World Bank
Core Course on Pensions
Washington, May 2015
Civil-Service
Agenda
Institutional arrangements for public-sector workers’ pensions
Demographic pressures on finances
Flexibility and portability of civil-service pensions
Origins
Civil-service pension schemes usually set up before national programmes
independence of civil servants
make working for the public sector attractive
shift the cost of remunerating civil servants into the future
Separate schemes then often persisted after national schemes established: ‘dualism’
Institutional arrangements
around the world
0 25 50 75 100
Eastern Europe/
Central Asia
Latin America/
Caribbean
OECD
East Asia
Middle East/
North Africa
Africa
South Asia
Separate
Integrated
7
32 7
12 15
27
13 12
9 6
7 4
Institutional arrangements
Fully integrated
Institutionally separate with similar benefits
Fully integrated with top-up arrangements
Partially integrated with top-up scheme
Entirely separate institutions and benefits
Chile Czech Republic Estonia Hungary Mexico Poland
Slovak Republic
Denmark Finland Iceland Israel Netherlands
Australia Canada Ireland Italy Japan New Zealand Norway Slovenia Spain Sweden Switzerland United States
United Kingdom Austria Belgium France Germany Greece Korea Luxembourg Portugal Turkey
Institutional arrangements
Separate
Angola
Benin
Burundi
Cameroon
Congo, DR
Congo, R
Cote d’Ivoire
Gambia
Guinea
Guinea-Bissau
Kenya
Madagascar
Malawi
Mali
Mauritania
Mozambique
Niger
Senegal
Sudan
Tanzania
Togo
Uganda
Zimbabwe
Partially integrated 1
Botswana
Lesotho
Mauritius
Namibia
South Africa
Swaziland
Integrated
Cape Verde
Central African Republic
Chad
Ethiopia
Ghana
Nigeria
Rwanda
Sao Tome e Principe
Seychelles
Sierra Leone
Zambia
Partially integrated 2
Liberia
No private-sector
scheme
Eritrea
South Sudan
Dualism
Arguments against
Integration gives civil servants direct, personal interest in the plan being well managed
Economies of scale
Mobility and portability
Equity
Transparency
Long-term goal should therefore probably be integration of civil-service and national pension plans
Economies of scale
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.25
2.50
2.75
3.00
3.25
3.50
3.75
4.00
Relative cost per beneficiary
Number of beneficiaries
Central-government employment
per cent of total population
early 1980s early 1990s
Africa 1.8 1.1Asia 2.6 1.1Latin America 2.4 1.5All developing countries 2.2 1.2
OECD 2.9 1.9
10
20
30
40
50
1995 2000 2005 2009
Australia
Estonia
Korea
Iceland
United States
Canada
Israel
FinlandNew Zealand
Austria
Portugal
United Kingdom
Percentage aged 50 and over
Belgium
10
20
30
40
50
1995 2000 2005 2009
Chile
Japan
Switzerland
Ireland
Netherlands
Italy
Germany
Denmark
Sweden
Norway
Greece
France
Percentage aged 50 and over
Mexico
Ageing central-government workforce
Example: Egypt
Example: Morocco
0
1
2
3
4
5
20 25 30 35 40 45 50 55 60
CMR
CNSS
RCAR
Labour-force demographics:
central government vs population
Central government employees Total labour force
0 10 20 30 40 50
Ireland
Canada
Netherlands
Finland
Norway
Israel
Greece
Slovak Republic
Denmark
United States
Germany
Belgium
Sweden
Iceland
Italy
Percentage aged 50 and over0 10 20 30 40 50
Korea
Chile
Japan
Estonia
Australia
Mexico
Poland
Slovenia
New Zealand
France
United Kingdom
Switzerland
Portugal
Hungary
Austria
Percentage aged 50 and over
Reform options 1
‘Parametric’ reforms to defined benefit plans reduce replacement rate
index pensions in payment to prices rather thancivil-service earnings
introduce/increase member contributions
raise pensionable age
extend averaging periods for ‘final’ salary
‘Systemic’ reforms introduce new system for new civil servants with some element
of pre-funding of obligations
Any reform must take account of all aspects of civil-service terms and conditions
Reform options 2
Increasing contributions: employer contributions are just re-labelling, unlike national systems
employee contributions may have an effect on wages or productivity
Increasing pension age: Civil service schemes are ‘closed’ systems
so increasing retirement age has different effects than it does in national schemes: labour supply effect in national schemes
Increase in retirement age cuts duration of benefit payments, but
without downward adjustment of accrual rates to compensate, benefit values increase
people might retire on higher pay if earnings continue to grow with age
affects both pay and pension bills
Lump-of-labour fallacy
across countries
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80
0.1
0.2
0.3
0.4
0.5
0.6
AUT
BEL
CZE
DEU
DNK
ESP
FIN
FRA
GBR
GRC
HUN
IRL
ITA
LUX
NLD
POL
PRT
SVK
SWE
Employment rate, 20-24 year olds
Employment rate, 60-64 year olds
Lump-of-labour fallacy
over time
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
New Zealand
1992
1995
2000
2001
2004
2010-12
France
19831985
1990
1995 2000
2005
2012
Employment rate, 20-24 year olds
Employment rate, 60-64 year olds
Individual-level model: Morocco
