Amounts and Accounts:
Reforming Private Pension Enrolment
Carl Emmerson and Matthew Wakefield
Institute for Fiscal Studies
© Institute for Fiscal Studies
IFS Retirement Saving Consortium
• Association of British
Insurers
• HM Revenue and Customs
• Bank of England
• Investment Management
Association
• Barclays
• Chartered Institute of
Personnel and Development
• Department for Work and
Pensions
• Financial Services Authority
© Institute for Fiscal Studies
• HM Treasury
• Pensions Regulator
• Personal Accounts Delivery
Authority
• Scottish Widows
• The Actuarial Profession
The 2012 pension reforms and
private pension holding in the UK
Carl Emmerson
© Institute for Fiscal Studies
State pension reforms
• Pension Credit Guarantee to be indexed to earnings-growth
• Increased generosity of basic state pension
• State Pension Age increased from 65 to 68 between 2024 and
2046
• Reduced generosity of Pension Credit Savings Credit
• Reduced accrual of State Second Pension for higher earners
• Impacts:
– increased income from state for many from State Pension Age
– state support less targeted on lower-income pensioners
– simpler pension system
© Institute for Fiscal Studies
2012 private pension reforms
• Employers to enrol employees automatically into a private pension
which complies with certain minimum standards
– all employees aged between 22 and State Pension Age earning more
than £5,035 (in 2006–07 earnings terms)
• Compliant schemes include:
– contracted-out defined benefit arrangements
– defined contribution schemes with certain minimum contributions
(includes new Personal Accounts)
• No increase in compulsion for employees
– free to choose to leave the scheme
– re-enrolled each time they move employer and might also be reenrolled periodically (but not more often than every 3 years)
© Institute for Fiscal Studies
Personal Accounts
• A new pension option for employers
– minimum default contributions of 5% of band earnings from employee
(1% being basic rate income tax-relief) and 3% from employer
– applies to earnings from £5,035 to £33,540 (2006–07 earnings terms)
• Employees can choose:
– to contribute less than the default minimum, but would risk losing the
employer contribution; or to contribute more than the default amount
• Employers can choose:
– to have a higher employer contribution; or to enrol employees at
higher default employee contribution rates
• Annual contribution cap of £3,600 (2005 earnings terms)
• In most cases no transfers between Personal Accounts and other
private pensions (review in 2017)
© Institute for Fiscal Studies
Reforms to boost private pension coverage
• Those currently not choosing to join an employer’s pension
scheme:
– standard economic model: slightly easier to contribute to a private
pension and slightly harder not to contribute to a private pension
– behavioural economics: some individuals might shy away from
making seeming complex decisions
© Institute for Fiscal Studies
Changing defaults can boost pension
coverage…
100%
Percentage of employees
90%
80%
70%
60%
50%
40%
30%
Before automatic enrolment
20%
After automatic enrolment
10%
0%
0
4
8
12
16
20
24
28
32
36
40
Months since joining employer
Source: This graph has been used in presentations by David Laibson (e.g. Laibson, 2008);
it draws on and is sourced to Choi et al. (2004), which in turn built on Madrian and Shea (2001).
© Institute for Fiscal Studies
44
48
…but some might contribute less
100%
Before automatic enrolment
Percentage of employees
90%
After automatic enrolment
80%
70%
65%
63%
60%
50%
40%
30%
20%
14%
10%
3%1%
4%
4%2%
11%
7%
6%5%
3%2%
4%3%
7–9%
10% 11–14% 15%
1%1%
0%
0%
1–2%
3%
4–5%
6%
Contributions as a share of earnings
© Institute for Fiscal Studies
Source: Madrian and Shea (2001).
Reforms to boost private pension coverage
• Those currently not choosing to join an employer’s pension
scheme:
– standard economic model: slightly easier to contribute to a private
pension and slightly harder not to contribute to a private pension
– behavioural economics: some individuals might shy away from
making seeming complex decisions
• Those currently not offered the chance to join an employer
scheme will face an increased incentive to join a private pension
– employer contribution of (at least) 3% of (band) earnings is contingent
on the employee not choosing to leave the scheme
– in addition to impact of change in defaults
© Institute for Fiscal Studies
Financial year
© Institute for Fiscal Studies
Source: Authors’ calculations using data from the FRS and the BHPS.
