At the Public Convenience?: The Role and Impact of the Pay Review Bodies

1
At the Public Convenience? The Role and Impact
of the Pay Review Bodies

• Peter Dolton
• Royal Holloway College, University of London
• Based on work with
• Gerry Makepeace
• Oscar Marcenaro
• and
• Adeline Aw

2
Two Main Questions

• What determines PRB Recommendations?
• PAPER What determines Public Sector Pay Review
  Body Decisions in the UK? with Adeline Aw.
• What has been the effect of PRBs?
• PAPER The Impact of the Public Sector Pay Review
  Bodies in the UK. with Gerry Makepeace and Oscar
  Marcenaro.

3
Outline of Talk

• 1. What are PRBs why do they exist?
• What are their remits
• 2. Do we have an Economic Theory to explain them?
• 3. What determines PRB Recommendations?
• 4. What has been the effect of PRBs?
• 5. What are the challenges for the future of PRBs

4
1. What are PRBs??
5
Pay Review Bodies (PRBs)

• Doctors and Dentists Review Body (DDRB),
• Senior Salaries Review Body (TSRB), the
• Armed Forces Pay Review Body (AFPRB),
• Nursing and Other Health Professions (NOHPRB),
• Prison Service Pay Review Body (PSPRB),
• School Teachers Review Body (STRB).
• Police Review Board (PAB)

6
Motivation

• After the hearing the arguments of the parties
  concerned the Tribunal retired and the Chairman
  outlined a possible decision for consideration by
  his colleagues. I recall with great vividness
  how, when asked whether I agreed with his
  proposal, I was aware only of a sensation of a
  great void opening for one cannot decide whether
  the salary of a particular post should be
  increased, or how great the increase should be,
  unless one has some principles from which to
  start.
• Baroness Barbara Wootton (1954) The Social
  Foundations of Wage Policy.

7
PRB Remit

• To reach recommendations having regard to
• Need to recruit, retain an motive staff.
• Regional/local variation in labour markets
• Government Departments output targets for the
  delivery of services.
• Government Departments expenditure limits.
• Governments Inflation target.

8
Key Features of PRB Activity

• Based on mutual consent of the parties
• Independent in their decisions
• Ensure Fair Play equal attention to the
  evidence.
• Free to determine their own methods of working.
• Theoretically not determined by an Envelope of
  Money

9
PM- Gordon Brown _at_ TUC Conference Brighton 10th
Sept 2007

• Talking about the upcoming public sector pay
  round being restricted to below inflation rises
• This government will always put stability first
  and there will be no loss of discipline, no
  resort to the easy options, no unaffordable
  promises and no taking risks with inflation.
• So let me be straightforward with you pay
  discipline is essential to prevent inflation, to
  maintain growth and create more jobs.. and so
  that we never return to the old boom and bust of
  the past.

10
Advantages to PRB System

• Avoid strikes and conflict
• Can be more efficient than collective bargaining
  with many rounds of negotiations.
• Objective gains trust of employees
• independent assessment of evidence.
• Government can be seen as not to blame for a
  low pay award.

11
Main Problems of PRBs

• De facto constrained by Expenditure Limits
• If not constrained then inflationary.
• If Government Departments step outside the PRBs
  and do contract deals separately e.g. GPs and
  Consultants.
• Not very easy to give differential awards by job,
  region etc.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.
• How can we create incentives in public sector
  delivery?

12
2. Do we have an Economic Theory of PRBs?
13
Is there a Economic Theory of Public Sector Pay?

• Assume
• Demographically determined inelastic demand.

14
(No Transcript)
15
Is there a Economic Theory of Public Sector Pay?

• Assume
• Demographically determined inelastic demand.
• Government Budget Constraint.

16
(No Transcript)
17
Is there a Economic Theory of Public Sector Pay?

• Assume
• Demographically determined inelastic demand.
• Government Budget Constraint.
• Homogenous Supply Government Sets Wages.

18
(No Transcript)
19
Is there a Economic Theory of Public Sector Pay?

• Assume
• Fairly inelastic demand.
• Government Budget Constraint.
• Powerful Monopoly Union.

20
(No Transcript)
21
3. What Determines PRB Remit Pay?
22
Differences between fitted and actual earnings
for Public and Private sector workers full time
workers aged 21-60
23
Evolution of PRB Pay Recommendations over 1971
2006
24
Evolution of Actual Pay Awards over 1971 2006
25
(No Transcript)
26
(No Transcript)
27
4. What has been the effect of PRBs?
28

• Compares the real weekly earnings of workers in
• (i) occupations covered by Pay Review Bodies
  (PRBs)
• (ii) the remainder of the public sector
  (non-PRB)
• (iii) private sector workers.
• Summarises LFS data from 1993 to 2006.

