JUDGMENT OF THE GENERAL COURT (Ninth Chamber)

12 February 2020 (*)

(Civil service – Members of the temporary staff – Pension – Decision determining pension rights – Pension statements – Action for annulment – Time limit for lodging a complaint – Delay – Purely confirmatory act – Partial inadmissibility – Recovery of undue payments – Classification in grade and step – Multiplication factor – Withdrawal of an unlawful act – Legitimate expectation – Reasonable time)

In Case T‑605/18,

ZF, represented by J.-N. Louis, lawyer,

applicant,

v

European Commission, represented by B. Mongin and L. Radu Bouyon, acting as Agents,

defendant,

APPLICATION under Article 270 TFEU for annulment of the note of 30 November 2017 by which the Office for Administration and Payment of Individual Entitlements (PMO) of the Commission amended, with effect from 1 April 2015, the applicant’s retirement pension rights, and the note of 31 January 2018 in which it informed the applicant of the amount of the balance of the European Union’s debts against him,

THE GENERAL COURT (Ninth Chamber),

composed of S. Gervasoni (Rapporteur), President, L. Madise and R. da Silva Passos, Judges,

Registrar: M. Marescaux, Administrator,

having regard to the written part of the procedure and further to the hearing on 7 November 2019,

gives the following

Judgment

 Legal framework

1        Article 5(1) of the Staff Regulations of Officials of the European Union, in the version applicable before amendment by Council Regulation (EC, Euratom) No 723/2004 of 22 March 2004 amending the Staff Regulations of officials of the European Communities and the Conditions of Employment of other servants of the European Communities (OJ 2004 L 124, p. 1) (‘the old Staff Regulations’), provided for four categories of posts, designated in descending order of rank, by the letters ‘A’, ‘B’, ‘C’ and ‘D’. Category A consisted of eight grades, ranging from grade A 8, which was the lowest grade, to grade A 1, which was the highest grade.

2        Article 5 of the Staff Regulations of Officials of the European Union, in the version resulting from Regulation No 723/2004 (‘the Staff Regulations’), provides as follows:

‘1.      The posts covered by the Staff Regulations shall be classified, according to the nature and importance of the duties to which they relate, in an administrators’ function group (hereinafter ‘AD’) …

2.      Function group AD shall comprise twelve grades, corresponding to managerial, conceptual and analytical as well as to linguistic and scientific duties …’

3        The first paragraph of Article 44 of the Staff Regulations provides:

‘An official who has been at one step in his grade for two years shall automatically advance to the next step in that grade …’

4        Article 66 of the Staff Regulations contains a table setting out the basic monthly salary for each grade and step in function group AD. That table contains, for each grade, five steps, with the sole exception of grade 16, which has only three. The table in Article 66 of the old Staff Regulations included, as regards Category A, two steps in grade A 8, six steps in each of the grades A 7, A 2 and A 1 and eight steps in grades A 6, A 5, A 4 and A 3.

5        Article 107a of the Staff Regulations reads:

‘Transitional provisions are set out in Annex XIII.’

6        Under Article 1(1) of Annex XIII to the Staff Regulations:

‘1.      For the period from 1 May 2004 to 30 April 2006 Article 5(1) and (2) of the Staff Regulations are replaced by the following:

“1.      The posts covered by the Staff Regulations shall be classified, according to the nature and importance of the duties to which they relate, in four categories A*, B*, C* and D*, in descending order of rank.

2.      Category A* shall comprise twelve grades, category B* shall comprise nine grades, category C* shall comprise seven grades and category D* shall contain five grades.”’

7        In accordance with Article 2(1) of Annex XIII to the Staff Regulations, on 1 May 2004, subject to Article 8 of that annex, former grades A 1, A 2, A 3, A 4, A 5, A 6, A 7 and A 8 of officials placed in one of the posts referred to in Article 35 of the Staff Regulations were renamed A*16, A*15, A*14, A*12, A*11, A*10, A*8 and A*7 respectively. These are intermediate grades. Article 8(1) of that annex provides, inter alia, that, with effect from 1 May 2006, the abovementioned grades are renamed AD 16, AD 15, AD 14, AD 12, AD 11, AD 10, AD 8 and AD 7 respectively.

8        Article 2(2) of Annex XIII to the Staff Regulations provides:

‘Subject to the provisions of Article 7 of this Annex, basic monthly salaries shall be determined for each grade and step as provided for in the following tables (in euro).’

9        Article 2(2) of Annex XIII to the Staff Regulations then contains four tables, one for each of the former A, B, C and D categories. More specifically, in the table relating to former category A, a basic monthly salary for each of the various steps in intermediate grades A*16 to A*5 is shown in bold type, corresponding to the basic monthly salary laid down for the same grade and step in the table in Article 66 of the Staff Regulations.

10      For all intermediate grades, with the exception of grades A*13, A*9, A*6 and A*5, the table in question also contains the following additional information:

–        for each new intermediate grade, the corresponding grade in the former category A;

–        in bold type, a basic monthly salary also for steps 6 to 8, provided that the corresponding grade in former category A included such steps; these amounts have no equivalent in the table in Article 66 of the Staff Regulations;

–        below each monthly salary in bold type, a figure is shown in italics; a first footnote states that these figures ‘refer to the former salaries as set out in Article 66 [of the old Staff Regulations]’ and that they ‘are included in these tables merely for explanatory reasons and do not have any legal implication’; the figures in italics indicate, in all cases, a lower amount than that shown in bold type above;

–        below the figures in italics, a third line indicating a figure which is, in all cases, less than one; in a second footnote, it is specified: ‘The figure on the third line corresponding to each step is a coefficient representing the ratio between the basic salary before and after 1 May 2004’; in all cases, the figure indicated in the third line corresponds to the result of dividing the amount indicated in italics immediately above (second line) by the amount indicated in bold type in the first line.

11      The figure given in the third line for grade A*12, step 6, is 0.9426565.

12      Article 7 of Annex XIII to the Staff Regulations provides:

‘Basic monthly salaries of officials recruited before 1 May 2004 shall be determined in accordance with the following rules:

1.      The renaming of grades pursuant to Article 2(1) of this Annex shall not lead to any changes in the basic monthly salary paid to each official.

2.      For each official, a multiplication factor shall be calculated at 1 May 2004. This multiplication factor shall be equal to the ratio between the basic monthly salary paid to an official before 1 May 2004 and the applicable amount defined in Article 2(2) of this Annex.

The basic monthly salary paid to the official on 1 May 2004 shall be equal to the product of the applicable amount and the multiplication factor.

The multiplication factor shall be applied in order to determine the official’s basic monthly salary following advancement in step or update of remunerations.

5.      Without prejudice to paragraph 3, for each official, the first promotion after 1 May 2004 shall, depending on the category occupied before 1 May 2004 and the step occupied at the time the promotion takes effect, lead to an increase in basic monthly salary …

6.      A new multiplication factor shall be determined upon this first promotion. That multiplication factor shall be equal to the ratio between the new basic salaries resulting from the application of paragraph 5 and the applicable amount in Article 2(2) of this Annex. Subject to paragraph 7, this multiplication factor shall be applied to the salary after advancement in step and adaptation of remunerations.

…’

13      Article 8(2) of Annex III to the Staff Regulations provides:

‘Without prejudice to the provisions of Article 7 of this Annex, basic monthly salaries shall be determined for each grade and step on the basis of the table in Article 66 of the Staff Regulations. For officials who have been recruited before 1 May 2004 and until their first promotion comes into effect after that date, the table shall be as follows: [that table provided, as of 1 April 2015, in respect of grade AD 12, step 8, for a basic monthly salary of EUR 13 322.22].’

