements the aforementioned provision by assessing the fulfilment of EU institutions’ recruitment needs, describing the process and outcome of the implementation of the rules for updating remuneration and pensions under Annex XI to the SR (hereinafter the "Method") as well as the solidarity levy under Article 66a for the period 2014-2021.

2.EU institutions recruitment needs

2.1.Statutory requirement to survey recruitment needs

Article 2 of Annex XI of the SR provides that the Commission shall regularly survey the recruitment needs of the institutions.

While from an operational point of view recruitment aims at filling vacant positions within EU institutions, Article 27 of the SR and Articles 12 and 82 of the Conditions of Employment of Other Servants require, on the one hand, that officials and agents meet the highest standards of ability, efficiency and integrity, and, on the other hand, that they are recruited on the broadest possible geographical basis from among nationals of Member States of the EU.

2.2.Observations on recruitment needs

The European Personnel Selection Office (EPSO) is in charge of selecting officials, who represent the largest share of EU staff. EU institutions regularly communicate their recruitment needs to EPSO, which establishes a yearly plan to run competitions and defines selection methods and procedures, including marking rules for each competition. Together with the checks carried out upon recruitment, this guarantees that all recruits meet the highest standard of ability, efficiency and integrity, in line with Article 27 of the SR.

2.2.1.Decrease of applications to EU competitions

By comparing EU institutions’ recruitment needs for non-linguistic administrators (expressed in number of laureates sought) with competitions’ actual outcome, Annex 1 shows that, as a general rule, EPSO managed to select the requested number of laureates.

This being said, the numbers of applicants to AD generalist competitions declined for all clusters of Member States (Annex 2). For pre-2004 Member States, such a decline started after the 2004 reform of the SR, while for post-2004 Member States, the decline started after the 2014 reform of the SR (Annex 3) 2 . This affected the EU institutions’ ability to fulfil their recruitment needs, as well as to ensure a geographically balanced recruitment.

2.2.2.Pragmatic solution implemented by the Commission to fulfill specific recruitment needs

To address its recruitment needs, the Commission increasingly recruited temporary agents on permanent posts which are normally occupied by permanent officials. In the Commission, their total number outside linguistic services grew from 461 in 2013 (of which 118 at administrator level) to 1672 in 2020 (of which 825 at administrator level).

2.2.3.A geographically imbalanced recruitment

The Commission has encountered difficulties to recruit staff on the broadest possible geographical basis. In its 2018 report adopted pursuant to Article 27 of the SR 3 , the Commission notably identified the issue of attractiveness, translating in low levels of participation in EU competitions when compared to the size of the relevant population of some Member States. As a result, some nationalities are not sufficiently represented on EPSO lists of laureates. Annex 4 illustrates how, since 2018, applications to EU competitions remained unbalanced 4 , thereby resulting in geographically unbalanced recruitments. In particular, the number of significantly under-represented nationalities among junior administrators has risen to 13: Austrian, Czech, Estonian, German, Danish, Finnish, Cypriot, Luxembourgish, Maltese, Dutch, Polish, Slovak and Swedish 5 .

2.3.Correlation between EU institutions’ attractiveness and different macroeconomic and socioeconomic factors, including remuneration

The table hereinafter provides information regarding the level of participation in EU competitions for each Member State over the period 2013-2019. All Member States were included in the analysis except Belgium, the smallest Member States in terms of population and those whose level of participation is within the 80-120% bracket, i.e. close to the ideal “target” participation 6 . More precisely, excluded Member States are:

-Belgium because the overrepresentation of Belgian citizens among EPSO candidates is linked to the status of Brussels as the seat of several EU institutions 7 ,

-Cyprus, Luxembourg and Malta, as, in light of their small population size, the analysis of the share of candidates in EU competitions compared to the national population is not statistically sound, and

-Hungary, Slovakia and Finland, as their level of participation (comparing level of participation with percentage of EU population) is between 80% and 120% measured over seven years and therefore close to the target.

In order to assess factors influencing the level of participation in EU competitions for each Member State, it is necessary to examine EU institutions’ attractiveness across the EU. As the choice to apply to an EU competition remains ultimately a personal one that can be based on various factors, the Commission assessed the existence of possible correlations between a set of macroeconomic and socioeconomic factors and observed trends in the level of participation observed across Member States (Annex 5 to 5.7). In particular, the Commission assessed relative net earnings in Members States (Annexes 5.1 and 5.2), unemployment levels (Annex 5.3), the national employment situation (Annex 5.4), the image of the EU (Annex 5.5), personal financial prospects (Annex 5.6), and the OECD better life index (Annex 5.7).

On the basis of these data, the Commission assessed possible correlations between low participation in EU competitions and these socioeconomic and macroeconomic factors. The following conclusions could be drawn:

-participation of EU nationals in EU competitions is fairly correlated with net earnings in Member States and the perception of employment prospects at national level,

-if the analysis is restricted to the Member States with the highest and lowest wages, the level of participation to EU competitions and the level of net earnings are almost perfectly correlated 8 . This is particularly true for the Germany, France, the Netherlands, Austria, Denmark, Ireland, Sweden, Portugal, Slovenia, Lithuania, Romania, Croatia, Latvia and Bulgaria. For mid-wage Member States, results are however more nuanced and require taking on board other socioeconomic factors,

-the level of participation in EU competitions is highly correlated with the OECD better life index,

-the level of participation is weakly or not correlated with the national unemployment level, the image of the EU among citizens and the perception of improvement of one’s financial situation in the coming year.

3.Implementation of Article 65 of the Staff Regulations

3.1.Principle of parallelism

The principle of parallel development of purchasing power means that the purchasing power of EU staff members follows, both upwards and downwards, the evolution of the average purchasing power of civil servants in national central governments. Assessing the implementation of the principle of parallelism during the observed period may further clarify difficulties observed when it comes to EU institutions’ recruitment needs.

Under Article 65(1) of the SR, the update of remuneration and pensions reflects the annual evolution between 1 July of the previous year and 1 July of the current year 9 of:

-the real salaries of civil servants in the central governments in a sample of 10 Member States representing at least 75% of the EU Gross Domestic Product (EU GDP)  10 . For each of these Member States, the yearly change in real salaries is calculated net of inflation and is called the Specific Indicator. The Global Specific Indicator (GSI) is the average of all specific indicators weighted by the respective GDP of each Member State.

-the annual inflation in Brussels and Luxembourg (in the same proportion as the distribution of the EU staff between the two cities). This element is called the Joint Index.

The GSI and the Joint Index are multiplied to calculate the value of the update. The value of the update is expressed as a percentage that is applied across-the-board to the net remuneration and pensions of all EU staff, with effect on 1 July 11 .

