Download full version (PPT 1680 KB) - The Institute For Fiscal Studies

skillfulbuyerUrban and Civil

Nov 16, 2013 (3 years and 8 months ago)

71 views

IFS



Pensioner poverty over the next decade:
what role for tax and benefit reform?

IFS Commentary 103


http://dx.nbdrs.com/10.1920/co.ifs.2007.0103


Mike Brewer, James Browne, Carl Emmerson, Alissa
Goodman, Ali Muriel and Gemma Tetlow

Institute for Fiscal Studies, UK

© Institute for Fiscal Studies, 2007

Policy background


2.5 million pensioners in the UK living below the relative poverty
line in 1997/98


Significant increases in benefit entitlements for pensioners since
1997


Relative poverty fallen by 0.3 million



Major reform in 2006


2007 (“Pensions White Paper”,
“Pensions Act”)


Impact on future pensioner poverty?



Forecasting pensioner poverty requires a dynamic simulation
model


current pensioners not a good guide to future pensioners

© Institute for Fiscal Studies, 2007

Technical content: micro
-
simulation
content & ageing


Sample of 11,500 adults aged 50+ from 2002


English Longitudinal Survey of Aging



Age the population by simulating life events (“Demographic simulation”)


“Transitions” based on equations estimated from various sources, sometimes
calibrated


Includes labour market behaviour, contributions to private pension, receipt of private
pension



Estimate tax liabilities and entitlement to state pension and means
-
tested
benefits (TAXBEN) (“Policy simulation”)



Examine population aged 65+ in each year from 2002 to 2017



Bespoke project, but learned from Pensim2 (Emmerson, Reed and Shephard
2004), DYNACAN (Morrison and Dussault 2000) and SAGEMOD (Zaidi and
Rake, 2001)


© Institute for Fiscal Studies, 2007

What’s coming up [redo]


Model

Results:


Demographics


Projected private incomes


Before taxes and benefits


Individual level


Projected living standards


After taxes and benefits (net income)


Family level


Poverty projections

© Institute for Fiscal Studies, 2007

Why use ELSA?


Survey of 11,500 individuals aged 50+ in England


Longitudinal, but only had 2 waves



Advantages


Rich data on pensions


Can simulate private pension income for those who have not yet retired


Demographic characteristics, incomes (including partner’s where relevant),
health & wealth



Disadvantages


Little known about household members other than spouse


Not quite best practice in recording details of income from state benefits


England, not UK



See IFS WP 05/09, IFS WP 7/12, IFS Report 67, + annual reports on
ELSA (http://www.ifs.org.uk/elsa/)

© Institute for Fiscal Studies, 2007

Demographic simulations

Mortality

Health

Disability Benefits

Labour Market

Private Incomes

2002/03 ELSA

2003/04 Simulated

2004/05 Simulated

Depends on:



Age



Sex



Social Class

Depends on:



Age



Sex



Income



Education



Marital Status

Depends on:



Age



Sex



Income



Education



Marital Status



Health



Time trend

Depends on:



Age



Sex



Income



Education



Marital Status



Health



Current benefits



Partner’s benefits

Assumes:



No real growth in
income components
except
-



2% real earnings
growth for 50
-
54
year olds

2005/06 Simulated

2017/18 Simulated

© Institute for Fiscal Studies, 2007

Demographic simulations

2003/04 Simulated

2004/05 Simulated

2017/18 Simulated


TAXBEN




Assesses tax liability &


benefit/tax credit eligibility



Different policy scenarios


can be modelled

2003/04 Net Incomes

& Poverty

2004/05 Net Incomes

& Poverty

2017/18 Net Incomes

& Poverty

© Institute for Fiscal Studies, 2007

Demographic simulations


Mortality


Use “life
-
tables” (age, sex) adjusted by social class


Health (“long
-
standing illness or disability that limits daily
activities”)


Age, sex, marital status, education, income quintile,


Labour market (full
-
time, part
-
time, no work


Age, sex, education, health (BHPS 1991
-
2005)


Can only “downsize” so calibrate to actual employment rates 2002
to 2005


Disability benefits (several)


Start: age, sex, education, health, private income (BHPS 1991
-
2005). Stop: age, sex


Calibrated to official on
-
flow and off
-
flow rates


Create 250 simulated populations, 2002/3 to 2017/18


© Institute for Fiscal Studies, 2007

Demographic simulations


We don’t model


Divorce or remarriage or fertility (remember, aged 50+)


Changes in household composition


Tenure change/geographical mobility


Moving into residential care


Assumptions on


Earnings growth (2% real a year if aged < 55) & earnings drop if go
part
-
time


Growth in non
-
earned, non
-
pension income (0% real)


Contributions to private pensions


Growth of pension wealth/fund


Saving/dissaving (none allowed, except pension wealth annuitized
when retire/downsize)