14      Article 10(1), Article 20(2) and Article 20(4) of the Conditions of Employment of Other Servants (‘the CEOS’) provide respectively that Article 5(1) and (2), Article 66 and Article 44 of the Staff Regulations are to apply by analogy to members of the temporary staff.

15      Article 1 of the annex to the CEOS, entitled ‘Transitional provisions applicable to the staff covered by the [CEOS]’, provides that the provisions of Annex XIII to the Staff Regulations are to apply by analogy to other servants employed on 30 April 2004.

 Background to the dispute

16      By a contract signed on 11 November 1999 with effect from 15 November 1999, the applicant was employed by the Council of the European Union as a member of the temporary staff. With effect from 1 January 2000, the applicant was classified in category A, grade 4, step 4.

17      At the time of the reform resulting from Regulation No 723/2004 (‘the 2004 reform’), the career system was amended. The Staff Regulations resulting from this reform established a career system in which the A, B, C and D categories of posts were grouped into two function groups, AD and AST. This new system was accompanied by a new salary scale. The salary scale applied before the 2004 reform provided for up to eight steps for certain grades of one of the four categories of posts, while the scale introduced in 2004 was amended with more grades (16), each with a smaller number of steps (up to five). Provision was made for a gradual transition between the two systems. The salaries of officials and servants remained unchanged, but were calculated by reference to the new salary scale. A multiplication factor was applied to allow for this to be maintained. This is a figure between 0 and 1, representing the ratio between the salary paid to the official or servant and the salary he or she would receive under the salary scale introduced by the 2004 reform.

18      On 1 May 2004, when the new salary scale entered into force, the applicant’s classification, namely grade A 4, step 6, was converted into grade A*12, step 6, with a multiplication factor of 0.9426565, in accordance with the table relating to former category A set out in Article 2(2) of Annex XIII to the Staff Regulations (see paragraphs 8 to 11 above).

19      On 1 May 2006, the applicant’s grade was renamed AD 12 (see paragraph 7 above).

20      On 1 November 2007, the applicant was classified in grade AD 12, step 8, after having been granted a second advancement in step since the entry into force of the 2004 reform.

21      Following the transfer to the European External Action Service (EEAS) of certain services and duties previously within the competence of the General Secretariat of the Council, the applicant himself was transferred to the EEAS as of October 2011. The applicant retained his grade and step, namely grade AD 12, step 8, and his seniority in step as on 1 November 2007.

22      The applicant’s salary, to which, until then, a multiplication factor of 0.9426565 had been applied, then had a multiplication factor of 1 applied to it, effective as of his February 2013 salary statement. A retroactive correction was also applied in this regard for the period from October 2011 to January 2013. As a result of that retroactive correction, an additional EUR 7 948.81 was paid to the applicant in February 2013.

23      On 2 July 2013, the applicant was appointed head of the European Security and Defence College (ESDC), while remaining in grade AD 12, step 8. He held that post until 31 December 2014 and then worked again at the EEAS from 1 January to 31 March 2015.

24      The applicant retired on 1 April 2015.

25      By note of 6 March 2015, the European Commission’s Office for the Administration and Payment of Individual Entitlements (PMO) determined the applicant’s retirement pension rights (‘the note of 6 March 2015’). The PMO took into account the period worked, namely from 15 November 1999 to 31 March 2015. The applicant was classified in grade AD 12, step 5. Nevertheless, a correction of 1.1314352 was applied so that, despite this change in step, the salary taken into account for the calculation of the pension was identical to the salary received by the applicant on the date of his retirement (‘the basic salary’). The applicant’s basic salary corresponded to that of grade AD 12, step 8, as shown in the table in Article 8(2) of Annex XIII to the Staff Regulations, that is to say, an amount of EUR 13 322.22 (see paragraph 13 above), that amount not being reduced because the applicant had a multiplication factor of 1 applied from February 2013 (see paragraph 22 above).

26      The pension subsequently paid to the applicant was consistent with that mentioned in the note of 6 March 2015.

27      When paying the pension of November 2015, the PMO mentioned on the applicant’s pension statement that there was overpayment resulting from the salary paid to him during the period from October 2011 to March 2015, that overpayment corresponding to a debt owed by the applicant to the European Union in the amount of EUR 22 896.98.

28      From February 2016, a monthly deduction of EUR 715.33 was applied.

29      In addition, by decision of 21 February 2013, the applicant was granted an extension of the dependent child allowance for his daughter for the period from 1 March 2012 to 28 February 2014. The applicant, however, continued to receive the dependent child allowance until March 2015.

30      When paying the June 2017 pension, the PMO mentioned on the applicant’s pension statement that there was an overpayment made during the period from March 2014 to March 2015 corresponding to a debt owed by the applicant to the European Union, by way of dependent child allowance, for a total amount of EUR 10 196.51.

31      The PMO also sent a letter dated 16 June 2017 to the applicant. By this letter, the PMO mentioned the amount of EUR 10 196.51 and presented a schedule providing for the application of a monthly deduction of EUR 728.32 between August 2017 and July 2018.

32      By letter of 19 October 2017, the applicant requested that the PMO terminate the deductions which had been applied to his pension (‘the letter of 19 October 2017’).

33      By note of 23 November 2017, responding to the letter of 19 October 2017, the PMO mentioned, inter alia, that the applicant’s salary had had applied to it, without justification, a multiplication factor equal to 1 as of his February 2013 salary statement, including, retroactively, for the period from October 2011 to January 2013 (‘the note of 23 November 2017’). Nonetheless, it was stated therein that the multiplication factor recorded in his personal file remained at 0.9426565. According to the PMO, that overpayment explains the existence of the applicant’s debt to the European Union in the amount of EUR 22 896.98 appearing on his November 2015 pension statement (see paragraph 27 above).

34      In the note of 23 November 2017, the PMO also stated that the applicant had continued to receive the dependent child allowance after the end, in February 2014, of the extension period which had been granted to him in that regard. According to the PMO, that overpayment explains the existence of the applicant’s debt to the European Union in the amount of EUR 10 196.51 (see paragraph 30 above).

35      That note also contained a summary of the deductions applied to the applicant’s pension corresponding to each of the two debts referred to in paragraphs 27 and 30 above respectively.

36      Lastly, in the same note of 23 November 2017, the PMO stated, with regard to the error relating to the multiplication factor (see paragraph 33 above), that the note of 6 March 2015 was based on an erroneous basic salary. The PMO then mentioned that a new note would be sent to the applicant on the matter by separate letter.

37      By note of 30 November 2017, the PMO amended the applicant’s retirement pension rights with effect from 1 April 2015 (‘the note of 30 November 2017’). His basic salary was modified, leading to the application of a correction of 1.066555 instead of 1.1314352 (see paragraph 25 above).

38      When paying the January 2018 pension, the PMO mentioned on the applicant’s pension statement the existence of an overpayment resulting from the amounts paid to him in respect of his pension during the period from April 2015 to December 2017. This overpayment corresponded to a debt owed by the applicant to the European Union for a total amount of EUR 7 389.51.

39      By letter of 4 January 2018 to the PMO, the applicant wished to produce evidence which, in his view, would help resolve the difficulties relating to the calculation of his past salaries and his pension. He concluded the letter by stating that the multiplication factor to be taken into account for the calculation of his salary and pension should be a value of 1 and not 0.9426565.

40      Following exchanges of letters and emails, the PMO, by note of 31 January 2018, informed the applicant that the balance of the European Union’s debts against him now amounted to EUR 22 409.61 (‘the note of 31 January 2018’). In a schedule setting out between three separate debts the monthly deductions applicable between February 2016 and January 2020, the PMO restated the total amount of each of those debts – namely EUR 22 896.98 for the first debt, EUR 10 196.51 for the second one and EUR 7 389.51 for the third one – and gave details of the current state of repayments for each of those debts.

