Union of India & Anr Vs.
A.K. Narula  Insc 652 (18 May 2007)
CJI K G Balakrishnan & R V Raveendran
[Arising out of SLP (C) No.20719/2005] R.V. RAVEENDRAN, J.
Union of India has filed this appeal by special leave against the judgment
and order dated 27.5.2005 of the Punjab & Haryana High Court in Civil W.P.
2. Respondent was appointed as Deputy Superintendent of Police in the
Central Reserve Police Force (CRPF for short) on 1.12.1971. He was promoted as
Assistant Commandant on 23.7.1983. Promotion from the post of Assistant
Commandant to the next higher post of Second-in-Command is on the basis of
3. The O.M. dated 10.3.1989 issued by the Ministry of Personnel, Public
Grievances and Pension laid down the procedure of selection where the
Recruitment Rules required promotions to be made by selection. It provided that
where vacancies to be filled are more than four, the number of officers to be
considered by the Departmental Promotion Committee ('DPC' for short) shall be
three times the number of vacancies. It authorized the DPC to decide its own
method and procedure for objective assessment of the suitability of the
candidates. It excluded interviews, unless specifically provided for in the
relevant recruitment Rules and directed that Confidential Reports (CRs) were to
be the basic inputs for assessment. The O.M. required the DPC to follow the
procedure laid down therein to ensure that the evaluation of the CRs was fair,
just and non-discriminatory. A summary of the relevant portions of the
procedure laid down in the OM is given below:
(i) Suitability should be assessed on the basis of service record, with
particular reference to CRs for five preceding years. CRs.
of equal number of years should be considered in respect of all officers.
(ii) DPC should not be guided merely by the overall grading that may be
recorded in the CRs but should make its own assessment on the basis of entries
made in the CRs. If the reviewing authority or the accepting authority had overruled
the reporting authority or the reviewing authority as the case may be, the
remarks of the latter authority should be taken as the final remarks for
purposes of assessment provided it is apparent from the relevant entries that
the higher authority had come to a different assessment consciously after due
application of mind.
If the remarks of the reporting authority, reviewing authority and accepting
authority are complementary to each other, and one does not have the effect of
overruling the other, then final assessment should be made by the DPC, by
reading the remarks together.
(iii) DPC should give an overall grading to each officer whose CRs are
assessed, as outstanding, very good, good, average or unfit.
The panel of promotion should be prepared by the DPC based on the overall
grading assigned to each of the candidates considered by the DPC. Officers
graded as 'outstanding' would rank en bloc senior to those who are graded as
'very good' and officers graded as 'very good' would rank en bloc senior to
those who are graded as 'good' and placed in the select panel accordingly.
Officers with the same grading would maintain their inter se seniority in the
feeder post. Appointments from the panel should be made in the order of names
appearing in the panel.
(iv) For promotion to posts which are in the level of Rs.3700-5000 and
above, the benchmark grade should be "very good". For promotions to
Group 'A' posts carrying lesser scales of pay, the benchmark grade should be
4. A Departmental Promotion Committee ('DPC' for short) was convened to fill
up 48 vacancies of Second-in-Command on 13.6.1990. The DPC considered the CRs
of 144 eligible Assistant Commandants. The benchmark for promotion to the rank
of Second-in-Command was 'very good'. As per the parameters adopted by the DPC
held on 13.6.1990 on the basis of their prevailing instructions, a candidate
should have four 'very good' reports of ACRs out of last 5 ACRs including the
ACR for the latest year and should have no adverse remarks in CRs for the other
years under consideration for securing the benchmark of 'very good'. Out of 148
officers considered, DPC rated 39 officers as 'very good' and 80 officers as
After due deliberations, the DPC recommended that the 39 officers who had
secured the benchmark grade of 'very good' should be empanelled for promotion.
It also recommended that an additional nine senior-most Assistant Commandants
from those who were graded as 'good' be empanelled for promotion to the rank of
Second-in-Command so that 48 vacancies could be filled. The DPC further
empanelled the next ten senior- most Assistant Commandants graded as 'good' for
promotion against 10 anticipated vacancies.
5. The respondent had three ratings of 'good' for the years 1985-86, 1987-88
and 1988-89 and two ratings of 'very good' for the years 1986-87 and 1989-90.
The DPC, therefore, gave the overall rating of 'good' to the respondent. As
only 19 officers were needed from the list of officers with 'good' rating to
fill the existing as well as anticipated vacancies, only the first nineteen in
the list of officers with the rating 'good' were empanelled and promoted. The
respondent, though rated 'good', was not empanelled for promotion, as his name
did not figure in the first nineteen, having regard to the inter-se-seniority
among those who secured the overall rating of 'good'.
