Analysis of Future Pension Incomes 2025
Published 21 July 2025
This analysis provides new estimates of the number and proportion of working age individuals, aged 22 to State Pension age, who are undersaving for their retirement. This analysis also provides case studies which estimate the size of individuals’ defined contribution (DC) pots at retirement and estimates of retirement income for future retirees. This builds on the work from the previous publication in 2023 Analysis of Future Pension Incomes.
The report seeks to answer:
- What is the current level of undersaving for retirement of working-age individuals?
- Which groups are undersaving?
- What is the depth of undersaving?
- How do retirement outcomes change, for example through retiring early?
- What do retirement incomes look like for future retirees?
It is important to state any assessment of the adequacy of future pension income is complex and involves subjective judgement, as the level of income an individual may need, or want, in retirement is dependent on their own individual preferences and expectations. Adequacy measures are also affected by uncertain future economic circumstances. Therefore, when measuring the level of undersaving in the current working age population it is important to be aware estimates can change based on the methodology, the measure used, and the assumptions made. This report therefore considers several adequacy metrics for measuring undersaving and explains the key assumptions used.
1. Main Headlines
This analysis explores the levels of undersaving for retirement. This can be measured in many ways; two key approaches are explored in this publication:
- Target Replacement Rate (TRR) – a percentage of pre-retirement earnings (the average over ages 50 to State Pension Age) an individual would need to replace to meet an adequate income in retirement, as set out in the Turner Commission (2004)
- Expenditure-based level – a level of income that might be deemed adequate – for example, the Pensions UK - Retirement Living Standards
Pension saving in Great Britain has changed significantly over the past decade due to the introduction of Automatic Enrolment (AE). As of 2023, workplace pension participation rate for eligible employees in Great Britain is 88% (20.8 million)[footnote 1]. The overall participation rate of all employees into a workplace pension in Great Britain was 80% (22.3 million).
The analysis shows there is still a large proportion of the population undersaving for retirement. Key findings include:
- 4-in-10 (43%) working-age people (equivalent to 14.6 million) are undersaving for retirement when measured against their TRRs Before Housing Costs (BHC). This is calculated using the assumption individuals will convert the full value of their DC pension pot into an annuity
- higher earners are more likely to be undersaving relative to TRRs. Around 1-in-8 (13%) of those in the lowest earning bands (less than £15,900 pre-retirement earnings per year)[footnote 2] are undersaving compared to 1-in-2 (48%) in the top earnings band (more than £67,000 per year). This is driven by the State Pension providing a greater proportion of a low earner’s target income
- of the 14.6 million people undersaving, 47% (6.9 million) meet more than 80% of their target income
- 1-in-8 (13%) of working age people are undersaving for retirement when measured against the PLSA Minimum RLS (an alternative way of measuring adequacy of retirement income). This increases to 3-in-4 (73%) and 9-in-10 (91%) for the Moderate and Comfortable RLS respectively
- low earners are more likely to be undersaving when measured against the PLSA RLS. Around 1-in-2 (47%) of all people in the lowest earnings band are projected to not meet the PLSA minimum RLS, compared to only 1-in-25 (4%) in the highest earnings band
- a male born in 2002, with median earnings and a typical earnings profile who is automatically enrolled from age 22 and works continuously to State Pension age (SPa) can expect an annual income of £24,000 when combining State and private pensions. This provides a TRR of 64%; below the level recommended by the Turner commission (67%)
- an individual born in 2002, earning National Living Wage, who is automatically enrolled from age 22 and works to SPa can expect an annual income of £20,000 when combining State and private pensions. This provides a TRR of 89%, which is above the level recommended by the Turner commission (70%). This is because for lower earners, the State Pension provides a large majority of their pre-retirement income and therefore a large proportion of their TRR
- for individuals who do not work or save their whole working-age life, TRRs are harder to achieve. For example, a median earning female who takes a 5-year career break and works part-time for 5 years following this break and retires at age 60 achieves a TRR of 52%; below the level recommended by the Turner commission (67%)
- total pension incomes for future retirees are projected to decline over the next 20 years compared to previous retirees. This is due to Defined Benefit (DB) income and coverage declining before people have realised a full career of auto-enrolment savings in DC. The value and proportion of total income from State Pension will continue to rise over the projected period
- total incomes for individuals retiring in 2050 could be, on average, only 1% higher than those retiring in 2025, whereas private pension income for individuals retiring in 2050 could be 8% lower than those retiring in 2025
2. What you need to know
Background
1. To support the monitoring of Automatic Enrolment (AE), DWP have published analysis of future pension incomes several times over the last decade, most recently in March 2023. The workplace pension participation rate for AE eligible employees in Great Britain was at 88% (20.8 million) in 2023. This is the same participation level as previously but an increase in the total number of eligible employees saving into a pension (0.4million). The overall participation rate for all employees in Great Britain was 80% (22.3 million).[footnote 3]
2. In previous analysis DWP have used TRRs (defined below) and the PLSA RLS to measure of adequacy of future pension incomes. There are multiple new measures of undersaving which have been developed over the last decade. This analysis presents undersaving estimates using these two widely used measures, TRRs and the PLSA RLS, to estimate who is and is not able to meet these living standards in retirement.