29
Occupational sub-groups

• PRB occupations
• Doctors and Dentists
• (Medical practitioners)
• Nurses and other health professions
• Nurses and Midwives
• Practices Allied to Medicine (separate from
  Nurses)
• School Teachers
• Primary School Teachers
• Secondary School Teachers
• Prison Officers
• Armed Forces
• Police (Not PRB)
• Not Senior Salaries Body

30
Occupational sub-groups

• PRB occupations (8 groups including Police)
• Other public sector workers are the non-PRB
  public sector.
• All others are private sector workers.

31
LFS Data

• Full time employees aged 20-60.
• Great Britain
• (Teachers England and Wales)
• It summarises LFS data from 1993 to 2006.
• Real weekly earnings.
• Dummies for occupational group and controls for
  compositional effects.
• (Age (7 groups), qualifications (6 NVQ levels),
  region (11)).

32

• Definition of occupations is based on SOC code.
• List of SOC codes define public sector (see
  Nickell and Quintini) e.g. General Administrators
  100 103, Local Government Officers 102,
  Civil Service Executive Officers 132, Fire
  service officers 153 etc
• PRB occupations defined as associated SOCs e.g.
  Medical Practitioners 220

33
A model

• Notation Individuals i Time t
• Thousands of individuals (i1,.., Nt) for each
  value of t .
• One value of t for each individual.
• Dummies
• PUB working in a non PRB public sector
  occupation
• PRBo (o1,.,O) in the oth PRB occupation.

34

• For each t
• Yit?t?tPUBit?o?otPRBoit?k?ktXkit?it
• Xkit are the control variables (age,
  qualification, region).
• ?ot Pay(oth PRB occupation) Pay(private
  sector)
• ?ot-?t Pay(oth PRB occupation) Pay(non-PRB
  public sector)

35

• Standard OLS assumptions and no X.
• Differences estimator. Causal effect.
• Standard OLS assumptions and X.
• Ceteris paribus effect of occupation.

36
Panel model

• Include fixed effects
• Yit?t?tPUBit ?o?otPRBoit?k?ktXkitAi?it
• True panel - take differences and apply OLS.
• Akin to a difference in differences estimator
• (at, bt, dot) estimates of (?t, ?t, ?ot)
• Suppose dot-dokgt0. Pay in the PRB occupation has
  improved relative to private sector pay.

37

• Basic argument
• OLS gives inconsistent estimators for each year.
  Asymptotic bias is constant from year to year.
• (ie. Assume that the occupational decision does
  nto change much between any two years
  reasonable not many changers in the Panel
  element of the LFS)

38

• Define the linear projections of Ak as
• Ai?PRIV?PUBPUBit?o?oPRBoit?k?kXkit?it.
• Estimating equation for each year
• Yit(?t?PRIV)(?t?PUB)PUBit?o(?ot?o)PRBoit
• ?k(?kt?k)Xkit ?it

39

• OLS estimates (at, bt, dot)
• at estimates ?t?PRIV
• bt ?t?PUB
• dot ?ot?o
• Consider estimates for two time periods.
• Plim(at-ak) ?t-?k
• Plim(bt-bk) ?t-?k
• Plim(dot-dok)?ot-?ok

40

• Suppose dot-dokgt0. Pay in the PRB occupation has
  improved relative to private sector pay.
• Suppose (dot-dok)-(bt-bk)gt0. Pay in the PRB
  occupation has improved relative to nonPRB
  public sector pay.

41
Comparisons with the private sectorValues of
estimate (see handout)

• Men with similar private sector men.
• Women with similar private sector women.

42
Table Effects of Working in PRB and Other
Occupations (Men)Changes in weekly wages from
year t-1 to year t
Non-PRB Public Sector (Men)
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
-9.4 -14.23 -16.08 -25.88 -27.46 -31.66 -34.94 -41.79 -60.68 -60.85 -59.12

• 1993 Non-PRB Public 9.4 less than private.

43
(No Transcript)
44
(No Transcript)
45
Comparisons of PRBs with the private sector

• Vertical distances in these figures to the axis
  show the extent of the difference between any PRB
  group and the Private sector.
• Vertical differences between any PRB group and
  the BOLD line of the Non-PRB line show the
  difference between each remit group and the
  Non-PRB public sector.