6. Feeling aggrieved, the respondent filed W.P. No.12316 of 1990 praying for
a direction to the appellants herein to include his name in the list of
officers approved for being appointed as Second-in-Command. The respondent
contended that his service record was similar to that of R. S.
Virk, another Assistant Commandant, who was promoted. It was contended that
both had secured 'very good' for three years and 'good' for two years, during
the five years under consideration (1985-86, 1986-87, 1987-88, 1988- 89 and
1989-90); that the DPC upgraded the rating of R.S. Virk for the year 1988-89
from 'good' to 'very good' and, consequently, gave him the overall rating of
'very good'; that in respect of the year 1987-88, his case was similar to that
of R.S.Virk for 1988-89, and therefore his rating for 1987-88 ought to have
been upgraded from 'good' to 'very good' and that if it had been done, he would
have also got the overall rating of 'very good' instead of 'good', thereby
becoming entitled to be placed in the block of 'very good' candidates selected
7. By order dated 11.2.1992, a learned Single Judge accepted the
Respondent's contention and allowed the writ petition holding that the DPC
having modulated the rating of R.S. Virk from 'good' to 'very good', failure to
do so in the case of respondent in similar circumstances would amount to unfair
and hostile discrimination. He also held that the appellants failed to prove that
those who were promoted had a better service record than respondent. He,
however, noted that as the respondent had been promoted as Second-in-Command
(during the pendency of the writ petition), all that survived for consideration
was whether respondent was entitled to be promoted from an earlier date. He
directed the appellants to include the name of the respondent in the approved
list of Assistant Commandants for appointment to the rank of Second-in-Command
from the date when anyone junior to him was promoted as Second-in-Command (in
pursuance of the selection by the DPC on 13.6.1990), with all consequential
8. The said judgment dated 11.2.1992 was challenged by the appellants herein
in LPA No.586 of 1992. The Division Bench found that in the case of R. S. Virk
(who was senior to respondent in the feeder post) for the year 1988-89, the
accepting authority had rated him as 'good', but DPC had upgraded the rating as
'very good' on the ground that the reporting authority and the reviewing authority
had rated him as 'very good', and the accepting authority had failed to record
any reason for downgrading him to 'good'. The Division Bench found that in the
case of Respondent for the year 1987-88, the reviewing authority had graded him
as 'very good' whereas the accepting authority had graded him as 'good' without
assigning any reason. The Division Bench was of the view that no plausible
explanation was disclosed for applying different yardsticks to R.S. Virk and
Respondent, that application of a different yardstick to Respondent would
amount to unfair treatment, and that if the same yardstick applied in the case
of R S Virk, had been applied to Respondent, the DPC would have rated the
Respondent also as 'very good' instead of 'good'. The Division Bench, however,
felt that the learned Single Judge ought not to have issued a direction to
include the name of Respondent in the approved list (and that too from the date
when a person junior to him was promoted) and ought to have left it to DPC to
reconsider the matter. The Division Bench, therefore, allowed the appeal in
part by judgment dated 12.7.2001, set aside the order of the learned Single
Judge and directed the DPC to reconsider the respondent's case, keeping in view
the observations made in the said judgment while adjudging his suitability. The
Division Bench also observed that if on reconsideration the respondent was
found suitable, he shall be promoted with retrospective effect from the date on
which R.S. Virk was promoted as Second-in- Command, and should be given all
9. Pursuant to the decision of the Division Bench, a review DPC was convened
on 4.12.2003. It reconsidered the case of the respondent. It held that there
was no ground to upgrade his rating for 1987-88 from 'good' to 'very good'.
Consequently, the overall rating of Respondent by the DPC remained 'good' and
DPC did not recommend him for promotion with respect to the DPC dated
13.6.1990. We extract below the reasoning given by the review DPC:
"In view of the observations made by the Hon'ble Court, the ACRs for
the period from 1.4.1985 to 31.3.1990 in respect of Shri A.K. Narula and Shri
R.S. Virk have been duly scrutinized by the DPC. Scrutiny of records reveals
that ACR of Shri A.K. Narula for the year 1987-88 is in two parts.
The first part is for 6 months from 1.4.1987 to 26.9.1987 in which he was
graded "Good" by Reporting Officer, "Very Good" by
Reviewing Officer and "Good" by Accepting Officer whereas, in the
second part of ACR for a period of six months from 3.10.1987 to 31.3.1988,
Reporting Officer, Reviewing Officer and Accepting Officer all had rated him as
Having gone through all these facts, the Review DPC is of the view that the
overall performance of Shri A.K. Narula during the year 1987-88 was "Good"
only. His case is not identical to that of Shri R.S. Virk since his ACR was for
full one year and Reporting and Reviewing Officer had graded him as "Very
Good" and it was the final Accepting Authority who had graded him as "Good"
without any reasons. Hence, no case is made out to upgrade the report of Shri
A.K. Narula from "Good" to "Very Good"."
