Paying for an ageing population: Pension, savings and retirement policy in the UK

James Banks Institute for Fiscal Studies and University College London

Background • Populations are ageing across the world as a result of: – Falling fertility rates – Increasing life-expectancy

• Leading to pressures on public systems for retirement income provision

Male life expectancy and projections, UK, 1841-2041 90


80







70

60

50




40 1841 1861 1881 1901 1921 1941 1961 1981 2001 2021 2041 Source: GAD

20

Additional years life expectancy, 1841-2001 UK Males and Females aged 65

15

10

5 1841

1881

Source: Govertment Actuary’s Department

1921

1961

2001

Additional years life-expectancy Males, Age 65 Additional Years over 65

18 17 16

Japan

15

France

14

US

13

Germany UK

12 11 10 1960

1970

1980 Year

Source: OECD Health Data 2001, CD-Rom

1990

2000

Birth rates have fallen since 1950 Average number of births per female 4 3 2 1 0

Western Europe

United States

United Kingdom

1950

2.4

3.4

2.2

2000

1.5

1.9

1.6

2050

1.8

2.1

1.9

Source: UN World Population Ageing

The age structure of the UK adult population and its projection, 1981-2051 100% 80+ 70-79

75%

60-69 50% 50-59 20-49

25%

0% 1981 1991 2002 2011 2021 2031 2041 2051 Source: Government Actuary’s Department

Ageing of the EU population Old age dependency ratio Ireland Netherlands Luxembourg Denmark Finland Portugal Austria UK France Germany Spain Belgium Greece Sweden Italy

2000

0.0

10.0

20.0

30.0

40.0

Percentage

Source: EUROSTAT

50.0

60.0

2040

70.0

Overview: Lecture 1 • Why does population ageing put pressure on public finances? • How can we measure the fiscal pressure in order to decide what reform is necessary? – Are Generational Accounts the answer?

• Does the UK have less of a problem than elsewhere?

Overview: Lecture 2 • Why is there still a raging UK debate then? • A guide to current policy issues – – – – –

Are private savings adequate? What will private pensions look like? Will retirement ages adjust? … or will there be many poor pensioners? … and hence pressure on government finances

• What other factors might change?

Paying for pensions • Public pensions have traditionally been funded on a Pay as You Go (PAYG) basis • PAYG Pension system needs to balance current contributions with current payments

twL=bR Where: t = tax rate w = wage L = number of workers

b = benefit (pension) R = number of retired

• Which means

L/R = (1/t) (b/w)

• If L/R falls, then: – Either t up – Or b/w down • pensions down • or wages (productivity) up

• Note role for retirement age (links ageing population to L/R)

Projected public pension costs in Europe UK

2050 Ireland Luxembourg Netherlands Sweden Spain Portugal Belgium Denmark Finland Germany France Greece Italy Austria

2000

0

5

10

15

Percentage of national income

20

25 Source: Eurostat

It’s not just pensions • In fact, conventional government deficit accounting is fundamentally on a short term basis, a bit like Pay as You Go • A higher dependency ratio will place greater demands on health and social services as well as pensions • Generational Accounts were developed by Kotlikoff and co-authors to measure overall intertemporal liabilities in fiscal policy in the face of changing population structure

Generational accounts Government intertemporal budget constraint can be broken into four parts: PV future Net PV net government = public + taxes of + consumption wealth currently alive generations

PV net taxes of future generations

More specifically: D s =0

N t ,t − s Pt ,t − s +

∞ s =1

N t ,t + s Pt ,t + s (1 + r )

t −s

=

∞ s =0

Gt + s (1 + r )

Where: Nt,t-s = remaining net tax payment at time t for an individual of generation t-s, Pt,t-s = population of that generation alive at time t Gt+s = value of government consumption Dt = current net debt

t −s

+ Dt

Generational Accounts are a set of Nt,t-s such that the government’s intertemporal budget constraint is satisfied. Kotlikoff and co-authors argue that comparing the generation account of a current new born (age = 0 in period t) and a future newborn provides a measure of the generational imbalance in the public finances. Note: both calculations involve taxes over entire lifetimes