In this context, it is important to point out that the Method does not allow ‘double counting’ of inflation, but guarantees the parallel evolution of salaries of EU staff and national civil servants, net of inflation. To this end, the GSI is first calculated net of inflation in the sample of 10 Member States, and only as a second step is it combined with the Joint Index. Detailed information on the evolution of the GSI and of the Joint Index are provided in part 3.

3.2.Specific indicators

This chapter explores whether the sample of Member States currently used to calculate the GSI translates correctly the real changes of purchasing power of all national civil servants and whether the general and specific derogations from the principle of parallelism inherent to the Method have led to a significant loss of purchasing power by EU civil servants.

3.2.1.Calculation of the Global Specific Indicator as defined in the 2013 Method

Under the Method adopted in 2013, the GSI was computed as of 1 July 2015, using a sample of eleven reference Member States (weighted by GDP): Belgium, Germany, Spain, France, Italy, Luxembourg, Netherlands, Austria, Poland, Sweden and the United Kingdom.

As of 1 January 2021 the United Kingdom is no longer considered an EU Member State and therefore since that date it is not included in the sample for the calculation of the GSI. Annex 6 shows that following the United Kingdom’s withdrawal from the European Union, the sample still meets the 75% threshold laid down in Article 1(4) of Annex XI to the SR and thus does not need to be amended. In implementation of the agreed inter-institutional approach concerning references to the United Kingdom in the EU legislation, for the purposes of Article 1(4) of Annex XI, Eurostat now uses a sample consisting of the 10 remaining Member States aforementioned.

Annual and cumulative data by Member State of the applicable sample

Annex 7 shows the time series of the specific indicators for the applicable sample of eleven Member States since 1 July 2015, together with a simple arithmetic mean value for the period. Annex 8 shows the same data, re-expressed as a cumulative index (2014=100).

Global Specific Indicator for the applicable sample

The cumulative impact of the GSI calculated and the GSI applied since I July 2015 is illustrated in Annex 9. The calculated GSI corresponds to the evolution of purchasing power of civil servants in the sample of Member States over the period (weighted by GDP expressed in Purchasing Power Standards). In accordance with the principle of parallelism (see part 3.1), the purchasing power of EU staff followed the same evolution as the sample, except since 1 July 2020, as a result of the implementation of the exception clause (see part 3.2.4).

Annex 10 shows the time series of the GSI since 1 July 2015, together with the Joint Index and the consequent annual update for EU staff. The GSI applied, computed as a simple average for the seven year period is 100.6, the cumulative impact of the GSI applied to July 2014 (2014=100) is 103.4.

3.2.2.The Global Specific Indicator as defined under previous Methods

The current legislative solution to use a sample of 10 Member States to calculate the GSI has not always been the approach chosen by the legislator. Prior to 2004, the GSI was calculated by reference to the statistical data for all Member States (e.g. 15 Member States in 2003), whereas in the period 2004-2012 a sample of 8 reference Member States was used.

The legislator’s decision to adopt the new Method was motivated by a retrospective data analysis suggesting that developments in that sample would reflect closely the average evolution in all Member States. However, it is to be noted that a subsequent analysis showed that the actual evolution of purchasing power in the selected sample after 2004 was lower than the average for all Member States. In 2013, the co-legislator maintained the principle of a sample of Member States, while increasing its size from 8 to 11.

To fulfil the statutory requirement to analyse whether the evolution of purchasing power of Union officials has been in accordance with the changes in the purchasing power of civil servants in central governments from all Member States, this chapter provides simulations of the evolution of the GSI in case a sample of EU 8 or in case no sample would have been used in the period 2015 - 2021.

With the sample of eight Member States 12 , the simple average of the yearly GSI for the period 2015-2021 would be 100.4 and the cumulative impact (2014=100) would be 102.5. With all Member States the simple average of the yearly GSI for the period 2015-2021 would be 101.1 and the cumulative impact (2014=100) would be 106.5. 

Annex 11 of this report compares the evolution of the GSI calculated on the basis of the above mentioned alternative samples. These data show that as a result of the sample of Member States defined in 2013 by the co-legislator, the purchasing power of EU staff only reached half of the changes in the purchasing power of salaries in national civil services in central governements, which may have resulted in a reduction in perceived attractiveness of the EU as an employer.

3.2.3.Specific Indicator: trends for individual Member States    

Calculations and figures relating to specific indicators in the Member States are based on data supplied and validated by the responsible statistical authorities of the Member States. To this end, a yearly remuneration questionnaire has been developed and updated over time. A set of country-specific methodology manuals and assessment reports is also being made public to raise the level of transparency 13 .

Annex 12 summarises the available data for each Member State for the period 2015-2021, together with a simple arithmetic average value for the period. Annex 13 shows the same data, re-expressed as a cumulative index (2014=100). Among Member States that are part of the sample, the highest net increases since 2015 are observed in Poland (121), Germany (109.8), Austria (109.6) and the Netherlands (109). The lower increases are observed in Luxembourg (101.6) and Belgium (101.8) and, within the sample, only central civil servants in France experienced a decrease of their net purchasing power (97.9).

Because the evolution of EU staff remuneration is calculated on the basis of the weighted average of a sample of Member States, and as detailed in part 3.2.1, over the same period, the remuneration of EU staff increased by 103.4.

Among Member States that are not part of the sample, the highest net increases of remuneration are observed in Slovakia (186.4), Czechia (167,1), Bulgaria (156.3) and Romania (148.1). Annex 14 shows that the vast majority of Member States that are part of the sample and have low participation rates in EU competitions organised by EPSO – as shown in part 2 – had specific indicators largely superior to the GSI on the basis of which EU staff remuneration has been updated. In other words, relative purchasing power attractiveness of the EU has declined for those countries.

3.2.4.General derogation to the principle of parallelism: application of the moderation and exception clauses

By derogation to the principle of parallelism of purchasing power evolution, a moderation clause limits high increases in purchasing power, as well as any high loss in purchasing power for a given year. If the calculated increase in purchasing power (GSI) exceeds 2% or if the purchasing power decreases by more than 2%, then the change in purchasing power for EU staff is limited to 2%. The part of the gain or loss in purchasing power exceeding 2% is applied nine months later, from 1 April of the following year.

A second general derogation is the exception clause, which limits the gain in purchasing power of EU staff when there is a downturn in the EU economy. If there is a forecast decrease in the EU GDP in real terms and there is a gain in purchasing power measured by the GSI, then depending on the magnitude of the GDP in real terms decrease, a part of the gain in purchasing power is postponed. If the EU GDP in real terms decreased by more than 3%, then the gain in purchasing power due to the Method is granted when the EU economy recovers, i.e. when the EU GDP in real terms reaches the pre-downturn level (recovery clause). If the final data delivered by the Commission on the EU GDP in real terms is different from the forecast, to the extent that it affects how the exception clause was applied, then the necessary corrections, including retroactive adjustments, either positive or negative, are made 14 .

