Pay and Benefits in the Public Sector

1
Pay and Benefits in the Public Sector

• Martin Rama
• Development Research Group
• Public Expenditure Analysis
• Management Staff Training Course
• Washington, DC, May 22-24, 2001

2
The Problem

• The wage bill is usually the largest item in
  public sector spending
• Salaries and benefits result from historic
  legacies, fiscal adjustments, political
  pressures, etc.
• Pay tends to be compressed compared to the
  private sector (low top-to-bottom ratios)
• Low pay could lead to the departure of qualified
  staff and encourage corruption

3
The Solution

• Pay and benefits should be similar in the public
  and the private sectors
• An immediate alignment may not be feasible, but
  the direction should be clear
• The alignment may lead to gains for some public
  sector workers, and losses for others

4
Methodological Issues

• The private sector is very heterogeneous in
  developing countries (formal versus informal
  activities)
• Benefits account for a different share of total
  compensation in the public sector and in the
  private sector(s)
• Some important benefits (job security, low
  effort) are difficult to quantify

5
In Practice, Two Strategies

• The jobs approach estimate the total
  compensation associated with similar jobs in
  the private sector (e.g. secretary, driver)
• The workers approach estimate the total
  compensation a public sector worker would get if
  he/she had to move to the private sector

6
The Jobs Approach

• Widely used in industrial countries, and even by
  the World Bank (Hay points)
• Relies on the identification of appropriate
  private sector comparators
• Problem the chosen comparators are often the
  best companies in the formal sector
• Experience from downsizing shows that many public
  sector workers end up in the informal sector.
• Example the 2000 arbitration award in Guyana,
  doubling government pay.

7
An illustration State-owned Enterprises in
Vietnam
8
The Workers Approach

• Relies on the estimation of the pay and benefits
  a specific worker would get if he/she moved out
  of the public sector
• This is the same as estimating the amount of
  compensation he/she would need in the event of
  job separation
• Implication a public sector worker cannot be
  both underpaid and reluctant to leave in exchange
  for generous compensation

9
Implementation of this Approach

• Rely on household survey data (e.g. LSMS) to
  estimate private sector earnings as a function of
  worker characteristics (age, education, etc.)
• Predict alternative earnings of public sector
  workers by plugging their individual
  characteristics into the estimated earnings
  function
• Use the distribution of estimated earnings gaps
  to infer the value of intangible public sector
  benefits (more on this below)

10
Main Potential Weakness

• Public sector workers may be different from
  private sector workers in unobservable ways (e.g.
  they could be better connected, or less
  entrepreneurial)
• Failure to control for unobservable
  characteristics will bias the estimates (e.g. we
  will compare with the private sector earnings of
  someone who is not well connected, or is more
  entrepreneurial)
• How serious is this potential bias?

11
An Illustration State-owned Enterprises in
Vietnam

• If the bias from unobservable characteristics
  matters, then it has to matter in Vietnams SOEs
• Under central planning, SOE jobs were
  deliberately allocated based on political loyalty
  and to compensate for sacrifices during
  independence wars
• In the early 1990s, a voluntary downsizing
  program led to the departure of the most
  entrepreneurial third of the SOE workforce.

12
Six Ways to Estimate the Gap

• OLS Estimates earnings functions based on
  observable characteristics only
• HECKMAN Standard approach to correct for
  self-selection on unobservable characteristics
• SWITCHING More elaborate version of the above
• FE-TIME Focuses on workers who shift from public
  to private sector or vice-versa
• FE-HHOLD Focuses on individuals who belong to
  the same household but work in public versus
  private sector.
• MATCHING Compares with most similar private
  sector worker(s), rather than with a regression
  line.