How to determine accounts of future generations? - Assume, apart from growth effects, that intertemporal burden is borne equally across all future generations, i.e: D s =0

N t ,t − s Pt ,t − s +

∞ s =1

N (1 + g ) Pt ,t + s (1 + r ) s

t −s

=

∞ s =0

Gt + s (1 + r )

so we have one set of Nt for each living generation and a single (average) N for an average future generation

t −s

+ Dt

Define Generational Budget Gap (GBG) and Intertemporal Budget Gap (IBG) : ∞

GBG = s =0

Gs (1 + r )

∞

IBG =

s=0

t −s

+ Dt −

D s =0

N t ,t − s Pt ,t − s +

∞ s =1

N t ,t (1 + g ) s Pt ,t + s (1 + r )

t −s

GDP G t + s (1 + r )

t−s

+ Dt −

D s=0

N t ,t − s Pt ,t − s +

∞ s =1

N t + s ,t Pt ,t + s (1 + r )

GDP

Alternatively, can calculate the immediate and permanent change in spending, transfers or in taxation that would close the IBG

t−s

Empirical methodology • Allocate each element of tax and spending for all future periods: – – – – –

Micro data for age/sex profile for each element Demographic forecasts for size of generations Forecasts for growth/wages/prices/interest rates … Policy projections for each element of tax/spending Forecast tax and spending then distributed to future generations according to age/sex profiles – Investment expenditure treated like consumption – allocated in year of expenditure

Generational Accounts for the UK Present value of remaining net tax payments (£ 000’s) 150 100

Future generations

50 0 -50 -100 0

10

20

30

40

50

60

70

80

90

Age in 1997 Source: Cardorelli et al (2000); accounts on 1997 basis

Generational Accounts around the world Percent reduction in government transfers required to achieve generational balance Italy Japan Netherlands Brazil US Germany Argentina Australia France Portugal Norway UK Belgium Denmark Canada New Zealand

-5

0

5

10

15

20

25

30

35

40

45

Source: Cardorelli et al (2000); accounts on 1997 basis

Issues : 1 • Is the ‘no change’ policy scenario the right one? • What do they really say about ‘sustainability’? • Is it appropriate to abstract from past taxes paid and benefits received? • Accounting incidence problematic in economic interpretation, e.g. implies no sharing within households (men to women, parents to children) • Also inherent in this framework is no behavioural or relative price changes (Economic Incidence) • Labour supply, Health, Education, Wages, Prices

Issues : 2 • How to estimate appropriate age profiles? • Sensitivity to predictions and forecasts: • Demographics, (including household structure) • Life-expectancy (and importantly health/morbidity • Growth, wages etc… • Why just age and sex? • Within-generation effects neglected, and may be more important in understanding redistribution (i.e. rich →poor) • Aggregation

Broad conclusions from Generational Accounts • Bring home the intertemporal nature of policy commitments and policy reform • Focus on all aspects of tax and spending, not just pensions • Basic message is that something will have to change in the future. This something could be many things - GA don’t really say what • N.B: Things may have begun to adjust already since life-cycle model is forward looking – but age-profiles may not pick it up

• Can’t get away from dependency on forecasts • Should not be used to target individual reforms • Incidence assumptions very extreme • At the very least, should be labelled ‘handle with care’

So why is the UK situation not so bad as elsewhere? • Introduction of Private Pensions as an alternative to State Pensions – Occupational/Employer (1948) – Individual/Personal Pensions (1988)

• Successive reforms to State Pensions have reduced projected generosity – Basic State Pension / Minimum Income Guarantee – State Earnings Related Pension Scheme (SERPS) / State Second Pension /

The UK Pension system, 1980 Third tier (voluntary)

Second tier (mandatory)

Unapproved private Pension plans (e.g. DC form)

Approved occupational pensions (DB)

Contracted out

First tier (mandatory)

Basic state (flat) pension

Other saving and private insurance

State Earnings Related Pension Scheme

Contracted in

Social assistance (Income-tested)

The UK Pension system, 2002 Third tier (voluntary)

Second tier (mandatory)

Additional voluntary contributions (AVCs)

Approved occupational pensions (DB & DC form)

Other saving and private insurance

‘Free-standing’ AVCs

Personal pensions (individual)