Moderation clause: between 2015 and 2021, Eurostat duly verified the figures of the updates against the criteria for the moderation and exception clauses, as set out in Articles 10 and 11 of Annex XI to the SR. The GSI exceeded the statutory limit of +2% / -2% only in 2020, when the GSI measured (+2.5%) was above the threshold required to trigger the moderation clause. Given the concurrent application of the exception clause in the same year (2020), and in line with Article 10 of Annex XI to the SR, the moderation clause did not apply.

Exception clause: the annual EU GDP growth forecast available at the time of each yearly Eurostat report was positive, except in 2020, when, as a result of the COVID-19 pandemic, the forecasted evolution of the Union GDP in real terms was negative (-7.4 %) 15 . Therefore, the exception clause under Article 11(1) of Annex XI to the SR applied in 2020. Since the Union GDP decrease was greater than 3%, the GSI was not used to calculate the value of the update of remuneration and pensions. Hence, the whole GSI that would have been factored in the annual update for 2020 (+2.5%) will form the basis of the calculation of a future update once the cumulative increase of the Union GDP in real terms becomes positive, in accordance with Articles 10 and 11(4) of Annex XI to the SR.

The implementation of statutory general derogations to the principle of parallelism and, more specifically, of the exception clause in 2020 has been implemented strictly, allowing to limit the annual update resulting from the Method in times of economic downturns and thus providing net savings for the EU budget. On the other hand, this application of the exception clause impacted the parallel evolution of the purchaising power of EU civil servants, as for the same period civil servants in national governments from the sample effectively benefited from a higher purchasing power of 2.5% on average, thus resulting in a decline of attractivity for the EU since 2020.

Likewise, the 2020 annual update demonstrated the effectiveness of the automatic moderation clause, which was also introduced in 2014. Should the exception clause not have been triggered at the same time, the moderation clause would have automatically limited the increase of the GSI until April 2021.

Overall, it could be concluded that the automaticity of the Method, (i.e. remuneration updates based on statistical data which produce effects on the basis of the SR and without the intervention of any other legal act) has worked well since 2014. The automatic Method was fit for purpose, allowing to take account economic circumstances and purchasing power evolutions, while significantly reducing administrative burden compared to the annual recourse to ordinary legislative procedure used before 2014.

3.2.5.Specific derogation from the general principle of parallelism

To take account of the particularly difficult economic and social context in the Union at the time, as part of the 2014 reform it was also decided that the update of remuneration and pensions of all staff of the Union institutions, other bodies and agencies through the Method would be suspended in 2013 and 2014, and that the potential advantages for officials and other servants of the European Union derived from the application of the Method would be balanced by the reintroduction as of 1 January 2014 of a solidarity levy deducted from EU staff salaries at an increased rate.

3.2.5.1.The suspension of salary updates for two years (2013-2014)

As part of the reform of the SR in 2013, the European Council called for the adjustment of remuneration and pensions of all staff of the Union institutions through the Method to be suspended for two years. Thus, no update of remuneration and pensions took place in 2013 and 2014 16 , pursuant to the clear text of the SR. According to the European Court of Auditors, this two-year freeze generated 1.5 billion savings over the 2014-2020 Multiannual Financial Framework 17 .

While the suspension of the update discontinued the application of the principle of parallelism for the period 2013-2014, the correction coefficients continued to be updated to ensure equality of purchasing power between the different places of employment.

3.2.5.2.The impact of the EU solidarity levy

The solidarity levy is a deduction from the remuneration of EU staff that was re-introduced at an increased rate as from 1 January 2014. The rate of the solidarity levy, which is applied to the base defined in Article 66a(3) of the SR, is 6%. It is increased to 7% for staff in grade AD 15, step 2, and above. In the draft 2022 budget 18 , revenues from the solidarity levy are expected to reach EUR 106 million. Annex 15 shows the evolution of these revenues since 2012.

3.2.5.3.Assessment of the specific derogations to the principle of paralelism

As a result of these specific derogations, EU staff endured a significant loss in terms of real purchasing power. EU staff lost around 10.3% of their purchasing power, mainly due to the combined effect of the reforms of the SR in 2004 and in 2013, cuts in salary adjustments and the implementation of the exception clause in 2020. Over the same period, civil servants in central governments of the Member States gained on average 0.9 % in terms of purchasing power 19 .

3.3.The Joint Index

The Joint Index measures changes in the cost of living in Belgium and Luxembourg for EU staff according to the distribution of staff serving in these two Member States, based on the Harmonised Indices of Consumer Prices (HICP) in the case of Belgium and the Consumer Prices Index (CPI) in the case of Luxembourg, in accordance with Article 1 of Annex XI to the SR. Due to the specific consumption weights used for aggregating the Joint Index 20 , there may be differences between its evolution and the evolution pattern of the HICP for Belgium and the CPI for Luxembourg 21 for a given year.

Annex 16 shows the Joint Index time series (annual increase June-June by reference to previous year), together with information for Belgian HICP and Luxembourg CPI. The average Joint Index over the period from 2015 to 2021 is 101.4. The cumulative index for the whole period to June 2021 (base June 2014 = 100) is 110.6. By comparison, the average HICP in Belgium over the period is 101.6 and the cumulative total for the period is 111.4. The average CPI in Luxembourg over the period was 101.2 and the cumulative total for the period is 108.8.

As a result of the introduction of the Joint Index by the 2014 reform of the SR, part of the update corresponding to the inflation measured in Brussels and Luxembourg allowed to ensure consistency between adjustments paid to EU staff and inflations measured in Brussels and Luxembourg. Annex 17 shows that in case the Brussels International Index had continued to serve as a basis for the update of remuneration as was the case under the previous Method, the cumulative inflation measured during the reference period would have been almost the same as the one calculated on the basis of the Joint Index 22 . The cumulative total for the Joint Index is lower than the Belgian HICP and higher than Luxembourg’s CPI.  

4.Implementation of Article 64 of the Staff Regulations

4.1.Principle of equality of purchasing power among EU staff – correction coefficients

The correction coefficient puts into practice the general principle of equal treatment, which in this particular case corresponds to the equality of purchasing power between all staff of the EU institutions, bodies and agencies regardless of their place of employment. Whereas the same salary grids and basic amounts apply to all EU staff, correction coefficients apply to staff serving in posts outside Brussels and Luxembourg, who should neither suffer financially from higher living costs in their duty stations nor benefit from lower living costs 23 .