3. The two measures are defined as:
1. Target Replacement Rates
- The first report of the Pensions Commission published in 2004 discussed the concept of an ‘adequate’ pension income and what role the state should take to ensure pensioners reach an adequate income. The report assumed that a pensioner would want to maintain the same standard of living in retirement that they had during their working life, therefore they would need to maintain a level of pension income related to their pre-retirement earnings.
- Changing patterns in consumption as people get older suggests their pension income can be somewhat lower than their pre-retirement earnings to maintain a similar standard of living in retirement. TRRs refers to the percentage of their pre-retirement income an individual would need to replace to meet an ‘adequate’ income. For a median earner, this is 67% of their pre-retirement earnings. The Pension Commission suggested those with lower earnings need to achieve a higher replacement rate to maintain their living standards (80% for those in the lowest earnings band) than higher earners (50% for those in the highest earnings band).
- In the current analysis, pre-retirement earnings are calculated as the average gross (pre-tax) earnings from age 50 until State Pension age (SPa) over the period individuals are employed or self-employed. Pension income is gross (pre-tax) and is calculated as the income from SPa onwards averaged over the whole of each individual’s retirement. Full details of the methodology can be found in section 9.
- This analysis considers pre-retirement earnings only, which is consistent with the foundations of TRRs and previous analytical reports. It is important to consider the role of benefits, such as Universal Credit, in supporting people in working life. Individuals may have different lived experiences when considering total incomes pre and post-retirement, for example having previously had earnings and in-work benefit support.
- Pension income includes state and private pension income and excludes means tested benefits such as Pension Credit. While financial wealth is not modelled directly, an amount is imputed and then annuitised based on Wealth and Assets Survey data 2018 to 2020.
2. PLSA Retirement Living Standards (RLS)
- The PLSA developed Retirement Living Standards to help savers plan for their desired retirement by setting out levels of expenditure for three standards of living in retirement described as ’minimum’, ’moderate’, and ’comfortable’. These are calculated for both individuals (single person households) and couples (two person households), with variants to reflect higher living costs in London.
- The RLS build on the Joseph Rowntree Foundation’s approach to design the Minimum Income Standards, which estimate the retirement income that different households require to reach a minimally acceptable standard of living in the UK. The research, carried out by Loughborough University, uses focus groups with members of the public aged 50 or over to establish a public consensus on the goods and services needed by a range of household types to achieve a set of retirement living standards. The standards are then calculated by determining the expenditure required to consume the goods and services required at each standard.
- Although the PLSA RLS have separate standards for London and outside of London, the estimates are modelled at a national level therefore all individuals are assessed against the outside of London standards. See section 9 for further information about the methodology.
Modelling future pension incomes
Pensim3
4. DWP has a long-term model of pensioner incomes (Pensim3). Pensim3 is a dynamic microsimulation model forecasting state and private pension contributions and incomes for the British population until the year 2100. As with any large model, many assumptions are made which are subject to change with model development. The modelling improvements are ongoing and DWP analysts are working on a next generation long term model which will eventually replace Pensim3. The modelling is confined to known and predictable changes from policies rather than any possible behavioural impacts such as random reduced pension saving.
5. The current analysis includes the current AE eligible group (those aged 22 to SPa) and legislated SPa changes are accounted for across individuals. These estimates are all based on the State Pension continuing to be uprated by the triple lock.
6. Pensions ‘freedoms’ enabled people to access their pensions in more ways. This includes drawing down a pension income or taking partial or full cash withdrawals. For the purposes of this analysis, individuals with DC pensions are assumed to take their entire pension pot as an annuity (annual income for life). This is referred to as the ‘all income’ scenario. Whilst there has been a steep decline in the sale of annuities since 2014, the long-term decumulation landscape is uncertain and using the annuity approach creates a consistent, predictable income measure across all retirees for the analysis.
7. Two scenarios are presented:
1) “All Income” - This reflects the entire pension pot being converted into an annuity (annual income). This is the preferred metric used throughout the report unless otherwise stated.