Private pension income when retire/downsize (IFS WP 05/09)



IFS



Demographic simulations: results

[add some more]



© Institute for Fiscal Studies, 2007

Projected English population 65+

0
2
4
6
8
10
12
03-04
04-05
05-06
06-07
07-08
08-09
09-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
Millions
80+
75-79
70-74
65-69
© Institute for Fiscal Studies, 2007

Projected English population 65+

0
2
4
6
8
10
12
03-04
04-05
05-06
06-07
07-08
08-09
09-10
10-11
11-12
12-13
13-14
14-15
15-16
16-17
17-18
Millions
<1926
1926-1930
1931-1935
1936-1940
1941-1945
1946-1950
1951-1955
© Institute for Fiscal Studies, 2007

Hazard of becoming “unhealthy”

0
0.1
0.2
0.3
0.4
0.5
0.6
50
55
60
65
70
75
80
85
90
95
100
Hazard rate/probability
Women
Men
% who are unhealthy is 40
-
45% amongst 65+ and little change over time

© Institute for Fiscal Studies, 2007

Hazard of stopping work

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
50
55
60
65
70
75
Hazard rate/probability
Women , from F/T
Men, from F/T
Women, from P/T
Men, from P/T
© Institute for Fiscal Studies, 2007

Simulated employment rates

0%
5%
10%
15%
20%
25%
30%
35%
2003–04
2004–05
2005–06
2006–07
2007–08
2008–09
2009–10
2010–11
2011–12
2012–13
2013–14
2014–15
2015–16
2016–17
2017–18
Employment rate
Women 65-69
Men 65–69
All Women 65+
All men 65+
IFS



Private income projections



© Institute for Fiscal Studies, 2007

Private income projections: results


Consistent growth in mean private income among
65+ population


despite

assuming zero real growth for almost all income
components


cohort effect is driving up average


Strongest growth in earnings from employment


new cohorts reaching 65 are higher earners, on average,
and
-


our model predicts rising proportions of individuals aged 65
and over remaining in work


© Institute for Fiscal Studies, 2007

0
20
40
60
80
100
120
140
160
180
200
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
£ per week
Total
Sources of income, 2003

04 to
2017

18

0
20
40
60
80
100
120
140
160
180
200
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
£ per week
Private Pensions
Savings
Employment
Other
Mean Private Income

© Institute for Fiscal Studies, 2007

Mean private income, by birth cohort,
2003

04 to 2017

18

0
50
100
150
200
250
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
£ per week
<1926
1926-1930
1931-1935
1936-1940
1941-1945
1946-1950
© Institute for Fiscal Studies, 2007

Average annual private income growth at
different points in the income scale:


2003

04 to 2017

18

0.00%
1.00%
2.00%
3.00%
4.00%
5.00%
6.00%
7.00%
8.00%
9.00%
Mean
25th
50th
75th
90th
Actual Growth (FRS) 2000/01 - 2005/06
Simulated Growth 2003/04 - 2017/18
IFS



Net income projections



© Institute for Fiscal Studies, 2007

State provision of retirement income

1.
Basic state pension

-

Flat
-
rate pension to any retired person who has paid sufficient
contributions over their working life (also credits for caring activities and
receipt of some out
-
of
-
work benefits)


Indexed to prices since 1981; value declining relative to earnings


2.
Earnings
-
Related Pension (SERPS, S2P)

-
Additional retirement income to those without access to a personal or
occupation pension (skewed towards lower earners)

-
If opt out, then pay lower payroll taxes


3.
Means
-
tested benefits (Pension Credit)


Guarantee Credit: Tops up income to an ‘appropriate’ minimum level


Savings Credit: rewards people over 65 who have saved for retirement


© Institute for Fiscal Studies, 2007

Pension credit

£0
£50
£100
£150
£200
£0
£25
£50
£75
£100
£125
£150
£175
Income other than means-tested benefits
Final income
Original non-means tested income
Final income with MIG
Final income after Pension Credit reform
Basic State Pension

© Institute for Fiscal Studies, 2007

Pensions Act 2007


Continued earnings
-
indexation of Pension Credit Guarantee, but
smaller increases in Savings Credit to stop 40% taper spreading up
income distribution



Increased basic state pension coverage from 2010


Earnings indexation of basic state pension from 2012 (aspiration)



Skew S2P towards low earners



“Personal Accounts”


compulsory retirement saving (not considered
today)


4% employee contribution (from net salary), 3% employer, 1% tax rebate


Can opt out if employer offers better alternative



Increase state pension age (not considered today)

© Institute for Fiscal Studies, 2007

Policy stability?

“This is our New Insurance Contract for pensions. This Contract will
deliver the security we all want, now and for the future.”