Hence the correction coefficient operates as a percentage adjustment to the salary to compensate the difference (either positive or negative) in the cost of living in each duty station. The correction coefficient is applied according to the following formula:

Correction coefficient

(= Economic parity / Exchange rate)

Salary in duty station 
(in euros for euro-countries)

Exchange rate 
(=1 for euro-countries)

Salary in Brussels 
(in euros)

=

X

X

   

Correction coefficients are updated at least once a year with effect on 1 July 24 .

The methodology for calculating correction coefficients is defined and regularly improved by Eurostat in cooperation with the national statistical institutes. For the period under consideration, it is worth mentioning the following developments:

-implementation of a more developed methodology for comparing healthcare and education costs, including fee-paying schools across Member States,

-data collection work has been adapted in the context of Brexit and of the Covid 19 pandemic’s consequences,

-an increasing number of organisations are willing to cooperate with Eurostat 25 and to use the data issued by Eurostat for different purposes.

Annex 18 shows correction coefficient values for individual duty stations in the EU 26 for the period 2013-2021. From this table it is apparent that the correction coefficients for different locations have followed different trends over time. In 15 locations they have decreased over the period, whilst they increased in 16 locations.

Since the entry into force of the amended Annex XI to the SR, a new procedure for creating or withdrawing correction coefficients outside of the capital city is in force, in accordance with Article 9 thereof 27 . Following the United Kingdom’s withdrawal from the European Union, the correction coefficient applicable in the United Kindom is calculated in line with the methodology applicable to third countries, since 1 February 2020 28 .

4.2.Issues faced by the Commission with the system of correction coefficients

4.2.1.Perception of correction coefficients as impacting negatively purchasing power and attractiveness of the EU public service

As identified in 2018 in the Commission’s interim report on the implementation of Annex XI 29 , correction coefficients are often mistakenly perceived as the driver of the decrease of purchasing power of EU staff. The Commission services are often asked by stakeholders, including staff, staff representatives, host State representatives about the role of correction coefficients and the methodology used for computing them. This issue arises in particular in cases involving staff whose place of employment is subject to a correction coefficient below 100, particularly when this specific coefficient is subject to downward updates 30 . 

Correction coefficients aim only at maintaining over time the equivalence of purchasing power between staff posted in different Member States and staff members posted in Brussels. Conversely, the correction coefficients do not aim at maintaining purchasing power at a given level. Hence, as the purchasing power of staff in Brussels decreased since 2004 (part 3.2.5), the purchasing power of staff in all places of employment decreased in the same proportion. Thus, the correction coefficients represent only the vehicle through which the adjustment is implemented in places outside Brussels and Luxembourg.

At the same time, the above perception of correction coefficients remains an issue related to attractiveness of the EU civil service in some Member States. In that regard, a number of EU agencies located in Member States where a correction coefficient below 100 applies reported to the Commission a visible negative impact on their capacity to recruit and retain highly-qualified and geographically balanced staff. In the same manner, agencies located in Member States with high correction coefficients also reported similar difficulties, highlighting the decrease of EU staff net purchasing power observed in many duty stations, regardless of the applicable correction coefficient.

In this context, the Commission ensures, at the occasion of each update of correction coefficients, comprehensive information is given to other institutions about the context of the update and provides detailed explanations on the fluctuation of different correction coefficients and the evolution of purchasing power of staff.

4.2.2.Lack of consideration for expenses made outside the place of employment

Under the SR, Eurostat and national statistical institutes are responsible for determining the statistical methodology applied to the calculation of correction coefficients and its implementation. The existing methodology developed by Eurostat in cooperation with national statistical institutes is based on the assumption that staff member's expenditure is entirely incurred in the place of employment. This approach is being analysed, as the overall consumption patterns of staff members also comprises expenditures outside the place of employment (‘out of area expenditures’).

In this regard, Eurostat and national statistical institutes agreed on the need to update family budget surveys in order to quantify the impact of ‘out of area expenditure’ on expatriate EU staff consumption. On that basis, experimental results were discussed among statisticians of the Commission and the Member States. In particular, Eurostat has started to explore the possibility to better reflect staff expenditure patterns by including expenses taking place outside the place of employment. Annex 19 illustrates results of this preliminary work and notably the flattening effect resulting from the inclusion of ‘out of area expenditures’ in EU staff consumption patterns.

When it comes to EU staff consumption patterns, the system of correction coefficients is focusing on bilateral comparisons between Brussels and the duty station, but it could be envisaged that it also takes into account ‘out of area’ expenditure. It is expected that Eurostat’s current experimental work on ‘out of area expenditure’ would be mature in the coming years as a statistical measure and could result in an evolution of the statistical methodology applied, pending agreement among statisticians in the Working Group on Articles 64 and 65 of the SR. Some of these changes could, however, be implemented under the current statutory framework.

4.2.3.Cost of living in Luxembourg

On the occasion of the last SR reform in 2014, the co-legislators maintained the established legislative solution that no correction coefficient shall be applicable to the remuneration of staff in Brussels and Luxembourg, having regard to the special referential role of those places of employment as principal and original seats of most of the institutions. At the same time, the co-legislators decided to take due account of inflation in Luxembourg by creating the Joint Index.

The absence of a correction coefficient in Luxembourg under the SR is being questioned by stakeholders and certain staff representatives. In this regard, following the Commission interim report on Annex XI 31 , this institution commissioned a study on the cost of living in Luxembourg to assess whether the legislative solution laid down in the SR remained fit for purpose. The Commission contracted AIRINC to carry out the study.

According to AIRINC, the accommodation in Luxembourg city in 2019 cost some 53% more than in Brussels. Rent has been the main driver for the deteriorating purchasing power of staff members working and residing in Luxembourg as compared with staff members working and residing in Brussels.

However, AIRNC concluded that Luxembourg is in a unique situation in that it has thousands of European institutions’ staff members residing in bordering regions, where the cost of living is significantly lower than in Brussels. These staff members nonetheless still reside at no greater distance from what would be compatible with the proper performance of their duties 32 . As a result of this geographical particularity of Luxembourg, the introduction of a correction coefficient calculated on the basis of the costs observed in Luxembourg may not fully grasp the particularities of Luxembourg as an EU place of employment 33 .

In addition, among 80 headings used by Eurostat to compare the differences between places of employment, housing is the main heading where the cost of living in Luxembourg is manifestly higher than that of Brussels 34 .

Moreover, some staff members may have acquired their dwellings at a time when the housing costs were lower or similar in Luxembourg compared to Brussels and have potentially experienced an asset price movement which sets them apart from new arrivals/existing staff who did not previously own property 35 . 

Thus, it is considered that introducing a correction coefficient at this stage might not contribute to ensure equality of purchasing power among EU staff. In addition, introducing a correction coefficient solely on the basis of a divergent price level/trend for a single component may introduce an element of instability, if the price level/trend of that single component ceases to diverge in future.