13
Some Results
14
A Crude Comparison

• The jobs approach suggests that SOE workers earn
  4 to 8 times less than private sector workers
• The workers approach shows that they actually
  earn about 30 percent more
• The mean and median gap are stable, regardless of
  the econometric technique used
• Even individual gaps are quite stable, as
  revealed by the significant correlation
  coefficients

15
Assessing Benefits

• Both the jobs approach and the workers approach
  assume that benefits are the same in the public
  and the private sectors
• The comparators chosen in the jobs approach may
  offer similar benefits, but they require higher
  effort and provide less job security
• Benefits such as health coverage and old-age
  pension are usually unavailable in the informal
  sector

16
Methods to Assess Benefits

• Direct method Calculate the value of the most
  significant benefits available in the public
  sector but not in the informal sector (e.g.
  old-age pension)
• Indirect method If a worker who could earn X
  more in the private sector does not leave
  voluntarily, he/she must value the intangible
  benefits at X of his/her salary at least.

17
An Illustration Civil Servants in Guinea-Bissau
18
Main Findings in Guinea-Bissau

• All civil servants in N-Z categories would lose
  if they moved to the informal sector, but a
  quarter of them would earn more in the private
  formal sector
• Based on the estimated gains, and the probability
  of finding a job in the private formal sector,
  the value of intangible benefits is around 20
  percent of the measurable earnings
• The direct method, based on the present value of
  old-age pension, yields similar results

19
Conclusions

• Having clear benchmarks is important to reform
  public sector pay over time, minimizing risks of
  either over-spending or losing qualified
  personnel
• The jobs approach yields an excessively high
  benchmark, because it focuses on the best
  companies, and not on the true alternatives of
  public sector workers
• The workers approach can be implemented in any
  country with decent household survey data, of the
  kind we use in our poverty work

20
Conclusions (Contd.)

• The main weakness of the workers approach is the
  potential bias created by unobservable worker
  characteristics
• However, this bias does not appear to be
  substantial, and even simple econometric
  techniques (such as OLS) could yield decent
  results
• The workers approach can also be used to estimate
  the value of intangible benefits offered by the
  public sector, which are part of total
  compensation

21
References

• Cited in the presentation
• Bales, Sarah and Martin Rama (2001) Are Public
  Sector Workers Underpaid? Appropriate
  Comparators in a Developing Country, work in
  progress, Washington, DC The World Bank.
• Chong, Alberto and Martín Rama (2001) Do
  Compensation Packages Need to Be that Generous?
  Simulations for Government Employees in
  Guinea-Bissau, in Shantayanan Devarajan, F.
  Halsey Rogers and Lyn Squire (eds.) World Bank
  Economists Forum, 1, p. 169-194, Washington, DC
  The World Bank.
• Other references
• Adamchik, Vera A. and Arjun S. Bedi (2000) Wage
  Differentials between the Public and the Private
  Sector Evidence from an Economy in Transition,
  Labour Economics, 7(2), p. 203-224, March.
• Filmer, Deon and David Lindauer (2001) Does
  Indonesia Have a Low Pay Civil Service?
  unpublished manuscript, Washington, DC The World
  Bank.
• Hou, Jack W. (1993) Public-Private Wage
  Comparison A Case Study of Taiwan, Journal of
  Asian Economics, 4(2), p. 347-362, Fall.
• Lindauer, David and Richard H. Sabot (1983) The
  Public-Private Wage Differential in a Poor Urban
  Economy, Journal of Development Economics,
  12(1-2), p. 137-152, February-April.
• Mengistae, Taye (1998) Wage Rates and Job
  Queues Does the Public Sector Overpay in
  Ethiopia?, Working Paper Series, WPS/98-20,
  Oxford, UK Centre for the Study of African
  Economies, December.
• Psacharopoulos, George, Jorge Valenzuela and Mary
  Arends (1996) Teacher Salaries in Latin
  America a Review, Economics of Education
  Review, 15(4), p. 401-406.
• Stelcner, Morton, Jacques van de Gaag and Wim
  Vijverberg (1989) A Switching Regression Model
  of Public-Private Sector Wage Differentials in
  Peru 1985-86, Journal of Human Resources,
  24(3), p. 545-559, Summer.
• Terrell, Katherine (1993) Public-Private Wage
  Differentials in Haiti Do Public Servants Earn a
  Rent?, Journal of Development Economics, 42(2),
  p. 293-314, December.