DWP, Green Paper, 2002

DSS, Green Paper, 1998

DWP, White Paper, 2006

“Pensions policy has to be for the long term. If we want people to
plan for the future, stability in the framework of pensions policy is a
key component.”


“These reforms set the direction for the long
-
term future of
pensions and retirement savings. They will create a system that is
coherent, comprehensive and which will stand the test of time”


Next reform: 2010?

© Institute for Fiscal Studies, 2007

Net (family) income and poverty rate


Demographic simulations passed through TAXBEN


Estimates tax liabilities & benefit entitlements


Incorporates estimates of non
-
take
-
up of means
-
tested benefits



Construct “equivalised net income”


Identical to that used in official poverty statistics, except at family level, not
household level


Income from private sources & state benefits/tax credits


Net of direct taxes including council tax, income tax and National Insurance Contributions


No imputed income for owner
-
occupiers, housing costs not deducted



Construct “poverty indicator”


Official measure is “proportion of those above pension age living in
households below 60% of median income”


We use “family” rather than “household” income, and for those aged 65 or
over; behaves in similar way to the official pensioner poverty rate

© Institute for Fiscal Studies, 2007

Simulated net family income

2003

04 to 2017

18

£250
£300
£350
£400
£450
£500
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
£ per week
Mean

Median

© Institute for Fiscal Studies, 2007

Mean net family income among those aged
65 and over, by birth cohort,

2003

04 to 2017

18


250
300
350
400
450
500
550
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
£ per week
<1926
1926-1930
1931-1935
1936-1940
1941-1945
1946-1950
© Institute for Fiscal Studies, 2007

Average annual net income growth at
different points in the income scale:


2003

04 to 2017

18

0.00%
0.50%
1.00%
1.50%
2.00%
2.50%
3.00%
3.50%
Mean
10th
25th
50th
75th
90th
Actual (HBAI) 1996–97 to 2005–06
Simulated 2003–04 to 2017–18
© Institute for Fiscal Studies, 2007

Our measure of poverty


Choose poverty line to calibrate 2005/06
poverty rate in ELSA to that in official dataset
(HBAI)


Up
-
rate poverty line at 1.8% a year for future
years


1.8% is long
-
run rate of growth of median


Roughly corresponds to 2% real growth in
earnings


© Institute for Fiscal Studies, 2007

Net income distribution: 2003

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

22%

© Institute for Fiscal Studies, 2007

2004

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2005

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2006

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2007

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2008

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2009

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2010

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2011

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2012

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2013

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2014

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2015

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2016

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

© Institute for Fiscal Studies, 2007

2017

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2003/04,
£189
Note: net incomes shown under “White Paper” policy baseline

5%

© Institute for Fiscal Studies, 2007

2017

0
100
200
300
400
500
0
100
200
300
400
500
600
700
800
900
1,000
£ per week, 2003/04 prices
Number of individuals (thousands)
Poverty
Threshold
2017,
£247
Note: net incomes shown under “White Paper” policy baseline

19%

Uprating poverty line
with long run median
income growth (1.8%
p.a) in future

© Institute for Fiscal Studies, 2007

Choosing a ‘preferred’ simulation

Model outcome depends on random draws


We repeat demographic simulation 250 times


Calculate poverty rates for each run


Choose simulation with least squared
deviation from median poverty rate (on
average) over 15 years



© Institute for Fiscal Studies, 2007

10%
12%
14%
16%
18%
20%
22%
2003–04
2004–05
2005–06
2006–07
2007–08
2008–09
2009–10
2010–11
2011–12
2012–13
2013–14
2014–15
2015–16
2016–17
2017–18
Poverty rate
Median
5th percentile
95th percentile
Preferred simulation
Choosing a ‘preferred’ simulation

© Institute for Fiscal Studies, 2007

Extent of means
-
testing before and
after Pensions Act

Proportion eligible for Pension Credit, Housing Benefit
or Council Tax Benefit

2008/09

2017/18

Previous
policy

White
Paper

Previous
policy

White
Paper

Singles

77%

77%

80%

78%

Couples

50%

50%

51%

44%

All

66%

66%

68%

64%

© Institute for Fiscal Studies, 2007

Poverty rate 1996
-
97 to 2017
-
18
under White Paper baseline

10%
12%
14%
16%
18%
20%
22%
24%
26%
28%
96–97
97–98
98–99
99–00
00–01
01–02
02–03
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
Poverty rate
actual 60%
simulated 60%
‘Calibrate’ poverty line
so that 2005
-
06 poverty
rate is same in ELSA
and ‘official’ measure

© Institute for Fiscal Studies, 2007

Components of Pensions Act

15%
16%
17%
18%
19%
20%
21%
22%
23%
24%
25%
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
Poverty rate
All White Paper reforms
Pre WP with earnings indexation
Pre WP without earnings indexation
© Institute for Fiscal Studies, 2007

Distributional impact of Pensions Act

-5%
0%
5%
10%
15%
20%
Poorest
2
3
4
5
6
7
8
9
Richest
Income decile group
% change in net income
Gain from earnings indexation of pension credit
Gain/loss from rest of White Paper
© Institute for Fiscal Studies, 2007

Why does poverty reduction stop
after 2007/08?