In this regard, together with other institutions concerned, the Commission is f assessing measures allowing to effectively address high housing costs, to which some of the staff residing in Luxembourg is exposed. Such measures would allow to address Luxembourg’s specifics and notably the situation of EU staff assigned in Luxembourg but residing in neighbouring countries.

A working group of representatives of the administrations of the EU Institutions with staff in Luxembourg is currently discussing the possibilities to address the challenges faced by most vulnerable staff members who are assigned to and reside in Luxembourg 36 .

In parallel, the Commission is looking into methodological developments on ‘out of area expenditure’, which could serve as an additional element in the reflection on the introduction of a correction coefficient for Luxembourg.

An assessment of the effects of the targeted social measure on the attractiveness of Luxembourg as a place of employment, together with the ‘out of area expenditure’ statistical work, should form part of the reflection on the opportunity to envisage any evolution in the current legal framework with a view of strengthening equality of purchasing power among EU staff.

5.Conclusions

Article 15(3) of Annex XI to the SR provides for the continued application of the Method and Article 66a of the SR beyond 2023, as long as the European Parliament and the Council have not adopted a regulation on the basis of a Commission proposal.

Since 2014, the Commission observed unequal attractiveness among applicants to EPSO generalist competitions resulting in geographical imbalances among Commission staff. Over the same period, the evolution of EU staff remuneration (+3.4) only reached half of the change in the purchasing power of salaries in national civil services in central governments (+ 6.5).

At the same time, the automatic Method provided under Annex XI to the SR:

-put aside inter-institutional tensions and court cases known from the past, while remaining under the constant scrutiny of the European Parliament and the Council via the annual reports delivered by the Commission.

-proved its efficiency, including in 2020, when the automatic implementation of the exception clause limited salary increases in a time of economic downturn and generated net savings for the EU budget,

-resulted in moderated salary increases and prevented – as from 2015 – EU staff from further losing purchasing power, by ensuring a cumulative annual update (114.3) close to the cumulative annual inflation measured by the Joint Index in Brussels and Luxembourg (110.6),

-resulted in a decrease of the purchasing power of EU staff by 0.2% in 2021,

-ensured EU staff’s continued contribution to the consolidation of public finances in the Union by re-introducing a solidarity levy at an increased rate,

-withstood legal scrutiny before the European Courts in the few individual cases brought against it and avoided social tensions, as no major strikes took place in the Institutions during that period.

Therefore, at this stage, the Commission considers it necessary to further evaluate possible methodological improvements to the system of correction coefficients, as well as to assess the opportunity of broadening the sample of Member States used for the calculation of the GSI, before submitting a proposal to amend Annex XI to the SR.

Consequently, in line with Article 15(3) of Annex XI to the SR, the Commission will rely on the provisional application of the current Method beyond 2023, whilst ensuring the continued annual reporting to the European Parliament and the Council on data pertaining to the budgetary impact of remuneration and pensions.



Annexes

Annex 1: EPSO non-linguistic AD competitions (all profiles) by publication year

AD'

laureates sought

laureates delivered

Fulfilment of needs

2013

474

457

96%

2014

340

359

106%

2015

375

383

102%

2016

351

361

103%

2017

290

313

108%

2018

423

435

103%

2019

585

591

101%

Annex 2: Number of Participants to AD level Generalists open competitions (Source: EPSO)

 

1998

September 2005

March 2010

March 2014

March 2015

March 2017

March 2018

March 2019

Avg 2017-19 vs Avg 1998-05

Avg 2014-15 vs Avg 1998-05

Avg 2017-19 vs Avg 2014-15

AT

755

1162

259

397

453

376

471

373

-58%

-56%

-4%

BE

2735

3905

1745

2036

2111

1635

1537

1423

-54%

-38%

-26%

DK

531

1179

130

273

233

153

161

169

-81%

-70%

-36%

FI

795

538

298

258

265

304

223

330

-57%

-61%

9%

FR

3647

4745

1696

2015

2176

1668

1643

1623

-61%

-50%

-22%

DE

2836

3573

1418

1947

2115

1473

1521

1464

-54%

-37%

-27%

EL

1075

873

1381

1812

2146

2533

2164

1844

124%

103%

10%

IE

425

279

153

182

172

167

176

191

-49%

-50%

1%

IT

8918

7212

3676

4608

6982

5071

4712

5331

-38%

-28%

-13%

LU

106

92

33

42

70

56

54

67

-40%

-43%

5%

NL

662

750

252

584

530

602

642

622

-12%

-21%

12%

PT

1098

1746

556

1083

1133

736

629

595

-54%

-22%

-41%

ES

4335

3397

1786

3602

3818

3108

3050

3356

-18%

-4%

-15%

SE

1314

943

276

248

253

221

208

211

-81%

-78%

-15%

EUR 14

29232

30394

13659

19087

22457

18103

17191

17599

-41%

-30%

-15%

CY

189

84

117

157

123

143

143

0%

CZ

420

319

283

348

214

235

404

-10%

EE

105

140

234

279

111

92

91

-62%

HU

667

572

747

673

459

442

380

-40%

LV

145

247

193

197

128

146

118

-33%

LT

342

447

377

468

290

302

267

-32%

MT

174

79

91

100

70

78

74

-23%

PL

1097

1086

1146

1300

855

838

778

-33%

SK

378

454

370

391

259

243

274

-32%

SI

201

209

237

318

201

196

162

-33%

EUR 10

29232

3718

3637

3795

4231

2710

2715

2691

-33%

BG

1078

1065

1142

773

734

656

-35%

RO

2332

2323

2278

1558

1530

1490

-34%

EUR 2

3410

3388

3420

2331

2264

2146

-34%

HR

725

713

435

393

372

-44%

Competition of March 2010 was affected by special circumstances and took place only in 2013. It is therefore not considered for comparison purposes

Annex 3: Number of Participants to administrator level generalists open competitions (by cluster)

Annex 4: Applications to EU competitions by nationality

2013 to 2019 Competitions

Applications
(number)

Applications per M. habitants

Applications
(share)

Population
(share)