41      On 28 February 2018, the applicant lodged a complaint under Article 90(2) of the Staff Regulations against, inter alia, the note of 30 November 2017 and the note of 31 January 2018.

42      By decision of 27 June 2018, the Commission rejected the complaint, taking the view that it was directed against acts confirming earlier decisions which had not been challenged within the required time limit and was therefore inadmissible.

 Procedure and forms of order sought

43      By application lodged at the Court Registry on 8 October 2018, the applicant brought the present action.

44      On 26 September 2019, in the context of the measures of organisation of procedure laid down in Article 89 of the Rules of Procedure, the Court put written questions to the parties. The parties replied to those questions within the time limit set.

45      The parties presented oral argument and replied to the Court’s oral questions at the hearing on 7 November 2019.

46      The applicant claims that the Court should:

–        annul ‘the Commission’s decision of 30 November 2017 fixing his pension rights with retroactive effect as of 6 March 2015’;

–        annul ‘the Commission’s decision of 31 January 2018 requiring an undue payment of EUR 22 409.61 to be recovered’;

–        order the Commission to pay the costs.

47      The Commission contends that the Court should:

–        dismiss the action;

–        order the applicant to pay the costs.

 Subject matter of the dispute

48      As is apparent from the background to the dispute, the PMO contends that there are three debts owed by the applicant to the European Union: the first one, which in November 2015 amounted to EUR 22 896.98, allegedly resulting from the erroneous application of a multiplication factor of 1 instead of 0.9426565 to the salary paid to the applicant during the period from October 2011 to March 2015 (‘the first debt’); the second one, which in June 2017 amounted to EUR 10 196.51, allegedly resulting from the unjustified receipt by the applicant of the dependent child allowance during the period from March 2014 to March 2015 (‘the second debt’); the third one, which in December 2017 amounted to EUR 7 389.51, allegedly resulting from the incorrect application, for the calculation of the applicant’s basic salary, and thus of his pension, of a multiplication factor of 1, leading the PMO to adopt a correction of 1.1314352 instead of 1.066555 during the period from April 2015 to December 2017 (‘the third debt’).

49      The applicant challenges two acts, namely the note of 30 November 2017 and the note of 31 January 2018.

50      The note of 30 November 2017, by retroactively applying to the applicant’s salary a multiplication factor of 0.9426565 instead of 1, modifies the basic salary taken into account for the calculation of the pension and gives rise to the third claim (see paragraphs 33, 36 and 37 above).

51      The note of 31 January 2018 informs the applicant of the amount, on that date, of the aggregate balance of the three debts owed by him to the European Union. It is accompanied by a schedule setting out the total amount of each of the three debts.

52      In this respect, it should be noted that the first and second debts gave rise, before January 2018, to deductions from the applicant’s pension, so that the net amount of the total balance of the three debts, namely EUR 22 409.61 on 31 January 2018, was, on that date, less than the sum of those debts.

53      As regards the second debt, the applicant stated at the hearing that he had withdrawn his claim relating to that debt, formal note of which was taken in the minutes of the hearing.

54      Furthermore, the schedule accompanying the note of 31 January 2018 gives details of the current repayments for each of the three debts (see paragraph 40 above).

55      In that regard, the applicant states, in his written pleadings, that he seeks the annulment of the ‘Commission’s decision of 31 January 2018 requiring that an undue payment of EUR 22 409.61 be recovered’. That clarification leads to the conclusion that only the existence of such a debt – not the arrangements for its repayment set out in the schedule mentioned in paragraph 51 above – is covered by the applicant’s claim for annulment. Such a conclusion is supported by the fact that the applicant did not at any time, either in his written pleadings or at the hearing, challenge those repayment arrangements.

56      It must therefore be concluded that the applicant merely requests the annulment of the note of 31 January 2018 in so far as it mentions the existence of debts owed by him to the European Union.

57      It follows from the foregoing considerations that the subject matter of the dispute is limited to the challenge of, first, the note of 30 November 2017, which concerns the third debt, relating to the pension paid to the applicant between April 2015 and December 2017, and, second, the note of 31 January 2018, in so far as it mentions the existence of that debt and in so far as it also mentions the existence of the first debt, relating to the salaries paid to the applicant for the period from October 2011 to March 2015.

 Admissibility

58      The Commission, which relied in the defence on the fact that the note of 30 November 2017 and that of 31 January 2018 were merely a reiteration of the decision reflected in the November 2015 pension statement, stated at the hearing that it had withdrawn its pleas of inadmissibility, formal note of which was taken in the minutes of the hearing.

59      Nevertheless, according to settled case-law, the periods for lodging complaints and bringing actions referred to in Articles 90 and 91 of the Staff Regulations are matters of public policy and cannot be left to the discretion of the parties or the Court, which must ascertain, of its own motion if need be, whether they have been complied with (judgments of 29 June 2000, Politi v ETF, C‑154/99 P, EU:C:2000:354, paragraph 15, and of 29 November 2018, WL v ERCEA, T‑493/17, not published, EU:T:2018:852, paragraph 64).

60      Under Article 90(2) of the Staff Regulations, any person to whom the Staff Regulations apply may submit to the appointing authority a complaint against an act affecting him adversely, which must be lodged within three months.

61      Salary or pension statements may be the subject of a complaint and ultimately an action, where a decision the subject matter of which is purely financial is liable, due to its nature, to be reflected in such a salary statement. In this case, notification of the monthly salary or pension statement has the effect of setting time running for the purposes of the time limits for lodging a complaint and bringing an action against an administrative decision, where the fact and scope of that decision become apparent, clearly and for the first time, from that statement (see judgment of 14 December 2017, Martinez De Prins and Others v EEAS, T‑575/16, EU:T:2017:911, paragraphs 31 and 32 and the case-law cited; see also, to that effect, judgment of 9 January 2007, Van Neyghem v Committee of the Regions, T‑288/04, EU:T:2007:1, paragraphs 39 and 40).

62      In the present case, as was stated in paragraph 27 above, when paying the pension of November 2015, the PMO informed the applicant of the existence of the first debt, which resulted, as is apparent from the pension statement, from an overpayment made monthly during the period from October 2011 to March 2015 and amounted in total to EUR 22 896.98.

63      The subject matter of the decision establishing the existence of the first debt is purely financial. It was therefore liable, due to its nature, to be reflected in the November 2015 pension statement. While it is regrettable that the PMO did not include any details in the statement as to the grounds of that decision, the fact remains that, since that statement clearly showed, for the first time, the fact and scope of that decision, its notification had the effect of setting time running for the purposes of the time limits for lodging a complaint and bringing an action against it.

64      As regards the date when notification of the decision establishing the existence of the first debt occurred, it is apparent from the documents in the file and it is confirmed by the applicant’s letter of 19 October 2017 (see paragraph 33 above) that he sent a message to the PMO on 19 February 2017 in which he stated that he had become aware of his November 2015 pension statement and of the deductions subsequently applied to his pension. He even stated in the letter of 19 October 2017 that he had ‘received’ that statement in November 2015.

65      Therefore, irrespective of the circumstances otherwise invoked by the applicant in order to show that he was not able to become immediately aware of his November 2015 pension statement, the statement may be regarded as having been notified to the applicant, in any event, no later than 19 February 2017.

66      However, the applicant does not mention, in his written pleadings, any complaint which he lodged against the November 2015 pension statement within the three-month time limit running from 19 February 2017.

67      It follows from the foregoing considerations that the applicant did not challenge within the prescribed time limits the decision establishing for the first time the existence of the first debt, which was reflected in his November 2015 pension statement.