Because recent fall in poverty has been
exceptional


Recent increases in means
-
tested benefits for
pensioners higher than earnings growth


Median income growth since 2002/3 low by
historical standards



Our simulations are also sensitive to how fast
median income grows in the future

© Institute for Fiscal Studies, 2007

Other scenarios


Basic State Pension reforms


Full take up of means tested benefits


Council Tax


Income Tax

© Institute for Fiscal Studies, 2007

Possible basic state pension reforms
in 2017
-
18

£8.3bn
£6.9bn
£1.5bn
£0.7bn
£1.9bn
£20.0bn
-14%
-12%
-10%
-8%
-6%
-4%
-2%
0%
Earnings index BSP from
2010
Earnings index BSP from
2008
Universal BSP for those retiring
after
2012
BSP made universal
Increase BSP to PC guarantee
Universal BSP at level of PC
guarantee
Change in poverty (ppts)
© Institute for Fiscal Studies, 2007

Impact of full take
-
up of means
-
tested benefits in 2017
-
18

£3.9bn
£1.8bn
£1.2bn
£0.7bn
-14%
-12%
-10%
-8%
-6%
-4%
-2%
0%
Full take up of housing benefit
Full take up of council tax benefit
Full take up of pension credit
Full take up of all means tested
benefits and tax credits
Change in poverty (ppts)
© Institute for Fiscal Studies, 2007

Potential tax changes in 2017
-
18

-£3.0bn
-£1.2bn
£1.8bn
£4.5bn
-10%
-8%
-6%
-4%
-2%
0%
2%
4%
Abolish higher pensioner tax
allowances
High council tax rises 2008-17 (4.1%
p.a.)
Council tax nominal freeze 2008-17
Double pensioner tax allowances
Change in poverty (ppts)
© Institute for Fiscal Studies, 2007

Policy conclusions


New pensioners will have higher private incomes, but
pensioner incomes will not necessarily increase
faster than the median


Commitment to earnings indexation of pension credit
guarantee important to stop poverty rising again


Taken as a whole, the rest of Pensions Act has little
impact on poverty


Poverty could be reduced further if the government


Reduced non
-
take up of means tested benefits


Made the Basic State Pension universal or


Increased the Basic State Pension to the level of the
Pension Credit Guarantee


IFS



End



© Institute for Fiscal Studies, 2007

Comparing our poverty measure
(family level, England, 65+ only) with
the official one

0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
96–97
97–98
98–99
99–00
00–01
01–02
02–03
03–04
04–05
05–06
Poverty rate
Official 70%
Official 60%
Official 50%
IFS 70%
IFS 60%
IFS 50%
© Institute for Fiscal Studies, 2007

Poverty rates 1996
-
97 to 2017
-
18
under Pensions Act baseline

0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
96–97
97–98
98–99
99–00
00–01
01–02
02–03
03–04
04–05
05–06
06–07
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
Poverty rate
actual 50%
actual 60%
actual 70%
simulated 50%
simulated 60%
simulated 70%
© Institute for Fiscal Studies, 2007

Declining replacement rates

State pension income at age 65, median earning male

Source: Disney and Emmerson (2005)

0.0%
10.0%
20.0%
30.0%
40.0%
50.0%
1950
1960
1970
1980
1990
2000
2010
2020
2030
2040
2050
% of earnings at age 50
S
2
P addition
SERPS
Basic state pension
Year reaches age 65

© Institute for Fiscal Studies, 2007


First tier
(mandatory)


Basic state
pension


-

Savings Credit




-

Guarantee


Second

tier

(mandatory)

Approved

occupational
pensions

(DB & DC form)

Contracted in

Additional voluntary

contributions
(AVCs)

Contracted out

State
Second
Pension

(S2P)

Third tier

(voluntary)

Personal
pensions

(individual,
DC form)

‘Free
-
standing’
AVCs

Stakeholder
pension
(individual,
DC form)

“…the most complex pension
system in the world”

Pension Credit

Other

saving

© Institute for Fiscal Studies, 2007

Earnings
-
indexation of basic state
pension

15%
16%
17%
18%
19%
20%
21%
22%
07–08
08–09
09–10
10–11
11–12
12–13
13–14
14–15
15–16
16–17
17–18
Poverty rate
2012
2010
2008
2015
Never