Guiding Rate

Applications
vs Population

Applications
vs

Guiding Rate

EL

19847

1851

8,7%

2,4%

3,1%

362%

281%

CY

1278

1459

0,6%

0,2%

0,8%

286%

72%

BE

16314

1423

7,1%

2,6%

3,1%

279%

229%

MT

676

1370

0,3%

0,1%

0,6%

268%

46%

BG

8722

1246

3,8%

1,6%

2,4%

244%

158%

LT

3235

1158

1,4%

0,6%

1,5%

227%

95%

HR

4185

1027

1,8%

0,9%

1,6%

201%

116%

EE

1304

984

0,6%

0,3%

0,8%

193%

71%

RO

18236

940

8,0%

4,3%

4,5%

184%

177%

SI

1927

926

0,8%

0,5%

1,0%

181%

88%

LU

524

856

0,2%

0,1%

0,8%

168%

31%

LV

1577

821

0,7%

0,4%

1,0%

161%

72%

PT

8397

817

3,7%

2,3%

3,1%

160%

120%

IT

47460

777

20,8%

13,6%

11,2%

152%

186%

ES

29352

625

12,8%

10,5%

8,9%

122%

144%

SK

2963

544

1,3%

1,2%

1,8%

106%

73%

HU

4862

498

2,1%

2,2%

3,0%

97%

71%

FI

2402

436

1,1%

1,2%

1,8%

85%

59%

AT

3296

373

1,4%

2,0%

2,6%

73%

56%

IE

1656

338

0,7%

1,1%

1,6%

66%

45%

FR

17009

254

7,4%

15,0%

11,6%

50%

64%

DK

1459

252

0,6%

1,3%

1,8%

49%

36%

NL

4338

249

1,9%

3,9%

3,9%

49%

49%

CZ

2539

241

1,1%

2,4%

3,1%

47%

36%

PL

9025

238

3,9%

8,5%

8,2%

47%

48%

SE

1880

184

0,8%

2,3%

2,7%

36%

30%

DE

14057

169

6,2%

18,5%

13,8%

33%

45%

Annex 5: Correlation between EU institutions’ attractiveness and different macroeconomic and socioeconomic factors including remuneration

The correlation coefficient is a measure of linear correlation between two sets of data. It is the ratio between the covariance of two variables and the product of their standard deviations; thus, it is essentially a normalized measurement of the covariance, such that the result always has a value between -1 and 1. Any figure between -0.7 and -1 or between 0.7 and 1 is considered – in the framework of this report – as illustrating a strong correlation.

Among Member States included in the analysis, some were excluded, as their inclusion would have jeopardised the statistical soundness of correlations for the following reasons. Belgium was excluded because the overrepresentation of Belgian citizens among EPSO candidates is linked to the status of Brussels as the seat of several EU institutions. Cyprus, Luxembourg and Malta were excluded since the analysis of the share of candidates to EU competitions compared to the national population is not statistically sound, in light of their small population size. Hungary, Slovakia and Finland were excluded as their level of participation (comparing the level of participation with percentage of EU population) is between 80% and 120%, measured over seven years and therefore close to the target.

Annex 5.1: Correlation between net earnings and the level of participation to EU competitions – Eurostat data base

Annual net earnings in MS, i.e. the “net pay taken home”, originate from the Eurostat database and stand for gross earnings from which income taxes and social contributions are deducted and to which family allowances are added. The analysis looks at annual net earnings in 2019 in Member States. Net earnings are calculated from four surveys carried out both in the private and public sector. To remove the effect of differences in price levels between Member States, purchasing power parities were used to have earnings expressed in purchasing power standards, in compliance with Eurostat standard methodology. Given that the European Institutions strive to attract both single candidates and candidates with families, two different datasets reflecting two different situations were considered. First, a single person with no children earning 167% of an average worker and second, a two-earner couple with two children, both earning 100% of the average worker earnings.

Member States

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

LV

LT

NL

AT

PL

PT

RO

SI

SE

Share of candidates vs share of population

244%

47%

49%

33%

193%

66%

362%

122%

50%

201%

152%

161%

227%

49%

73%

47%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

1

-1

1

1

-1

1

1

1

1

-1

-1

-1

1

1

1

-1

Net earnings in 2019 for a single person without children earning 167% of the average earning (in PPS). BG = 1

1

1.33

2.12

2.40

1.36

2.03

1.46

1.81

2.00

1.09

1.64

1.02

1.20

2.49

2.30

1.48

1.28

1.15

1.26

1.98

Correlation

-0.76

Net earnings in 2019 for a couple with two children, both earning 100% of the average earning (in PPS). BG = 1

1

1.48

2.43

2.80

1.54

2.44

1.76

1.99

2.29

1.29

1.93

1.12

1.32

2.99

2.79

1.69

1.46

1.18

1.44

2.41

Correlation

-0.76

Annex 5.2: Correlation between net earnings in Member States with the highest and lowest wages and the level of participation to EU competitions – Eurostat data base

When restricting the analysis to the Member States with the highest and lowest wages, the level of participation to EU competitions and the level of net earnings are almost perfectly correlated. This is particularly true for Germany, France, the Netherlands, Austria, Denmark, Ireland, Sweden, Portugal, Slovenia, Lithuania, Romania, Croatia, Latvia and Bulgaria. For mid-wage Member States, results are however more nuanced and require taking on board other socioeconomic factors.

Member States

BG

DK

DE

IE

FR

HR

LV

LT

NL

AT

PT

RO

SI

SE

Share of candidates vs share of population

244%

49%

33%

66%

50%

201%

161%

227%

49%

73%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

-1

1

1

1

-1

-1

1

1

1

-1

Net earnings in 2019 for a single person without children earning 167% of the average earning (in PPS). BG = 1

1

2.12

2.40

2.03

2.00

1.09

1.02

1.20

2.49

2.30

1.28

1.15

1.26

1.98

Correlation

-0.96

Net earnings in 2019 for a couple with two children, both earning 100% of the average earning (in PPS). BG = 1

1

2.43

2.80

2.44

2.29

1.29

1.12

1.32

2.99

2.79

1.46

1.18

1.44

2.41

Correlation

-0.96

Annex 5.3: Correlation between unemployment levels and the level of participation to EU competitions – Eurostat data base

In this Annex, it is assessed whether the economic situation in Member States determines, to some extent, the willingness to work in a foreign country and for the European Institutions. A macroeconomic indicator that might influence this choice is the unemployment rate, also provided by Eurostat. In order to match the demographics of EPSO candidates, only unemployed persons aged between 25 and 49 in the second quarter of 2019 were incorporated in the analysis.

Member States

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

LV

LT

NL

AT

PL

PT

RO

SI

SE

Share of candidates vs share of population

244%

47%

49%

33%

193%

66%

362%

122%

50%

201%

152%

161%

227%

49%

73%

47%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

1

-1

1

1

-1

1

1

1

1

-1

-1

-1

1

1

1

-1

Unemployment level

4.3%

1.8%

4.8%

3.2%

4.0%

4.2%

17.6%

12.7%

7.5%

6.9%

10.5%

6.1%

6.0%

2.5%

4.5%

2.9%

5.3%

3.2%

3.9%

5.3%

Correlation

0.43

Annex 5.4: Correlation between the perception of national employment situation and the level of participation to EU competitions – Eurobarometer 2019)

Among the more than 200 questions in the Eurobarometer 37 , one relates to the judgment citizens have on the employment situation in their country.