68      The applicant’s arguments based on the provisions of Articles 25 and 26 of the Staff Regulations, according to which there is an obligation to notify the staff member of specific decisions concerning him or her, are not capable of calling into question the conclusion referred to in paragraph 67 above, in so far as those provisions are not intended to determine the conditions for the application to officials and other servants of the European Union of the procedural time limits (see, to that effect, judgments of 30 May 2002, Onidi v Commission, T‑197/00, EU:T:2002:135, paragraph 156, and of 5 October 2009, de Brito Sequeira Carvalho v Commission and Commission v de Brito Sequeira Carvalho, T‑40/07 P and T‑62/07 P, EU:T:2009:382, paragraph 92), contrary to the provisions of Title VII of the Staff Regulations, entitled ‘Appeals’, on the basis of which the judgments referred to in paragraph 61 above were delivered.

69      Before concluding that the applicant’s claims are inadmissible, in so far as they relate to the existence of the first debt, it must, however, be ascertained that the note of 31 January 2018 was not such as to reopen the procedural time limits in that regard. It is therefore necessary to determine whether or not the note of 31 January 2018, in so far as it mentions the existence of the first debt, constitutes a purely confirmatory decision of the decision reflected in the November 2015 pension statement.

70      According to the case-law, the confirmatory or other nature of a measure cannot be determined solely by comparing its content with that of the previous decision which it confirms but must also be appraised in the light of the nature of the request to which it constitutes a reply (see judgment of 17 November 2016, Fedtke v EESC, T‑157/16 P, not published, EU:T:2016:666, paragraph 17 and the case-law cited).

71      It is apparent, in particular, from that case-law that, if the act constitutes the reply to a request in which substantial new facts are relied on, and whereby the administration is requested to reconsider its previous decision, that act cannot be regarded as merely confirmatory in nature, since it constitutes a decision taken on the basis of those facts and thus contains a new factor as compared with the previous decision. The existence of substantial new facts may justify the submission of a request for reconsideration of a previous decision which has become final. Conversely, where the request for reconsideration is not based on substantial new facts, an action against a decision refusing to carry out the reconsideration must be declared inadmissible (see judgment of 17 November 2016, Fedtke v EESC, T‑157/16 P, not published, EU:T:2016:666, paragraph 18 and the case-law cited).

72      As regards the question of the criteria which determine whether facts are to be classified as new, it is clear from the case-law that, in order for a fact to be new, it is essential that neither the applicant nor the administration was aware of, or in a position to be aware of, the fact in question when the previous decision was adopted (see judgment of 17 November 2016, Fedtke v EESC, T‑157/16 P, not published, EU:T:2016:666, paragraph 19 and the case-law cited).

73      In this respect, it is certainly true that, in the judgment of 13 November 2014, Spain v Commission (T‑481/11, EU:T:2014:945, paragraph 38), it is stated that a factor must be classified as new both where that factor did not exist at the time of adoption of the earlier measure, and where that factor already existed when the earlier measure was adopted but, for whatever reason, including a failure by the author of that measure to act diligently, was not taken into consideration at the time of its adoption.

74      Nevertheless, despite its very broad wording, that clarification may not be interpreted as allowing an official, whose first application has been rejected, in whole or in part, by a decision which has become final, to invoke as new facts, in support of a second application having the same purpose as the first one, factors which were already known to him and which he failed to present in support of his first application (judgment of 17 November 2016, Fedtke v EESC, T‑157/16 P, not published, EU:T:2016:666, paragraph 22).

75      If that were not the case, not only would the case-law referred to in paragraphs 70 to 72 above be infringed, but also the case-law according to which the possibility of submitting a request as provided for in Article 90(1) of the Staff Regulations does not allow an official to set aside the time limits laid down in Articles 90 and 91 of the Staff Regulations for lodging a complaint and bringing an action by indirectly calling in question, by means of such a subsequent request, a previous decision which had not been challenged within the time limits (see judgment of 17 November 2016, Fedtke v EESC, T‑157/16 P, not published, EU:T:2016:666, paragraph 24 and the case-law cited).

76      In the present case, the factors to which the applicant refers in his letter of 4 January 2018, which are intended, in essence, to show that his appointment as head of the ESDC constituted an appointment to a new post with higher responsibilities, are factors which were known to the applicant when he held that post between July 2013 and December 2014 and which were thus known to him when he acquainted himself with the November 2015 pension statement. However, he did not present them to the PMO within the time limit prescribed for lodging a complaint. These factors are therefore not such as to enable him to reopen the time limit.

77      Therefore, the note of 31 January 2018, in so far as it mentions the existence of the first debt, must be regarded as a purely confirmatory decision of the decision reflected in the November 2015 pension statement.

78      It follows from the foregoing considerations that the claims made against the note of 31 January 2018, in so far as they mention the existence of the first debt, are inadmissible.

79      Since the applicant declared at the hearing that he had withdrawn his claims in respect of the second debt (see paragraph 53 above), the merits of his argument should be examined only in so far as it relates to the third debt, the establishment of which, resulting from the note of 30 November 2017, was reiterated in the note of 31 January 2018.

 Substance

80      In support of his claims for annulment, the applicant relies on six pleas in law, the first alleging an error of law, the second alleging infringement of Article 85 of the Staff Regulations, the third alleging infringement of the principles applicable to the withdrawal of lawful acts, the fourth alleging infringement of the principles applicable to the withdrawal of unlawful acts, the fifth alleging an inadequate statement of reasons and the sixth alleging manifest error of assessment.

 Error of law

81      The applicant submits that the retroactive correction of his pension is not justified and that the decisions the annulment of which he seeks are vitiated by an error of law.

82      The applicant disputes that the decision reflected in his February 2013 salary statement was adopted without cause.

83      The Commission contends that the applicant’s change of post and his entry into the service of the ESDC did not give rise to any change in grade or step.

84      In this respect, it should be noted (see paragraphs 18 to 20 above) that, on 1 May 2004, when the new salary scale entered into force, the applicant’s classification, namely grade A4, step 6, was converted into grade A*12, step 6, with the application, in accordance with the table relating to former category A in Article 2(2) of Annex XIII to the Staff Regulations, of a multiplication factor of 0.9426565.

85      On 1 May 2006, the applicant’s grade was renamed AD 12.

86      On 1 November 2007, the applicant was classified in grade AD 12, step 8, after having been granted a second advancement in step since the entry into force of the 2004 reform. The possibility of classifying an official or other servant in post before 1 May 2004 at a step higher than the highest step provided for in Article 66 of the Staff Regulations (see paragraph 4 above) is apparent from the table relating to former category A in Article 2(2) of Annex XIII to the Staff Regulations (see paragraph 10 above).

87      In the absence of promotion, the applicant continued in grade AD 12, step 8, as long as he was in post, on the basis of Article 8(2) of Annex XIII to the Staff Regulations (see paragraph 13 above).

88      When he retired in April 2015, the applicant was classified in grade AD 12, step 5. Nonetheless, a correction equal to 1.1314352 was applied to the salary corresponding to grade AD 12, step 5 so that the salary taken into account for the calculation of his pension was identical to the basic salary, that is to say, to that which he was receiving on the date of his retirement, namely EUR 13 322.22 (see paragraph 25 above).

89      It should be pointed out that the applicant did not, at the time of his retirement, dispute the change of step of which he was the subject.

90      The amount of the basic salary, referred to in paragraph 88 above, resulted from the application of the multiplication factor of 1 which had been applied in February 2013 (see paragraph 22 above).

91      However, it follows from Article 7(2) of Annex XIII to the Staff Regulations that the multiplication factor is calculated at 1 May 2004 (see paragraph 12 above).

92      In view of the applicant’s classification on that date (see paragraph 18 above), the multiplication factor should, in accordance with the table relating to former category A in Article 2(2) of Annex XIII to the Staff Regulations, be 0.9426565 (see paragraph 11 above).

93      Furthermore, the third subparagraph of Article 7(2) of Annex XIII to the Staff Regulations provides for the application of the multiplication factor calculated on 1 May 2004 following advancement in step or update of remunerations. Paragraph 6 of that same article provides for the determination of a new factor only upon the first promotion after 1 May 2004 (see paragraph 12 above).