46
Comparisons of PRBs with the private sector

• Men have consistent negative differentials for
• Nurses and Midwives, Primary School Teachers,
• Secondary School Teachers, PAMs and
• Prison Officers (1997-2003)
• SO Armed Forces, Police and Doctors have done
  well.
• Women have a consistent negative differentials
  for
• Nurses and Midwives,
• Primary School Teachers (1999-2002)
• NOTE again Doctors have done well.

47
BUT

• Such comparisons
• PRBs with Private Sector
• are arguably just Compensating Wage
  differentials.
• This is not the Impact of the PRBs.
• Also does not tell us what we want to know we
  are looking for a trend DIFF-in-DIFF.
• i.e. difference in occupation over time compared
  to difference in NonPRB over time.
• This nets out for unobserved heterogeneity.

48
DIFF-in-DIFF IMPACT OF PRBs
49
Table Effects of Working in non PRB Public
Sector Changes in weekly wages from year t-1 to
year t
Non-PRB Public Sector (Men)
1993 1994 1995 1996 1997 1998 1999 2000
-9.4 -14.23 -16.08 -25.88 -27.46 -31.66 -34.94 -41.79
t-(t-1) -4.83 -1.85 -9.8 -1.58 -4.2 -3.28 -6.85

• 1993 Non-PRB Public 9.4 less than private.

50
Table Effects of Working in PRB Changes in
weekly wages from year t-1 to year t
Non-PRB Public Sector (Men)
1993 1994 1995 1996 1997 1998 1999 2000
-9.4 -14.23 -16.08 -25.88 -27.46 -31.66 -34.94 -41.79
t-(t-1) -4.83 -1.85 -9.8 -1.58 -4.2 -3.28 -6.85
Secondary School Teachers
1993 1994 1995 1996 1997 1998 1999 2000
-43.6 -40.8 -32.5 -57.7 -56 -71.9 -79.4 -77.4
t-(t-1) 2.8 8.3 -25.2 1.7 -15.9 -7.5 2.0
Diff.PRB 7.63 11.15 -15.4 3.28 -11.7 -4.22 8.85
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
Conclusions to Impact Evaluation

• No significant difference between any PRB group
  and the NonPRB Public Sector average.
• (Possible exception of the Doctors but then this
  was not a PRB decision to give GPs and
  Consultants a new contract.)

55
5. What are the Challenges for the Future for the
PRBs?
56
Challenges for the PRBs

• Not very easy to give differential awards by job.

57
(No Transcript)
58
(No Transcript)
59
Challenges for the PRBs

• Not very easy to give differential awards by job.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.

60
Challenges for the PRBs

• Not very easy to give differential awards by job.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.
• How can we create incentives in public sector
  delivery?

61
Challenges for the PRBs

• Not very easy to give differential awards by job.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.
• How can we create incentives in public sector
  delivery?
• How to take account of Pension Provisions in
  deliberations.

62
Challenges for the PRBs

• Not very easy to give differential awards by job.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.
• How can we create incentives in public sector
  delivery?
• How to take account of Pension Provisions in
  deliberations.
• Regional pay. Is it a Cost compensation issue?

63
Challenges for the PRBs

• Not very easy to give differential awards by job.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.
• How can we create incentives in public sector
  delivery?
• How to take account of Pension Provisions in
  deliberations.
• Regional pay. Is it a Cost compensation issue?
• Internationalisation of Labour Markets The
  Polish Dentist problem.

64
Challenges for the PRBs

• Not very easy to give differential awards by job.
• Not very easy to introduce PRP or relate
  remuneration to productivity in this setting.
• How can we create incentives in public sector
  delivery?
• How to take account of Pension Provisions in
  deliberations.
• Regional pay. Is it a Cost compensation issue?
• Internationalisation of Labour Markets The
  Polish Dentist problem.

65
Conclusions

• Time Series 1971-2006
• Inflation, Earnings rises, Unemployment, Public
  Sector Expenditure and industrial unrest all have
  a role to play in the determination of PRB pay
  awards.
• Blair era is characterised by lower
  recommendations.
• Cross Section (LFS) 1993-2006
• Conditional on attributes those in Public Sector
  are paid lower that those in the private sector.
• PRBs have no real impact on Earnings from
  1993-2006.
• But Different Conclusions for Different PRBs -
  depends on which PRB is considered.