Baseline
Contribution rate 20%Contribution years 33Lifetime contributions (x annual salary)
6.6
Accrual rate 2.5%Gross replacement rate 82.5%Net replacement rate(average earner)
106.7%
Pension age 60Indexation half pricesLifetime benefits(x annual salary)
13.2
Individual-level model
Baseline
Contribution rate 20%Contribution years 33Lifetime contributions (x annual salary)
6.6
Accrual rate 2.5%Gross replacement rate 82.5%Net replacement rate(average earner)
106.7%
Pension age 60Indexation half pricesLifetime benefits(x annual salary)
13.2
Benefit/cost ratio 2.0Equilibrium contribution rate 40%Sustainable replacement rate 41%
Individual-level model
Baseline Age 65: same accrual
Contribution rate 20% 20%Contribution years 33 38Lifetime contributions (x annual salary)
6.6 7.6
Accrual rate 2.5% 2.5%Gross replacement rate 82.5% 95.0%Net replacement rate(average earner)
106.7% 122.3%
Pension age 60 65Indexation half prices pricesLifetime benefits(x annual salary)
13.2 11.2
Benefit/cost ratio 2.0 1.5Equilibrium contribution rate 40% 38%Sustainable replacement rate 41% 50%
Individual-level model
Baseline Age 65: same accrual
Age 65: loweraccrual
Contribution rate 20% 20% 20%Contribution years 33 38 38Lifetime contributions (x annual salary)
6.6 7.6 7.6
Accrual rate 2.5% 2.5% 2.2%Gross replacement rate 82.5% 95.0% 82.5%Net replacement rate(average earner)
106.7% 122.3% 106.7%
Pension age 60 65 65Indexation half prices prices pricesLifetime benefits(x annual salary)
13.2 11.2 9.7
Benefit/cost ratio 2.0 1.5 1.3Equilibrium contribution rate 40% 38% 26%Sustainable replacement rate 41% 50% 64%
Pension possibilities
Three key variables:
accrual rate
pension eligibility age
contribution rate
Look at the sustainable combinations
Pension possibilities
0
5
10
15
20
25
30
35
40
45
50 55 60 65 700
0.25
0.5
0.75
1
1.25
1.5
50 55 60 65 70
0
0.5
1
1.5
2
2.5
0 5 10 15 20 25 30 35
Contribution
rate (%)
Pensionable age
Accrual rate: 2% Accrual
rate (%)
Pensionable age
Contribution rate: 17%
Accrual
rate (%)Pensionable age: 60
Contribution
rate (%)
Pension possibilities:
replacement rates
0
10
20
30
40
50
60
70
80
50 55 60 65 70
0
10
20
30
40
50
60
70
80
0 5 10 15 20 25 30 35
Pensionable age
Contribution rate: 17%
Replacement
rate (%)
Replacement
rate (%)
Contribution rate (%)
Pensionable age: 60
Flexibility and portability
Civil service schemes are inflexible: ill designed to deal with people without full careers
But flexible schemes are increasingly important
‘revolving doors’: cross-fertilisation between public and private sectors
transfer of employees due to privatisation or contracting out
Penalties to moving jobs
Vesting periods: when individual qualifies for a pension
<1 year in Finland, Netherlands, Sweden, Switzerland, UK
5 years in Belgium, Germany, Ireland, Italy
15yrs in Austria, France, Spain, Mauritius, Senegal
people can leave with nothing
Treatment of ‘early leavers’: what happens to the benefit between leaving the job and claiming the pension?
full transferability (Finland, Netherlands, Sweden)moves to occupational plan with same benefits in private sector
full preservation (France)accrued rights uprated in line with civil-service earnings
In other countries, a pension cost to moving jobs
Example: Mauritius
age25 30 35 40 45 50 55 60
Value of accrued
pension
Example: Mauritius
age25 30 35 40 45 50 55 60
Value of accrued
pension
Staying to retirement:
1/50th of final salary
Example: Mauritius
age25 30 35 40 45 50 55 60
Value of accrued
pension
Lump sumDeferred pension:
1/50th of current salary
Staying to retirement:
1/50th of final salary
Example: Mauritius
age25 30 35 40 45 50 55 60
0
2
4
6
8 Cost of leaving,
proportion of
earnings
Example: UK
age25 30 35 40 45 50 55 60
0
.5
1
1.5Cost of leaving,
proportion of
earnings
Early leaver’s benefit depends on
earnings uprated in line with prices:
‘partial preservation’
Germany
age25 30 35 40 45 50 55 60
0
2
4
6Cost of leaving,
proportion of
earnings
Early leaver is retrospectively transferred
to national scheme with lower benefits
Reforms to improve portability
Shorten vesting periods
Preserve pension rights of early leavers
Extend averaging period for ‘final salary’
career average uprating eliminates the mobility problem
also deals with problems of incentives for abuse
but requires improvements in record-keeping
Introduce a defined contribution scheme
fully portable
Conclusions
Reform of civil-service pension schemes is important in low- and middle-income countries
often, larger expenditure than national schemes
crowds out important social programmes
Many options to put civil-service pension schemes on a sustainable footing
Structural issues as important as fiscal ones
single national scheme would be more administratively efficient, equitable and increase labour-market flexibility
equity and efficiency also improved by longer averaging periods for earnings, shorter vesting periods, preservation for early leavers, DC option