94      However, it is common ground that the applicant was not promoted between 1 May 2004 and 1 April 2015, the date of his retirement.

95      Therefore, a multiplication factor of 0.9426565 had to continue to be applied until his retirement.

96      Nonetheless, a multiplication factor of 1 was applied to the applicant’s salary, effective as of his February 2013 salary statement (see paragraph 90 above).

97      There is nothing in the documents before the Court to suggest that the decision relating to the multiplication factor, reflected in the applicant’s February 2013 salary statement, is lawful whereas it has just been established that the applicant did not satisfy the conditions laid down in the legislation for the application of a new multiplication factor.

98      Contrary to what the applicant submits, such an amendment was therefore not justified in the light of the conditions laid down by the applicable provisions of the Staff Regulations.

99      As is apparent from the note of 6 March 2015, that unjustified amendment led to the calculation of the applicant’s pension on the basis of an amount of basic salary resulting from the application of a multiplication factor of 1 (see paragraph 90 above).

100    The correction made by the decision reflected in the November 2015 pension statement (see paragraph 33 above) and subsequently that effected by the note of 30 November 2017 was aimed at remedying the unjustified change mentioned in paragraph 98 above.

101    Thus, in the note of 30 November 2017, the PMO determined the amount of the basic salary this time, not on the basis of the salary which the applicant had received on the date of his retirement, which was determined on the basis of a multiplication factor of 1, but on the salary which the applicant should have received, which is determined on the basis of a multiplication factor of 0.9426565. Such a factor was therefore applied to the salary corresponding to grade AD 12, step 8, as shown in the table in Article 8(2) of Annex XIII to the Staff Regulations (see paragraph 87 above). On the date on which the applicant retired, the amount shown in that regard in that table was EUR 13 322.22 (see paragraph 13 above). The basic salary thus obtained was EUR 12 558.28, which led to a change in the correction applied so that the salary taken into account for the calculation of the pension was identical to the basic salary. The value of the correction thus changed from 1.1314352 to 1.066555 (see paragraph 37 above).

102    However, the applicant has not established that the application of the legislation thus carried out by the PMO was erroneous.

103    In particular, the applicant is wrong to claim that, due to his transfer to the EEAS in October 2011 (see paragraph 21 above), the transitional provisions of Annex XIII to the Staff Regulations were no longer applicable to him, which would have justified the abolition of the multiplication factor of 0.9426565 and the application of a multiplication factor of 1.

104    Even though the applicant’s taking up employment at the EEAS gave rise to the signing of a new contract, it resulted, as is apparent from a letter of 8 December 2010 sent to the applicant by the Secretary-General of the Council, from a transfer of staff accompanying the transfer to the EEAS of certain services and duties which had hitherto been covered by the General Secretariat of the Council. It was therefore not a recruitment marking the beginning of a new career to which the transitional provisions of Annex XIII to the Staff Regulations would not apply.

105    Moreover, when the applicant took up employment at the EEAS, his career was not interrupted. Therefore, the applicant’s classification of grade AD 12, step 8, which is applicable only by virtue of the transitional provisions of Annex XIII to the Staff Regulations, was maintained. Furthermore, his seniority in step prior to his entry into service at the EEAS on 1 October 2011 was maintained (see paragraph 21 above).

106    Thus, contrary to the applicant’s submission, the transitional provisions of Annex XIII to the Staff Regulations were still applicable to him after he took up employment at the EEAS on 1 October 2011.

107    It follows from the foregoing considerations that the applicant is wrong to claim that the retroactive correction made to his pension in November 2017 is not justified and that the note of 30 November 2017 is, for that reason, vitiated by an error of law.

108    Consequently, the present plea must be rejected.

 Breach of Article 85 of the Staff Regulations

109    The applicant submits that the multiplication factor applied to his salary while he was in service did not appear in his contract with the EEAS and that, therefore, he could not have been informed of any possible irregularity in the decision setting his rights when he joined the EEAS.

110    The applicant also relies on the complexity of the provisions at issue.

111    Lastly, the applicant states that no provision in the Staff Regulations enabled him to check the accuracy of the multiplication factor which had been applied to his salary, in particular, since February 2013.

112    The Commission contends that the conditions of Article 85 of the Staff Regulations were satisfied in the present case.

113    The Commission adds that the mere comparison of the applicant’s administrative situation as it appeared from a computer application enabling him to consult his personal file and his administrative situation stated in his salary statements enabled the applicant to identify an unexplained difference in the multiplication factor.

114    In this respect, it is apparent from Article 85 of the Staff Regulations, applicable to members of the temporary staff pursuant to Article 45 of the CEOS, that, for a sum paid without justification to be recovered, evidence must be produced to show that the recipient was actually aware that there was no due reason for the payment or that the fact of the overpayment was patently wrong such that he could not have been unaware of it. If, in the latter case, the recipient disputes that he was aware of the overpayment, the circumstances in which the payment was made must be examined in order to determine whether the overpayment was patently evident (judgment of 10 February 1994, White v Commission, T-107/92, EU:T:1994:17, paragraph 32).

115    The phrase ‘patently such’ relating to overpayment within the meaning of Article 85 of the Staff Regulations does not mean that an official who receives undue payments need make no effort to reflect or check but rather that recovery is due where the error is one which does not escape the notice of an official exercising ordinary care, who is deemed to know the rules governing his salary (judgments of 11 July 1979, Broe v Commission, 252/78, EU:C:1979:186, paragraph 13, and of 10 February 1994, White v Commission, T‑107/92, EU:T:1994:17, paragraph 33).

116    Article 85 of the Staff Regulations must be interpreted as meaning that it is not a question of whether or not the error was patent to the administration, but of whether it was patent to the person concerned. The situation of the administration, which is responsible for the payment of thousands of salaries and allowances of all kinds, cannot be compared to that of the official, who has a personal interest in checking the payments made to him every month (judgment of 11 July 1979, Broe v Commission, 252/78, EU:T:1979:186, paragraph 11). Although it is regrettable that it sometimes takes a long time for the administration to become aware of an overpayment, the fact remains that the staff member concerned, far from needing to make no effort to reflect or check, must detect an error which could not have escaped the attention of an official exercising ordinary care (see, to that effect, judgments of 10 February 1994, White v Commission, T‑107/92, EU:T:1994:17, paragraph 39, and of 18 June 2019, Quadri di Cardano v Commission, T‑828/17, not published, EU:T:2019:422, paragraph 63).

117    Furthermore, it is apparent from the case-law that the factors taken into account by the EU judicature in determining whether the error by the administration was ‘patently such’ include the level of responsibility of the official linked to his grade and seniority, and also the degree of clarity of the applicable Staff Regulations provisions setting out the conditions for granting the emoluments owing to the party concerned and the significance of the changes in his personal or family circumstances where payment of the sum at issue is linked to an assessment of such circumstances by the administration (see judgment of 18 June 2019, Quadri di Cardano v Commission, T‑828/17, not published, EU:T:2019:422, paragraph 48 and the case-law cited).

118    In addition, according to settled case-law, it is not necessary for the official concerned, in the exercise of his duty of diligence, to be able to determine the precise extent of the error made by the administration. The fact that he has doubts about the validity of the payments in question is sufficient for him to be obliged to contact the administration so that it can carry out the necessary checks (see judgment of 18 June 2019, Quadri di Cardano v Commission, T‑828/17, not published, EU:T:2019:422, paragraph 49 and the case-law cited).

119    In the present case, the error committed by the PMO results from the application to the applicant of a multiplication factor of 1 instead of 0.9426565. That error, which was first made in February 2013, continued thereafter, including at the time of the applicant’s retirement and the determination of his pension rights.