Stakeholder pension

Contracted in

Contracted out

First tier (mandatory)

Basic state (flat) pension

State Second Pension (SSP)

Minimum Income guarantee

UK Second tier pension arrangements, 1979-2001 25,000

Number of members (thousands)

Em ployer DC Hybrid

20,000

Em ployer DB 15,000

Public DB Individual DC

10,000

SERPS 5,000

0 1979 1981

1983

1985 1987

1989

1991

1993 1995

Source: Underlying data taken from Department for Work and Pensions, 2004

1997

1999 2001

Pension coverage, by earnings Occupational pension

Both

Personal pension

No private pension

100

Percentage

80 60 40 20

st ri c he

9

8

7

6

5

4

3

2

po or es t

Earnings

Ze

ro

ea rn

in gs

0

Source: British Household Panel Survey, 1992 to 1998

Reforms to the UK state pension system, 1980 to present day a. Reductions in generosity of the state system: Social Security Act 1980

State pension payments to be increased by growth in prices instead of the greater of growth in prices or earnings.

Social Security Act 1986

Entitlement to SERPS to be calculated on the basis of earnings in 49 years (16 to 64) rather than across the best 20 years for those reaching the state pension age after April 1999. The accrual factor on SERPS to be reduced from 25% of relevant earnings to 20% of relevant earnings. This is to be phased in on contributions between April 1999 and March 2008, although accrued entitlement from before April 1988 is protected. Surviving partners of those who die after April 2000 to inherit 50 percent of their spouses state pension instead of 100 percent. This change was later put back to October 2002 after the Department for Social Security failed to correctly inform individuals of this change.

Social Security Act 1995

State pension age for women to be increased from 60 to 65 gradually between 2010 and 2020 Technical change made to the formula used to calculate SERPS entitlement. This reduced the generosity of SERPS to those reaching the state pension age after April 1999.

b. Increases in generosity of the state system: Child Support, Pensions and Social Security Act 2000

The State Second Pension to replace SERPS from April 2002 onwards. This is more generous to lower earners and to some individuals not in paid employment but with caring responsibilities.

State pension entitlements, by year of retirement

35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0%

Bas ic State Pens ion SERPS + Second State Pens ion

SERPS Pens ion Credit Guarantee

Notes: Calculations for an individual with full contribution history at average earnings throughout working life. Assumes annual real earnings growth of 2%.

2060

2055

2050

2045

2040

2035

2030

2025

2020

2015

2010

2005

2000

1995

1990

1985

1980

1975

1970

1965

1960

1955

0.0% 1950

Percentage of average earnings at retirement

40.0%

Reduction in UK pension generosity: 1 Earnings indexation of benefits from 1980 onwards

4.8 6.3

Earnings indexation from 1999–2000 onwards

3.6 4.8

3.5

Basic state pension

2000

3.0

0

1

2 3 4 Percentage of national income

2030 5

6

7

Source: Calculations from Emmerson and Johnson (2002)

Reduction in UK pension generosity: 2 SERPS with original formula, retirement ages and inheritance rights

0.5 2.9

SERPS after the 1995 Act

0.5

State Second Pension after the 1999 Act

0.5

0.8

2000 1.1

0

1

2030

2 3 4 Percentage of national income

5

6

7

Source: Calculations from Emmerson and Johnson (2002)

Result: More retirement income from private sources already Income sources, by income group

800

Other income Earnings Investment income

600

Occupational pension Benefit income

400 200

2001-02

st R ic he

4

3

2

t or es Po

st R ic he

1979

4

3

2

or es

t

0

Po

£ per week, 2001-02 prices

1,000

Pensioner couples only

Result: More income will come from means tested benefits in the future

Percentage of total

100% MIG / pension credit

75%

SERPS / State Second Pension

50%

25%

Basic state pension

0% 2001/2

2011/2

2021/2

2031/2

2041/2

2051/2 Source: DWP

Overview: Lecture 2 • Why is there still a raging UK debate then? • A guide to current policy issues – – – – –

Are private savings adequate? What will private pensions look like? Will retirement ages adjust? … or will there be many poor pensioners? … and hence pressure on government finances

• What other factors might change?