2007–08
2006–07
2005–06
2004–05
2003–04
2002–03
2001–02
British Household Panel Survey
1999–00
1998–99
1997–98
1996–97
1995–96
1994–95
1993–94
Family Resources Survey
2000–01
100
90
80
70
60
50
40
30
20
10
0
1992–93
Percentage
Private pension coverage in the UK
Detailed pension status, 2005
Offered employer's pension & joined
59
Offered employer's pension & declined,
joined Personal Pension
3
Offered employer's pension & declined,
no Personal Pension
16
Not offered employer's pension,
joined Personal Pension
5
Not offered employer's pension,
no Personal Pension
18
0
10
20
30
40
Per cent
© Institute for Fiscal Studies
Source: Authors’ calculations using data from the 2005 BHPS.
50
60
70
Pension coverage by earnings, 2005
• Median earnings among those not contributing to private pension
was £14,000 compared to £21,600 among those who did
– increase in private pension coverage to be associated with only a
small, at least in absolute terms, increase in contributions
© Institute for Fiscal Studies
100
90
80
70
60
50
40
30
20
10
0
All
Gross annual earnings
© Institute for Fiscal Studies
Source: Authors’ calculations using data from the 2005 BHPS.
50,000
45,000
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
No private pension
0
Cumulative percentage
Pension coverage by earnings, 2005
Pension coverage by liquid assets, 2005
• Those not contributing to a private pension are not more likely to
have gross debts, or greater gross debts, than those who are
© Institute for Fiscal Studies
100
90
80
70
60
50
40
30
20
10
0
All
Current gross non-mortgage debt
© Institute for Fiscal Studies
Source: Authors’ calculations using data from the 2005 BHPS.
10000
9000
8000
7000
6000
5000
4000
3000
2000
1000
No private pension
0
Cumulative percentage
Pension coverage by non-mortgage debt, 2005
Pension coverage by liquid assets, 2005
• Those not contributing to a private pension are not more likely to
have gross debts, or greater gross debts, than those who are
• But they are less likely to have savings/investments to offset debts
© Institute for Fiscal Studies
100
90
80
70
60
50
40
30
20
10
0
All
Current net liquid financial wealth
© Institute for Fiscal Studies
Source: Authors’ calculations using data from the 2005 BHPS.
40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0
-5,000
No private pension
-10,000
Cumulative percentage
Pension coverage by liquid assets, 2005
Average liquid assets, by detailed pension
status
All
1,000
All offered employer's pension
1,500
… & joined
2,800
... & refused
0
… refused & joined PP
4,000
… refused & no PP
0
Not offered employer's pension
0
… & joined PP
3,800
… & no PP
0
0
© Institute for Fiscal Studies
1,000 2,000 3,000 4,000 5,000
Median net
liquid financial
wealth
Source: Authors’ calculations using data from the 2005 BHPS.
Pension coverage by liquid assets, 2005
• Those not contributing to a private pension are not more likely to
have gross debts, or greater gross debts, than those who are
• But they are less likely to have savings/investments to offset debts
• Suggests they should be saving more but not in a private pension?
• Limited scope for those brought into private pensions to use other
savings or investments to finance contributions
– more likely that new pension saving will be new overall saving
– individuals brought into private pensions might pay off existing debts
less quickly
© Institute for Fiscal Studies
Pension coverage by other characteristics
• Within couples pension status is positively correlated
– but many not contributing to a private pension have a partner who does
© Institute for Fiscal Studies
Pension holding within couples
Partner’s pension status
Offered, accepted
Offered, refused, PP
Not offered, PP
Offered, refused, no PP
Not offered, no PP
Own pension status
Offered, accepted
Offered, refused, PP
Offered, refused, no PP
Not offered, PP
Not offered, no PP
0%
© Institute for Fiscal Studies
20%
40%
60%
Percentage
Source: Authors’ calculations using data from the 2005 BHPS.
80%
100%
Pension coverage by other characteristics
• Within couples pension status is positively correlated
– but many not contributing to a private pension have a partner who does
• Pension coverage lower among those aged 22 to 29
– 21ppt less likely to be currently contributing than those aged 40 to 49
– less likely to accept offer to join an employer’s pension scheme, and
less likely to arrange an individual private pension
– how might this change as they age?
© Institute for Fiscal Studies
Pension status by age
Offered, accepted
Offered, refused, no PP
Offered, refused, PP
Not offered, PP
Age group
Total
50 to SPA
40 to 49
30 to 39
22 to 29
0%
© Institute for Fiscal Studies
20%
40%
60%
Percentage
Source: Authors’ calculations using data from the 2005 BHPS.