120    In this respect, as was stated in paragraph 22 above, the applicant’s salary, to which, until then, a multiplication factor of 0.9426565 had been applied, had applied to it, as from his February 2013 salary statement, a multiplication factor of 1, which corresponded to an increase in his salary of EUR 737.75 per month. In addition, a retroactive correction of his net remuneration amounting in total to EUR 7 948.81 was applied for the period from October 2011 to January 2013 and was paid to him in February 2013.

121    Such amendments, in view of their importance, were bound to attract the applicant’s attention.

122    Moreover, the applicant did not benefit, in October 2011, from a promotion which would have justified the determination of a new multiplication factor (see paragraphs 12 and 93 above). Nor did he receive any advancement in step which could have justified an increase in his salary.

123    Furthermore, the applicant’s transfer to the EEAS in October 2011 could not justify both the maintenance of a classification in grade AD 12, step 8, from which the applicant benefited under the transitional provisions of Annex XIII to the Staff Regulations and, at the same time, the abolition of the multiplication factor whose application resulted from the transitional provisions of Annex XIII to the Staff Regulations (see paragraphs 103 and 105 above).

124    Lastly, while it is true that the applicant was appointed head of the ESDC in 2013, he was maintained at grade AD 12, step 8 (see paragraph 23 above). Moreover, this appointment, which took place in July 2013, could not justify the payment, in February 2013, of a salary supplement for the period from October 2011 to January 2013.

125    Given the absence, in the applicant’s career, of events capable of justifying the changes referred to in paragraph 120 above, those changes should have been all the more obvious to the applicant.

126    Furthermore, it should be added that, in February 2013, the applicant had been in post for more than 13 years and was classified in grade AD 12, step 8, for which the corresponding salary is equivalent to that of grade AD 14, step 1. In addition, the applicant has claimed in his written pleadings that, subsequently, as head of the ESDC, he had ‘exercised the powers of the authority entitled to represent the ESDC for any legal act with financial and administrative implications’. Such a level of responsibility and the nature of the duties carried out by the applicant support the Commission’s argument that he could not have been unaware, by March 2015 at the latest, of the existence of the error made by the PMO.

127    Lastly, the Commission states, without being challenged on this point, that the mere comparison of the applicant’s administrative situation, as it appeared from the computer application enabling him to consult his personal file, and that administrative situation, as it appeared from his salary statements, revealed an inconsistency in relation to the multiplication factor applied to him (see paragraph 113 above).

128    It follows from the foregoing that, when he benefited, in February 2013, from an increase in his salary of EUR 737.75 and when he received, in the same month, a retroactive correction of his net remuneration in the total amount of EUR 7 948.81, the applicant should have had doubts, at the very least, as to whether he fulfilled the conditions for receiving such an increase without any change in his grade which could have justified the multiplication factor applied to him increasing from 0.9426565 to 1. With such doubts as to the lawfulness of the decision reflected in the salary statement for February 2013, it was incumbent on the applicant to refer the matter to the competent PMO departments (see, to that effect, judgments of 11 July 1979, Broe v Commission, 252/78, EU:C:1979:186, paragraph 13, and of 10 February 1994, White v Commission, T‑107/92, EU:T:1994:17, paragraph 42).

129    It does not appear from the documents in the file that the applicant ever undertook such steps, either at the time when the February 2013 salary statement was notified to him, or subsequently, in particular, when he received the note of 6 March 2015 determining his pension rights, although the patent error committed by the PMO was still continuing at that time.

130    Therefore, the infringement of Article 85 of the Staff Regulations has not been established.

131    That conclusion cannot be called into question by the fact that the PMO was negligent or erred when adopting the decision reflected in the February 2013 salary statement and then the note of 6 March 2015. Such circumstances do not affect the application of Article 85 of the Staff Regulations, which in fact presupposes that the administration has made a mistake in making the overpayment (judgments of 24 February 1994, Stahlschmidt v Parliament, T‑38/93, EU:T:1994:23, paragraph 23; of 30 November 2006, J v Commission, T‑379/04, EU:T:2006:368, paragraph 100; and of 16 May 2007, F v Commission, T‑324/04, EU:T:2007:140, paragraph 139).

132    Similarly, the finding in paragraph 130 above is not called into question by the applicant’s other arguments.

133    First, the argument that the applicant could not have been aware of a possible irregularity, since neither his contract nor the decision to recruit him referred to the multiplication factor, must be rejected in the light of the considerations set out in paragraphs 120 to 127 above.

134    In addition, the multiplication factor of 1 appeared on the February 2013 salary statement. The applicant was therefore able to be aware of its existence and its value.

135    Second, the determination of the multiplication factor which should have been applied to the applicant, namely 0.9426565, resulted from the combined application of the provisions set out in paragraphs 91 to 93 above to the applicant’s situation. Consequently, the applicant is wrong to claim that there was no provision in the Staff Regulations enabling him to check the accuracy of the multiplication factor applied to his salary.

136    Furthermore, it must be borne in mind that any official or other servant is deemed to be familiar with the Staff Regulations (judgment of 19 May 1999, Connolly v Commission, T‑34/96 and T‑163/96, EU:T:1999:102, paragraph 168). Thus, the applicant cannot claim that he was unaware of the fact and scope of those provisions, a fortiori in view of his level of responsibility and seniority.

137    It follows from the foregoing that the present plea must be rejected.

 Breach of the principles relating to the withdrawal of lawful acts

138    It should be recalled that, according to the case-law, the retroactive withdrawal of a lawful administrative act which has conferred individual rights or similar benefits is contrary to the general principles of law (see judgment of 27 June 2017, Ruiz Molina v EUIPO, T‑233/16 P, EU:T:2017:435, paragraph 26 and the case-law cited).

139    The applicant submits that the decision fixing his grade and step at the time of his transfer to the EEAS on 1 October 2011 was lawful and conferred individual rights on him. Therefore, in his view, it was not possible to withdraw that act.

140    The Commission submits that the PMO did not withdraw a lawful act.

141    It should be observed that the notes of 30 November 2017 and 31 January 2018, which the applicant seeks to have annulled, did not, contrary to what the applicant seems to argue, alter his grade or step. It was the note of 6 March 2015, which the applicant does not seek to have annulled and which, in that respect, was not amended by the note of 30 November 2017, which altered the applicant’s step (see paragraph 25 above).

142    In any event, the note of 6 March 2015, in so far as it was not amended in that respect by the note of 30 November 2017, became final, since, first, the applicant himself stated that it had been communicated to him in March 2015 and, second, it is not apparent from the documents in the file that the note was the subject of a complaint within the time limits laid down in the Staff Regulations.

143    It follows from the foregoing that the present plea must be rejected.

 Breach of the principles applicable to the withdrawal of unlawful acts

144    The applicant claims, in essence, that the Commission failed to have regard to the principles relating to the withdrawal of unlawful acts. He states that an institution may withdraw an unlawful act which confers individual rights only in so far as the withdrawal occurs within a reasonable period. He also invokes, in that regard, the principle of legal certainty and the application of a time limit of three months, corresponding to that set for an official or other servant to challenge an act of the administration.

145    The applicant also relies on his contractual relationship with the institution which employed him and on the fact that the employment conditions resulting from such a relationship could not be changed more than three months after he entered service and, a fortiori, after he had left the service. He adds that the EEAS was required to guarantee him all his rights, in particular his rights to remuneration. He also states, again in relation to the changes in his contractual relationship with the institution which employed him, that it was not until 23 November 2017, and without any decision to that effect having been notified to him, that he was informed ‘that his post was at the level of head of sector and that, as a result, he was not entitled to the managerial step’.

146    Lastly, the applicant states that the decision to apply a new multiplication factor to the remuneration received during his period of employment was never notified to him.