The competent authority approved the recommendations of the review DPC and
the decision was communicated by order dated 5.1.2004.
10. This led to the second round of litigation. The respondent filed W.P.
No.4455 of 2004 wherein he sought quashing of the order dated 5.1.2004 and a
direction to the appellant to promote him to the rank of Second-in- command
with reference to DPC held on 13.6.1990 with all consequential benefits. A
Division Bench of the High Court allowed the said petition by judgment and
order dated 27.5.2005. The Division Bench held that the judgment dated
12.7.2001 of the Division Bench in the earlier round of litigation had held
that the respondent was not treated fairly and had specifically directed the
DPC to reconsider his case in the light of its observations; that the said
decision attained finality as the SLP filed by Appellants against the said
judgment dated 12.7.2001 was dismissed by this Court on 11.8.2003; that inspite
of the above, the review DPC had reiterated its earlier decision and thereby
failed to comply with the observations in the judgment dated 12.7.2001; and
that therefore, the decision of the review DPC could not be upheld. The High
Court, therefore, quashed the order dated 5.1.2004 and directed the appellants
to reconsider the case of the respondent for promotion with reference to the
DPC held on 13.6.1990 by treating the entry for the year 1987-88 as 'very
good'. The court also directed that respondent shall be granted all
consequential reliefs. The said decision is challenged in this appeal by
11. On the contentions urged, two questions arise for consideration:
(i) Whether the case of respondent was similar to or identical with that of
R.S. Virk and, consequently, DPC ought to have treated the CR of respondent for
the year 1987-88 as 'very good' thereby upgrading the overall grading from
'good' to 'very good'.
(ii) Whether the decision dated 17.2.2001 in the first round of litigation
had concluded the issue of entitlement of the respondent to the rating of 'very
good' for 1987-88 and consequently, all that was required of the review DPC was
to include him in the approved list of Assistant Commandants with the overall
rating of 'very good' and promote him to the rank of Second-in-Command' with
reference to DPC dated 13.6.1990 ? Re : Question (i) :
12. R.S. Virk had the rating of 'good' for 1985-86 and 1988-89. He had three
'very good' ratings and two 'good' rating during the five year period between
1985-86 and 1989-90. The records of R.S. Virk disclosed that for the year
1988-89, the reporting authority had graded him as 'very good'; the reviewing
authority had concurred and also rated him as 'very good'; but the accepting
authority had downgraded his rating from 'very good' to 'good' without
assigning any reason for not accepting the concurrent rating of 'very good' by
the reporting authority and reviewing authority. DPC was, therefore, of the
view that where both the reporting authority and reviewing authority had graded
the performance of R. S. Virk as 'very good', the accepting authority was not
justified in downgrading the rating without assigning any reason. Therefore
exercising the power given under the guidelines contained in the OM dated
10.3.1989, DPC modulated his rating for that year as 'very good'. As a
consequence, the 'very good' rating of R.S.
Virk increased from three to four, resulting in upgradation of overall DPC
rating from 'good' to 'very good'.
13. The respondent had also secured three 'very good' ratings for 1986-87,
1988-89 and 1989-90. For the other two years, that is for the years 1985-86 and
1987-88, the respondent had secured the rating of 'good'. The respondent
contended that his rating for 1987-88 was similar to that of R.S. Virk for
1988-89 as the accepting authority had downgraded his rating without assigning
any reason, and therefore, DPC ought to have modulated his rating for 1987-88
as 'very good'. On careful consideration, we find that respondent's case was
different from that of R. S. Virk. The CR of respondent for the year 1987-88
consisted of two parts. For the first half period, that is, from 1.4.1987 to
26.9.1987, the reporting authority had graded him as 'good' and reviewing
authority had graded him as 'very good' and the accepting authority had graded
him as 'good'. In regard to the second half, that is, 3.10.1987 to 31.3.1988,
all the three authorities (reporting authority, reviewing authority and the
accepting authority) had rated him as only 'good'. There was no question of
reviewing or upgrading the rating of respondent for the second half of 1987-88,
as all the three authorities had concurrently graded him as 'good'. Even in
regard to the first half of that year (1.4.1987 to 26.9.1987), unlike the case
of R.S. Virk where both the reporting authority and the reviewing authority had
unanimously given the rating 'very good', the reporting authority had rated the
respondent as 'good' and the reviewing authority had rated him as 'very good';
and in view of divergence between the reporting authority and reviewing
authority, the accepting authority chose to rate him as 'good'. The DPC, on
assessment, had rightly found that there was no case for revising the grading
of respondent. The review DPC also found that the facts of respondent's case
were different from that of R.S. Virk. The CRs demonstrated that R.S.Virk
deserved upgrading and respondent did not.