80%
100%
Pension coverage by other characteristics
• Within couples pension status is positively correlated
– but many not contributing to a private pension have a partner who does
• Pension coverage lower among those aged 22 to 29
– 21ppt less likely to be currently contributing than those aged 40 to 49
– less likely to accept offer to join an employer’s pension scheme, and
less likely to arrange an individual private pension
– how might this change as they age?
• Coverage higher among public sector workers
– 26ppt more likely to be currently contributing than private sector workers
– more likely to be offered chance to join an employer’s scheme and, if
offered, more likely to accept an offer
© Institute for Fiscal Studies
Conclusions
• Reforms to enrolment and default contributions to boost private
pension coverage
• Most not currently contributing to a private pension have lower
earnings
– pounds increase in contributions will be small
• Majority not contributing to a private pension do not have positive
liquid wealth
– reshuffling small but some to repay debts less quickly?
– is a pension the best savings vehicle for them?
• Those aged 22 to 29 are less likely to be contributing than older
individuals
– a key issue is how their behaviour will change as they age
© Institute for Fiscal Studies
The market for Personal Accounts
Matthew Wakefield
© Institute for Fiscal Studies
Outline for part 2
• How much current disposable income would those not currently
contributing to a private pension forego if they were to contribute
5% of their earnings in the band £5,035 to £33,540?
• How many not currently offered an employer pension might be
brought into private pensions – perhaps often Personal Accounts
– through the change in enrolment?
• How much might these individuals build up in pensions through
minimum default contributions?
© Institute for Fiscal Studies
How much disposable income foregone?
• Take the most recent available year of data on incomes, pension
contribution status and family circumstances
– FRS, 2006/07
• Change pension contributions such that all those not in fact
contributing to a pension make individual contributions of 5% on
the on the band between £5,035 and £33,540 of earnings
• Model how this affects current disposable income, given the tax,
benefit and credit system.
• Look at effect on average across the population, and also at how
this varies across the income distribution
© Institute for Fiscal Studies
Income devoted to employee contribution
• Contribute a proportion of a band of earnings
– A key determinant of income taken is level of earnings
• Take an individual earning £33,540 with no other income
– Contributes 0.05*(33,540 – 5,035) = £1,425
– 20% is tax relief, reduction in income: 0.8*£1,425 = £1,140
– This is 3.4% of the £33,540 gross income
– Disposable income – net of income tax and NI – would have been
£24,561 w/o the pension contribution
– £1,140 is 4.6% of disposable income
• This is the biggest ‘loss’ somebody could have
– No interaction with higher-rate income tax or benefit withdrawal
– Lowest income on which one could pay 5% of the whole band
© Institute for Fiscal Studies
Income devoted to employee contribution
• Take an individual earning £7,885.50 (one-tenth of the way from
£5,035 to £33,540) with no other income
– Contributes 0.05*(7,885.5-5,035) = £142.5
– Contribution less tax relief is £114
– This is 1.4% of gross income, 1.6% of disposable income
• Take an individual earning £60,000 with no other income
– Contributes 0.05*(33,540 – 5,035) = £1,425
– 40% tax relief, reduction in income: 0.6*£1,425 = £855
– This is 1.4% of gross income, 2.1% of disposable income
• Level of earnings a major determinant of proportion of disposable
income accounted for by contribution
© Institute for Fiscal Studies
Income devoted to employee contribution
• Level of earnings a major determinant of proportion of disposable
income accounted for by contribution
• Amount of non-earned income will also matter
– Contribution a proportion of earnings
– No extra contribution for “other income”, but it is extra disposable
• Example of higher-rate taxpayer also showed tax-rate matters
– This is effective tax-rate, not just a distinction between basic- and
higher-rates
– Similar effect for those on steep taper of new tax credits
– For such an individual, at least some of pension contribution “costs”
only 41p of disposable, per pound of contribution, thus reducing the
proportion of disposable that is foregone
© Institute for Fiscal Studies
Proportion of disposable accounted for by 5% employee
contribution from those not currently contributing
Among families containing an “affected” individual
-0.5
-1.0
-1.5
-2.0
Source: IFS Tax and Benefit model, TAXBEN, using data from 2006–07 Family Resources Survey.