147    The Commission, which submitted in its written pleadings that it had not withdrawn an unlawful act, but had merely corrected an error, reversed that assertion at the hearing.

148    According to settled case-law, while it must be acknowledged that any EU institution, which finds that an act which it has just adopted is vitiated by unlawfulness, has the right to withdraw that act within a reasonable period, with retroactive effect, that right may be restricted by the need to respect the legitimate expectations of a beneficiary of the act, who has been led to rely on the lawfulness thereof (judgments of 20 June 1991, Cargill v Commission, C‑248/89, EU:C:1991:264, paragraph 20, and of 27 June 2017, Ruiz Molina v EUIPO, T‑233/16 P, EU:T:2017:435, paragraph 27).

149    Moreover, it should be borne in mind that an EU institution is entitled to withdraw an unlawful act only within a reasonable period (judgments of 17 April 1997, de Compte v Parliament, C‑90/95 P, EU:C:1997:198, paragraph 35, and of 27 June 2017, Ruiz Molina v EUIPO, T‑233/16 P, EU:T:2017:435, paragraph 27).

150    The withdrawal of an unlawful act which has benefited its recipient is therefore subject to two legal conditions, the first one being that it respects the legitimate expectation of the person concerned, and the second that it takes place within a reasonable period.

151    First, regarding the respect for the legitimate expectation of the person concerned, it is settled case-law that the right to rely on the principle of the protection of legitimate expectations presupposes the fulfilment of three cumulative conditions. First, precise, unconditional and consistent assurances originating from authorised and reliable sources must have been given to the person concerned by the administration. Second, those assurances must be such as to give rise to a legitimate expectation on the part of the person to whom they are addressed. Third, the assurances given must comply with the applicable rules (see judgment of 27 January 2016, Montagut Viladot v Commission, T‑696/14 P, EU:T:2016:30, paragraph 43 and the case-law cited).

152    Although the principle of the protection of legitimate expectations may restrict the administration’s right to withdraw an unlawful act with retroactive effect in a case where the addressee of the act has been led to rely on its apparent lawfulness, that condition is not deemed to be satisfied where there are objective circumstances which should have led the person concerned to realise the error in question or, in other words, where there are factors casting doubt on the lawfulness of the act. Thus, the person cannot rely on the apparent lawfulness of the withdrawn act, in particular where that act has no legal basis or was evidently adopted contrary to the applicable rules of law (judgment of 12 May 2010, Bui Van v Commission, T‑491/08 P, EU:T:2010:191, paragraph 44).

153    The case-law on withdrawal, with retroactive effect, of unlawful acts conferring individual rights aims to reconcile two principles, that of the protection of legitimate expectations, and that of lawfulness. In accordance with that case-law, where an official exercising due care could not fail to be aware of the unlawful nature of the act, his expectations cannot be regarded as legitimate and therefore the principle of legality is fully applicable (judgment of 12 May 2010, Bui Van v Commission, T‑491/08 P, EU:T:2010:191, paragraph 45).

154    It should be noted that the case-law relating to the application of the principle of legitimate expectations in relation to the withdrawal, with retroactive effect, of unlawful acts conferring individual rights, in particular the case-law referred to in the previous paragraph, echoes the case-law applicable to the recovery of undue payments (see paragraphs 115 to 118 above, in particular paragraph 115). Such convergence is not surprising, since Article 85 of the Staff Regulations is itself a manifestation of the principle of the protection of legitimate expectations (judgment of 13 March 1990, Costacurta v Commission, T‑34/89 and T‑67/89, EU:T:1990:20, paragraph 43).

155    Therefore, if an irregularity is such that it falls within the scope of Article 85 of the Staff Regulations, it cannot be such as to give rise to a legitimate expectation on the part of the person whom it benefits.

156    As was stated above (see paragraph 129), the PMO was able to apply, in the present case, the provisions of Article 85 of the Staff Regulations as interpreted by the case-law of the Court of Justice and the General Court, without erring (see paragraphs 115 to 118 above).

157    When the applicant received, in February 2013, an increase in his salary of EUR 737.75 from October 2011 and received, in the same month, a retroactive correction of his net remuneration in the total amount of EUR 7 948.81, he should have had doubts, at the very least, as to whether he fulfilled the conditions for receiving such an increase without any change in his grade which could have justified the multiplication factor applied to him increasing from 0.9426565 to 1 (see paragraphs 119 to 128 above).

158    It follows from the foregoing that, in the present case, the existence of a legitimate expectation has not been established.

159    As regards, in the second place, compliance with a reasonable time limit, it should be borne in mind that whether a period is reasonable must be assessed on the basis of all the circumstances of the case (judgment of 15 October 2002, Limburgse Vinyl Maatschappij and Others v Commission, C‑238/99 P, C‑244/99 P, C‑245/99 P, C‑247/99 P, C‑250/99 P to C‑252/99 P and C‑254/99 P, EU:C:2002:582, paragraph 187).

160    In particular, where the withdrawal of an unlawful act gives rise to the recovery of sums unduly paid, it is necessary to determine whether the subject matter of the act withdrawn is purely financial.

161    Where the subject matter of the act in question is purely financial, its withdrawal, which has the same effect as the recovery of sums unduly paid on the basis of that act, is the result merely of the application of the provisions of Article 85 of the Staff Regulations. In such a case, in order to ensure that the provisions of the first sentence of the second subparagraph of Article 85 of the Staff Regulations are of practical effect, the act in question must be withdrawn within the five-year time limit laid down in those provisions.

162    It should be noted that the concept of ‘decision the subject matter of which is purely financial’ has already been used by the Courts of the European Union in order to delimit the scope of the case-law which allows the time limits for lodging complaints and bringing actions to run from the notification of the salary or pension statement of the official or other servant concerned (see paragraph 61 above).

163    In that regard, a non-exhaustive list of decisions whose subject matter is purely financial – the fact and scope of which, by reason of their very subject matter, may clearly be apparent from a salary or pension statement sent individually to the official or other servant concerned – was drawn up by the Civil Service Tribunal in the judgment of 28 June 2006, Grünheid v Commission (F‑101/05, EU:F:2006:58, paragraphs 43 and 44). The Civil Service Tribunal thus referred, inter alia, to measures fixing correction coefficients, to the annual adjustment of remuneration, to the flat-rate repayment of travel expenses, to the refusal of entitlement to expatriation allowances or to the deductions of family allowances received from other sources.

164    It is necessary to distinguish decisions whose subject matter is purely financial from those which, while having financial effects, have a subject matter which goes beyond the fixing of the purely financial rights of the person concerned. This may consist, for example, in a decision to classify a newly recruited official definitively or a promotion decision.

165    In the present case, by the note of 30 November 2017, the PMO withdrew the note of 6 March 2015, in so far as it set the amount of salary taken into account for the calculation of the applicant’s pension on the basis of a multiplication factor of 1 (see paragraph 50 above). It therefore retroactively reduced the amount of the applicant’s pension and demanded repayment of the overpayment resulting from the payment of a higher pension amount between April 2015 and December 2017.

166    It is therefore an act the subject matter of which is purely financial and which was withdrawn by the note of 30 November 2017.

167    That withdrawal was effected within a period of approximately two years and nine months, which is shorter than the five-year time limit applicable in the present case (see paragraph 161 above).

168    It follows from the foregoing that infringement of the rules on the withdrawal of unlawful acts has not been established in the present case.

169    The conclusion in the previous paragraph is not called in question by the other arguments put forward by the applicant.

170    First, the arguments mentioned in paragraph 145 above relate to a change in the applicant’s step resulting from the note of 6 March 2015 and not the contested decisions.