Member States

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

LV

LT

NL

AT

PL

PT

RO

SI

SE

Share of candidates vs share of population

244%

47%

49%

33%

193%

66%

362%

122%

50%

201%

152%

161%

227%

49%

73%

47%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

1

-1

1

1

-1

1

1

1

1

-1

-1

-1

1

1

1

-1

Percentage of the population who thinks that the employment situation in their country is good

28%

80%

90%

73%

60%

65%

7%

13%

22%

24%

14%

37%

45%

85%

66%

65%

38%

38%

45%

65%

Correlation

-0.73



Annex 5.5: Correlation between the perception of the EU and the level of participation to EU competitions – Eurobarometer 2019

Many questions in the Eurobarometer address the image that citizens have about the European Union as a whole and the EU Institutions in particular. Even though EPSO candidates may not be necessarily representative of the EU citizens (in terms of age groups in particular), it may be posited that citizens in Member States that subscribe the most to the European project are those more likely to apply to the EU competitions.

Member States

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

LV

LT

NL

AT

PL

PT

RO

SI

SE

Share of candidates vs share of population

244%

47%

49%

33%

193%

66%

362%

122%

50%

201%

152%

161%

227%

49%

73%

47%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

1

-1

1

1

-1

1

1

1

1

-1

-1

-1

1

1

1

-1

Percentage of the population who have a positive image of the EU

61%

31%

55%

50%

49%

63%

31%

39%

36%

41%

33%

42%

50%

43%

38%

50%

59%

52%

44%

50%

Correlation

-0.04

Annex 5.6: Correlation between personal financial prospects and the level of participation to EU competitions – Eurobarometer 2019

In the Eurobarometer, citizens are asked to provide their expectations regarding the financial situation of their household in the following year. Expectations on the personal financial situation are dependent on global employment prospects, as well as on personal considerations regarding expected salary level, job stability, etc.

Member States

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

LV

LT

NL

AT

PL

PT

RO

SI

SE

Share of candidates vs share of population

244%

47%

49%

33%

193%

66%

362%

122%

50%

201%

152%

161%

227%

49%

73%

47%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

1

-1

1

1

-1

1

1

1

1

-1

-1

-1

1

1

1

-1

Percentage of the population who think that their own financial situation will improve in the next 12 months

26%

24%

21%

17%

31%

18%

23%

20%

24%

29%

23%

31%

25%

15%

16%

23%

27%

31%

20%

27%

Correlation

-0.56

Annex 5.7: Correlation between the OECD better life index and the level of participation to EU competitions – OECD

The OECD’s ‘better life index’ is a set of statistics aiming at comparing well-being across countries. Well-being encompasses material living conditions, housing conditions, employment, education, etc. For comparability purposes, the indicator chosen in the context of the study is the overall life satisfaction.

Member States

BG

CZ

DK

DE

EE

IE

EL

ES

FR

HR

IT

LV

LT

NL

AT

PL

PT

RO

SI

SE

Share of candidates vs share of population

244%

47%

49%

33%

193%

66%

362%

122%

50%

201%

152%

161%

227%

49%

73%

47%

160%

184%

181%

36%

Under represented = -1

Over represented = 1

1

-1

-1

-1

1

-1

1

1

-1

1

1

1

1

-1

-1

-1

1

1

1

-1

OECD better life index

6.7

7.6

7.0

5.7

7.0

5.4

6.3

6.5

6.0

5.9

5.9

7.4

7.1

6.1

5.4

5.9

7.3

Correlation

-0.84

Annex 6: Evolution of the sample of Member States and their respective weight (2014 = 100)



Annex 7: Annual specific indicators of Member States from the sample (2014 = 100)

Annex 8: Cumulative annual specific indicators of Member States from the sample (2014 = 100)

Annex 9: Calculated and applied Global Specific Indicator

Annex 10: Global Specific Indicator, Joint Index and annual update (2014 = 100)

 

GSI

Cumulative
2014 = 100

Joint Index

Cumulative
2014 = 100

Annual update

Cumulative
2014 =100

2015

101,2

101,2

101,2

101,2

102,4

102,4

2016

101,9

103,1

101,4

102,6

103,3

105,8

2017

100,4

103,5

101,1

103,7

101,5

107,4

2018

99,6

103,1

102,1

105,9

101,7

109,2

2019

100,5

103,6

101,5

107,5

102,0

111,4

2020

102,5

103,6

100,7

108,3

100,7

112,2

2021

99,8

103,4

102,1

110,6

101,9

114,3



Annex 11: Comparison of alternative GSI calculated based on different samples

Annex 12: Annual specific indicators of all Member States (2014 = 100)

Annex 13: Cumulative annual specific indicators of all Member States (2014 = 100)

Annex 14: Cumulative specific indicators of Member States from the sample and whose participation to EU competition is low

Annex 15: Proceeds from the solodarity levy between 2012 and 2022

Annex 16: Joint Belgium-Luxembourg index, Belgian HICP and Luxembourg CPI

Annex 17: Evolution of the cost of living in Brussels (Brussel international index)