14. The guidelines give a certain amount of play in the joints to the DPC by
providing that it need not be guided by the overall grading recorded in the
CRs, but may make its own assessment on the basis of the entries in the CRs.
The DPC is required to make an overall assessment of the performance of each
candidate separately, but by adopting the same standards, yardsticks and norms.
It is only when the process of assessment is vitiated either on the ground of
bias, malafides or arbitrariness, the selection calls for interference.
Where the DPC has proceeded in a fair, impartial and reasonable manner, by
applying the same yardstick and norms to all candidates and there is no
arbitrariness in the process of assessment by the DPC, the court will not
interfere (vide State Bank of India v. Mohd. Mynuddin [1987 (4) SCC 486], Union
Public Service Commission v. Hiranyalal Dev [1988 (2) SCC 242] and Badrinath v.
Government of Tamil Nadu [2000 (8) SCC 395]). The review DPC reconsidered the
matter and has given detailed reasons as to why the case of the respondent was
not similar to that of R S Virk. If in those circumstances, the Review DPC
decided not to change the grading of the respondent for the period 1.4.1987 to
31.3.1988 from 'good' to 'very good', the overall grading of the respondent
continued to remain as 'good'.
There was no question of moving him from the block of officers with the
overall rating of 'good' to the block of officers with the overall rating of
'very good' and promoting him with reference to the DPC dated 13.6.1990. In the
absence of any allegation of mala fide or bias against the DPC and in the
absence of any arbitrariness in the manner in which assessment has been made,
the High Court was not justified in directing that the benefit of upgrading be
given to respondent, as was done in the case of R. S. Virk.
Re : Question (ii)
15. We have gone through the judgment dated 11.2.1992 of the learned Single
Judge in CWP No.12316/1990 and the judgment of the Division Bench dated
12.7.2001 in LPA No. 586 of 1992 in the first round of litigation.
16. In the case of R. S. Virk, the DPC held that there was no justification
for the accepting authority to downgrade the rating as 'good' without assigning
any reason, when both the reporting authority and the reviewing authority had
rated his performance as 'very good'. Therefore, it made its own assessment and
upgraded the rating of R.S.Virk for the year 1988-89 as 'very good'. As a
consequence, R.S. Virk had four 'very good' ratings and was placed in the block
of 'very good' candidates. The High Court was of the view that if the
downgrading of respondent's rating for the year 1987-88 by the accepting
authority from 'very good' to 'good' was similar to that of R. S.
Virk, then the same yardstick should be applied to Respondent and the rating
that has been downgraded by the accepting authority as 'good' should be
upgraded as 'very good' for 1987-88. It also observed that failure to do so
would be 'unfair'. Having made these observations, the High Court set aside the
decision of DPC for 1990 and directed a review DPC to reconsider the case of
the respondent. It follows, therefore, that review DPC was required to decide
whether the case of the respondent for the year 1987-88 was similar to that of
R. S. Virk for the year 1988-89, and if so, upgrade the rating of respondent
for the year 1987-88 as 'very good'. It is true that in that event respondent
also would have had four 'very good' rating and his overall rating would have
jumped from 'good' to 'very good'. But all that depended on the finding by
review DPC as to whether respondent was entitled for upgradation of rating for
17. The decision of the Single Judge dated 11.2.1992 merged with the
judgment dated 12.7.2001 in the LPA. There is no conclusive finding in the
judgment dated 12.7.2001 of the Division Bench that the respondent was entitled
for upgradation of the entry for 1987-88 from 'good' to 'very good'.
The Division Bench held that no plausible explanation was given for adopting
different yardsticks in the case of R S Virk and the respondent, if
respondent's case was similar to that R. S. Virk. The effect of the decision
was that if the downgrading of rating of Respondent for 1987-88 was similar to
the downgrading of the rating of R. S. Virk for 1988-89, the DPC should
modulate the rating of respondent for 1987-88 as 'very good' instead of 'good',
as was done in the case of R.S.Virk. The direction by the Division Bench that
'if on reconsideration, the respondent is found suitable, then he shall be
promoted with retrospective effect' and that 'DPC shall reconsider his case'
make it clear that DPC was required to examine whether the case of respondent
was identical to that of R. S. Virk. Only if the facts were the same,
respondent was entitled to be treated in a manner similar to R.S.Virk.
The Review DPC was, therefore, entitled to examine the case of respondent on
merits and take a decision.
18. In view of our finding that the case of respondent was different from
that of R.S.Virk, the decision of review DPC that the respondent did not
deserve upgradation from 'good' to 'very good', deserves to be upheld.
19. We, therefore, allow this appeal, set aside the judgment dated 27.5.2005
passed by the High Court and dismiss the writ petition upholding the decision
of the review DPC. Parties to bear their respective costs.