171    Second, the principle of legal certainty (see paragraph 144 above) does not prevent the institutions of the European Union from withdrawing an unlawful administrative act after a period of three months. As was stated in paragraph 161 above, where, as in the present case, the subject matter of the act at issue is purely financial, the applicable time limit is that of five years laid down in the first sentence of the second subparagraph of Article 85 of the Staff Regulations.

172    Third, the argument referred to in paragraph 145 above does not concern the note of 30 November 2017, nor that of 31 January 2018, which did not alter either the remuneration which the applicant had received during his period of employment, or his grade, step or seniority in step, but merely applied to him a multiplication factor of 0.9426565 instead of 1, thereby altering the salary taken into account for the calculation of his pension.

173    In any event, it is not apparent from the documents before the Court that the applicant was able to benefit from an additional step to take into account the managerial duties he held.

174    Consequently, the present plea must be rejected.

 Insufficient statement of reasons

175    The applicant submits that the decisions which he contests are vitiated by a lack of any relevant statement of reasons.

176    He adds that the Commission refers, as regards the multiplication factor, to contradictory numerical values.

177    The Commission contends that sufficient reasons were given for the note of 30 November 2017 and that of 31 January 2018.

178    It should be noted that the purpose of the requirement laid down in Article 296 TFEU, and also contained in the second paragraph of Article 25 of the Staff Regulations, is to enable the EU Courts to review the legality of contested decisions and to provide the officials concerned with sufficient information to assess whether those decisions are well founded or subject to a defect enabling their legality to be challenged (judgments of 26 November 1981, Michel v Parliament, 195/80, EU:C:1981:284, paragraph 22; of 14 June 2018, Spagnolli and Others v Commission, T‑568/16 and T‑599/16, EU:T:2018:347, paragraph 68; and of 14 December 2018, UC v Parliament, T‑572/17, not published, EU:T:2018:975, paragraph 57).

179    Moreover, according to settled case-law, the statement of reasons for an act must be assessed with regard not only to its wording but also to its context and to all the legal rules governing the matter in question. Thus, grounds for a decision are sufficiently stated when the measure has been taken in circumstances known to the official concerned, which enable him to apprehend the scope of a measure taken in his regard (see judgment of 3 July 2019, PT v EIB, T‑573/16, EU:T:2019:481, paragraph 375 (not published) and the case-law cited).

180    In the present case, by a note of 23 November 2017, the PMO informed the applicant that, since the 2004 reform, the number of steps was limited to five and that, for that reason, he had been classified, at the time of his retirement, at grade AD 12, step 5, instead of grade AD 12, step 8 (which was his classification when he was still in post). A correction of 1.1314352 had, however, been applied to him so that his pension could be calculated on the basis of a salary equivalent to that which he received when he was still in service, namely EUR 13 322.22.

181    The PMO also recalled in that note that the multiplication factor, on the basis of which the applicant’s salary had been calculated in March 2015, before his retirement, had been modified in February 2013, from 0.9426565 to 1. Nevertheless, according to the PMO, there was no reason to do so as it was not linked to a promotion. It had thus caused an overpayment established in November 2015.

182    Also according to the PMO, in the note of 23 November 2017, the error relating to the multiplication factor affected the assessment of the applicant’s pension rights, the note of 6 March 2015 having been based on an incorrect multiplication factor and therefore on an incorrect basic salary. The PMO concluded in that regard by informing the applicant that a new note relating to his pension would be notified to him by separate letter.

183    In the note of 30 November 2017, the basic salary was amended so that it was set at EUR 12 558.28. The correction applied as a result of the change in step referred to in paragraph 180 above was, consequently, also amended to change from 1.1314352 to 1.066555. The document states that these changes were to take effect as of 1 April 2015.

184    In the January 2018 pension statement, in which the new value of the correction appears, namely 1.066555, reference is made to a debt owed to the European Union amounting to EUR 7 389.51, which corresponds to the sum of the debts established between April 2015 and December 2017, which also appear in that statement.

185    Lastly, as was stated in paragraph 40 above, in the schedule accompanying the note of 31 January 2018, the PMO gave details of the total amount of each of the three debts, this amount being EUR 7 389.51 for the third debt.

186    It is true that, in the note of 23 November 2017, the presentation of the link between, on the one hand, the multiplication factor allowing staff members’ salaries calculated by reference to the new salary scale introduced after the 2004 reform to be maintained (see paragraph 17 above) and, on the other hand, the correction applied so that the applicant’s pension can be calculated on the basis of a salary equivalent to the basic salary, that is to say to that which was his basic salary (or should have been his basic salary) when he was still in service (see paragraph 25 above), was made less clear by the fact that the PMO sometimes used the same term, ‘multiplikationsfaktor’, to designate without distinction the multiplication factor or the correction.

187    Nevertheless, it was possible, for a former staff member as experienced and reasonably informed as the applicant, who, moreover, had had a multiplication factor applied to him since the entry into force of the 2004 reform, whereas he has had a correction applied only since his retirement, in April 2015, to distinguish between those two factors.

188    It follows from the foregoing considerations that the applicant was in a position to ascertain the reasons for the note of 30 November 2017 and those of 31 January 2018, in so far as it established the existence of the third debt.

189    It follows that the plea that the statement of reasons was inadequate must be rejected.

 Manifest error of assessment

190    The applicant, after recalling several legislative provisions concerning the ESDC, states that, until his retirement, he assumed important administrative responsibilities and exercised staff management duties. The information that the EEAS provided to the PMO is therefore erroneous in his view.

191    The Commission states that the PMO took into account the applicant’s classification as established by the competent authorities during his career.

192    The fact, assuming it were established, that the applicant has performed high‑level duties, in particular staff management duties, is not capable of affecting the lawfulness of the note of 30 November 2017 and that of the note of 31 January 2018, in so far as it mentions the existence of the third debt, since the correction applied to the amount of the applicant’s pension as from the January 2018 pension statement is not based on the type of duties he carried out, but on the fact that, following the reform of 2004, he did not obtain any promotion capable of justifying that the multiplication factor applied to his salary would be retroactively amended in February 2013 and would continue to be applied thereafter (see paragraphs 181 to 183 above).

193    Consequently, the plea in law alleging a manifest error of assessment must be rejected.

194    Moreover, the argument that the Commission ‘has failed to provide the decisions pursuant to which [the applicant’s] pension statements were drawn up since 1 April 2015’ has no connection with the lawfulness of the note of 30 November 2017 or the note of 31 January 2018 which were notified to the applicant. That argument must therefore be rejected.

195    It follows from all of the foregoing that the action must be dismissed.

 Costs

196    Under Article 135(2) of the Rules of Procedure, the Court may order a party, even if successful, to pay some or all of the costs, if this appears justified by the conduct of that party, including before the proceedings were brought.

197    In the present case, as was noted in paragraph 63 above, it is regrettable that the PMO did not provide any explanations with the November 2015 pension statement as to the reasons for the decision establishing the existence of the first debt. Furthermore, after identifying the error relating to the multiplication factor applied to the applicant’s salary, the PMO took two years to draw all the appropriate conclusions from that error. Moreover, it was only through the note of 31 January 2018 that the applicant was able to obtain a complete schedule of all past and future repayments relating to each of the three debts. Lastly, in rejecting the complaint dated 27 June 2018, the Commission confined itself to invoking, wrongly in part, the inadmissibility of that complaint without replying on the substance to the applicant’s arguments, which were nonetheless substantive.

198    In the light of all the circumstances set out in paragraph 197 above, the Commission must be ordered to pay, in addition to its own costs, half of the applicant’s costs.

On those grounds,

THE GENERAL COURT (Ninth Chamber)

hereby:

1.      Dismisses the action;

2.      Orders the European Commission to pay, in addition to its own costs, half of ZF’s costs.

Gervasoni

Madise

da Silva Passos

Delivered in open court in Luxembourg on 12 February 2020.

[Signatures]


*      Language of the case: French.