Annex 18 : Evolution of correction coefficients



Annex 19: Effects of exploratory out of area work by ESTAT

Current correction coefficients –

July 2021

Estimated correction coefficient

including out of area expenditures

Change

BE

100

100

0.0%

BG

61.7

64.8

5.0%

CZ

88.1

89.4

1.5%

DK

134.2

129.8

-3.3%

DE

101.4

100.9

-0.5%

DE Karlsruhe

96.9

96.8

-0.1%

DE Munich

113.4

111.6

-1.6%

EE

86.3

87.7

1.6%

IE

133.6

128.7

-3.7%

EL

85.2

86.2

1.2%

ES

96.3

96.4

0.1%

FR

119.9

117.2

-2.3%

HR

78.3

80.3

2.6%

IT

95.2

95.3

0.1%

IT Varese

91.2

91.8

0.7%

CY

82.2

83.7

1.8%

LV

80.0

82.1

2.6%

LT

80.1

81.8

2.1%

HU

76.1

78.6

3.3%

MT

94.0

94.5

0.5%

NL

111.4

109.6

-1.6%

AT

109.6

107.3

-2.1%

PL

70.6

73.7

4.4%

PT

91.4

91.9

0.5%

RO

68.5

71.2

3.9%

SI

84.9

86.1

1.4%

SK

79.9

81.6

2.1%

FI

118.6

115.7

-2.4%

SE

130.3

126.4

-3.0%

(1)      Regulation (EU, Euratom) No 1023/2013 of the European Parliament and of the Council of 22 October 2013.
(2)      Only applications of Greek and Cypriot nationals increased or remained stable.
(3)      Commission report to the European Parliament and the Council of 15 June 2018, as corrected on 24 August 2018 by COM(2018)377 final/2.
(4)      In the framework of the present report, the level of participation for each Member State is defined by the ratio between the share of candidates and the share of EU population over the period 2013-2019. A 7-year period is used to smoothen data, as EU competitions are held irregularly. Moreover, due to the pandemic and its consequences on the organisation of these competitions, 2020 and 2021 are considered atypical years and are therefore excluded from the analysis.
(5)      Situation at 31 December 2021.
(6)      As the “target” (level of participation of 100%) can never be reached for statistical reasons, the choice was made to analyse only Member States where there is a clear imbalance. The cut-off level was set at +/- 20% compared to a 100% participation level. This is in line with the approach taken in the Commission report adopted pursuant to Article 27 of the Staff Regulations of Officials and to Article 12 of the Conditions of Employment of Other Servants of the European Union (Geographical balance) – COM(2018) 377 final.
(7)      The same phenomenon is observed in the headquarters of other international organisations. For example, German staff are overrepresented at the European Central Bank.
(8)      The United Nations (UN) deals with this issue by applying the so called ‘Noblemaire principle’, i.e. a formula which essentially sets the salary level of the UN by reference to the salary level paid by the civil service of the Member State which pays the highest (in recent years this has consistently been the United States). In consequence, the UN is able to guarantee interest from candidates of all Member States.
(9)      The methodology is based on the comparison of a snapshot of a national remuneration system of a Member State in the month of July of the current year with the equivalent snapshot in the same Member State in the month of July of the previous year.
(10)      Belgium, Germany, Spain, France, Italy, Luxembourg, the Netherlands, Austria, Poland and Sweden. Despite the withdrawal of the United Kingdom from the European Union, the sample of 10 Member States still represents more than 75% of the EU Gross Domestic Product as required under Article 1(4) of Annex XI to the SR.
(11)      To account for substantial changes in the cost of living in Brussels and Luxembourg, an intermediate update may take place as of 1 January, in accordance with Articles 4 to 7 of Annex XI to the SR.
(12)      Belgium, Germany, Spain, France, Italy, Luxembourg, Netherlands and United Kingdom.
(13)      The general methodology manual for calculating Specific Indicators is available on Eurostat’s website. In addition, 13 country-specific assessments have already published as of late 2018 and work is ongoing with other national authorities to increase the number of published assessments (see the following link: https://ec.europa.eu/eurostat/web/civil-servants-remuneration/specific-indicators/country-assessments ).
(14)      The Working Group on Articles 64 and 65 of the SR decided that the "final data" delivered by the Commission on the EU GDP for a given year would be interpreted as the data available by 30 September of the following year. This clarification was deemed to be necessary since GDP data can often be revised many years after the reference year.
(15)      In 2021, the final figure measured by Eurostat for the 2020 decrease of the EU GDP in real terms was -5.9%.
(16)      Pursuant to Article 65(4) of the SR.
(17)      European Court of Auditors, Special report 15/2019, Implementation of the 2014 staff reform package at the Commission - Big savings but not without consequences for staff.
(18)      Draft General budget of the European Union for the financial year 2022, COM (2021) 300.
(19)      Report from the Commission to the European Parliament and the Council on data pertaining to the budgetary impact of the 2021 annual update of remuneration and pensions of the officials and other servants of the European Union and the correction coefficients applied thereto COM(2021) 729.
(20)      As compared with the regular weights used for price statistics, the main difference for aggregating the Joint Index is the use of weights for rents that include owner-occupiers (instead of considering only tenants).
(21)      Values shown are for the overall aggregate index; actual calculations are done from detailed level.
(22)      Note that during the reference period, inflation rates were historically low for all component sub-indices. Note also that the calculation is using a consumption pattern for Belgium alone and might give slightly different results if combined with a specific consumption pattern for Luxembourg.
(23)      For staff members assigned in delegations, under Article 11 of Annex X, staff members have the possibility to request the payment of their remuneration in euros with the weighting applicable in Brussels.
(24)      An intermediate update may take place with effect on 1 January in the event of a substantial change in the cost of living between June and December.
(25)      For instance, the “family budget surveys” conducted to define consumption weights have been fully harmonised with the Coordinated Organisations (e.g. OECD, Council of Europe) and the United Nations decided in 2015 to use Eurostat source data on correction coefficients as input to adjust the salaries of their staff working in the EU (like Eurostat uses UN data for many locations outside the EU)..
(26)      In accordance with Articles 12 and 13 of Annex X to the SR, specific weightings (correction coefficients) may also apply to the remuneration of staff posted outside the EU. The updated value of these weightings is annexed to the yearly reports adopted by the Commission on data pertaining to the budgetary impact of the updates. Eurostat has developed a specific methodology in cooperation with national statistical authorities.
(27)      This procedure involves notably the adoption by the Commission of a delegated act following a formal request submitted by the appropriate authorities of the Member States concerned, the administration of an institution of the Union or the representatives of officials of the Union in a given place of employment. For staff assigned in Bonn, the German national statistical office stopped providing data for the calculation of this correction coefficient in Bonn in 2021. In the absence of data provided by the relevant national statistical office, and with regards to the very limited number of staff concerned, the correction coefficient applicable in Bonn has been suspended and replaced since 2021 by the correction coefficient applicable in Germany.
(28)      OJ C 211/8 of 25 June 2020 and OJ C 501, 13 December 2021.
(29)      Report from the Commission to the European Parliament and the Council on the application of Annex XI to the Staff Regulations and Article 66a thereof, COM (2018) 830.
(30)      A regular question received concern the local scope on the basis of which correction coefficients are calculated. The correction coefficients are calculated on the basis of a comparison of the cost of living in capital cities of Member States except in cases where local correction coefficients are calculated for some Member States in which case, local costs are reflected (see in that sense case T-618/20, FZ v. Commission).
(31)      COM(2018) 830.
(32)      This requirement is laid down by Article 20 of the SR.
(33)   https://ec.europa.eu/info/sites/default/files/study_on_the_cost_of_living_for_eu_staff_posted_in_luxembourg_-_final_report_v26-09-2019_0_0.pdf  
(34)      As the EU begins to exit a long period of historically low inflation, and the economies of Luxembourg and Belgium continue to evolve, differences seem starting to appear for certain other goods and services such as healthcare, education, furnishings, alcoholic beverages and tobacco.
(35)      Arguably this may be true in other duty stations but to a much more limited extent as a result of recruitment turnover.
(36)      A social measure could be adopted pursuant to Article 1(e) of the SR and would be limited to lowest grades staff while taking due account of their individual situation.
(37)      The Eurobarometer is a set of surveys that are used by the EU Institutions to monitor public opinion on various topics. In the context of this study, data come from the standard Eurobarometer 92 carried out in all EU Member States at the end of 2019.