9
Richest
Income decile group
8
7
6
5
4
3
2
Poorest
-2.5
All
% of disposable income
0.0
Proportion of disposable accounted for by 5% employee
contribution from those not currently contributing
Across all families
-0.5
-1.0
-1.5
-2.0
Source: IFS Tax and Benefit model, TAXBEN, using data from 2006–07 Family Resources Survey.
9
Richest
Income decile group
8
7
6
5
4
3
2
Poorest
-2.5
All
% of disposable income
0.0
Proportion of disposable accounted for by 5% employee
contribution from those not currently contributing
Across all families
-0.5
-1.0
-1.5
-2.0
Source: IFS Tax and Benefit model, TAXBEN, using data from 2006–07 Family Resources Survey.
9
Richest
Income decile group
8
7
6
5
4
3
2
Poorest
-2.5
All
% of disposable income
0.0
How many might be automatically enrolled into
Personal Accounts?
• Examine one group who might have been enrolled into Personal
Accounts in the past
– Those not offered employer’s scheme
• Excludes any whose employer might choose Personal Accounts
when previously offering a different scheme
• Also supposes employers that did not offer pension scheme in
the past would now offer Personal Accounts
• Think of the group identified as a group:
– Relatively likely to be brought in to Personal Accounts
– Who would be new to being offered pension through the workplace
• Use BHPS data
– Baseline year 2005, but also look at evolution 2001-05
© Institute for Fiscal Studies
Personal Account defaultees?
% of those aged 22 to SPA
30%
25%
20%
15%
10%
14.6%
5%
0%
2001
© Institute for Fiscal Studies
2002
2003
2004
2005
Any year
Personal Account defaultees?
© Institute for Fiscal Studies
Personal Account defaultees?
% of those aged 22 to SPA
30%
25%
20%
15%
26.6%
10%
5%
12.3%
13.2%
12.7%
14.1%
14.6%
2001
2002
2003
2004
2005
0%
© Institute for Fiscal Studies
Any year
Personal Account defaultees?
% of those aged 22 to SPA
30%
25%
Without Personal Pension or Stakeholder…
With Personal Pension or Stakeholder Pension
20%
13.7%
15%
10%
11.6%
9.7%
11.2%
3.4%
3.1%
2.9%
3.0%
2002
2003
2004
2005
9.3%
9.8%
3.0%
2001
13.0%
5%
0%
© Institute for Fiscal Studies
Any year
Persistence of being in PA default group
All 5 years,
27.9%
2005 only,
21.0%
2 years,
17.8%
4 years,
17.2%
© Institute for Fiscal Studies
3 years,
16.1%
Amount of default minimum contributions
• How much would the group identified in 2005, have contributed
over the period 2001 – 2005
• Default minimum contributions are 8% of earnings between
£5,035 and £33,540 so can be computed straightforwardly
– Assessing actual contributions would be much more complex
• The amounts will be a reflection of the earnings distribution
• … and of how this and group membership shifted over the period
2001-2005, for those identified in 2005
© Institute for Fiscal Studies
Distribution of PA default minimum contributions
Cumulative percentage
100
90
80
70
60
50
40
30
20
10
0
2005
...&2004
...&2003
...&2002
...&2001
£0
£2,000
£4,000
£6,000
£8,000
£10,000
Default PA contributions among contributors in 2005
© Institute for Fiscal Studies
£12,000
Averages of PA default contributions
© Institute for Fiscal Studies
Averages of PA default contributions
© Institute for Fiscal Studies
Conclusions (1)
• Reforms to enrolment and default contributions to boost private
pension coverage
• Most not currently contributing to a private pension have lower
earnings
– Pounds increase in contributions will be small
• Majority not contributing to a private pension do not have positive
liquid wealth
– Reshuffling small but some to repay debts less quickly?
– Is a pension the best savings vehicle for them?
© Institute for Fiscal Studies
Conclusions (2)
• Default minimum employee contribution to reduce disposable
income by 0.5%
• Number brought in to Personal Accounts likely to increase
quickly
– Although also a persistent group of defaultees
– Some would have saved in a private pension without the reform
• Among those not offered an employer’s pension scheme in 2005:
– Aggregate contributions £4.2 billion from 4.7 million individuals
– Median contributions of £770
– Over 2001-05, median contributions of £2,170
© Institute for Fiscal Studies
Amounts and Accounts:
Reforming Private Pension Enrolment
Carl Emmerson and Matthew Wakefield
Institute for Fiscal Studies
© Institute for Fiscal Studies