2) “Income after lump sum” - This reflects an individual taking the 25% tax-free allowance as a lump sum with the remainder of the pension pot then being converted into an annuity. This is presented just for comparison as DWP’s research[footnote 4]
8. All estimates are presented at the individual level. In the analysis, pension incomes are combined at the benefit unit level (a benefit unit is defined to be a single adult or a married or cohabiting couple and any dependent children) and equivalised to adjust a couple’s income (section 9). This enables comparisons across all individuals against the same benchmark.
9. When measuring undersaving using TRRs, Before Housing Cost (BHC) and After Housing Cost (AHC) income is used. DWP previously made this adjustment to the original methodology to reflect the different costs faced by owner occupiers and renters. The Pensions Commission benchmarks are adjusted to reflect the impact of different housing costs between working life and retirement (section 9 covers). Unless otherwise stated, this report focuses on the BHC measure. For the full AHC results see the published data spreadsheet.
10. The PLSA RLS do not include any housing costs, reflecting that most pensioners currently own their home. As a result, when measuring undersaving using PLSA’s RLS the analysis compares pension income AHC.
Changes from the previous published 2023 undersaving estimate
11. DWP’s long-term models are continuously improved reflecting new data, updated assumptions and model development. As a result, the latest version of the model is an update from the previous publication. Any long-term model is highly sensitive to assumptions. The changes made have resulted in the estimate of undersaving (measured by TRRs BHC) changing from 38% to 43%. This has been driven by modelling and data changes rather than necessarily changes to the population’s saving behaviours. Due to the changes outlined the following analysis cannot be directly compared to DWP’s previous undersaving estimates. Key modelling changes are listed in the annex.
IPEN
12. IPEN is a micro-simulation model used to create case studies of different individual types, such as median earners or National Living Wage earners, and different policy scenarios. Changes in both state and private pension rules can be made in the model to see the impacts on people with different working patterns, employment history, pension contributions, ages, relationship status and sex. Outputs include private pension pot size, and pension income from both state and private pension in both future and current price terms. The pension income is assumed to be annuitised. State and private pension income is compared to the individual’s pre-retirement income to determine a TRR and give an adequacy verdict. This model allows analysis of individual outcomes, whereas Pensim3 is effective for analysis of groups or the aggregate population. This is important because it enables the analysis to look at the details for individuals with fundamentally different savings and work profiles.
3. Undersaving analysis
13. Figure 1 shows the proportion of the working-age population (22-SPa) projected not to meet their TRR BHC and AHC, as well as the proportion of the population projected to not meet each PLSA RLS.
Figure 1: The proportion of working-age people projected to not meet their TRR or PLSA RLS
Table 1: Volumes of working-age people saving enough or undersaving by each measure of adequacy, and total number of individuals included in analysis for each measure[footnote 5]
Measure | Saving Enough | Undersaving | All[footnote 6] |
---|---|---|---|
TRR BHC | 19,020,000 | 14,590,000 | 33,610,000 |
TRR AHC | 18,230,000 | 15,380,000 | 33,610,000 |
PLSA Minimum | 30,200000 | 4,610,000 | 34,810,000 |
PLSA Moderate | 9,450,000 | 25,360,000 | 34,810,000 |
PLSA Comfortable | 3,300,000 | 31,510,000 | 34,810,000 |
14. In the ‘all income’ scenario BHC, it is estimated 43% of the working age population are undersaving for retirement (14.6 million people). This is slightly higher, 46% (15.4 million people) AHC. In the ‘lump sum’ scenario, the proportion undersaving is about 5 percentage points higher, 48% (16.2 million people) BHC.
15. When measuring adequacy against the PLSA RLS, 13% (4.6 million people) of the working-age population are projected to have pension income that falls below the PLSA minimum RLS. Almost 3-in-4 (73%) of the population are projected to have a pension income below the PLSA moderate RLS. The PLSA estimates of moderate were significantly revised in their February 2024 update[footnote 7], which means a greater number of individuals are unlikely to meet this threshold. A large majority, 91% (31.5 million) are projected to have a pension income below the PLSA comfortable RLS.
16. The Pensions Policy Institute[footnote 8] have previously conducted similar analysis, using the Wealth and Assets survey to project how many people will fall below the PLSA RLS and TRRs. Within this analysis, the additional capital scenario is most like the analysis DWP have completed using Pensim3. PPI estimate that 51% of working-age households will miss their TRR, whereas this analysis estimates 4-in-10 will fall below their TRR. It is important to note that the methodology is different across the two projects, and that PPI’s project uses an older version of the PLSA RLS. Despite this, the results are largely consistent with the analysis completed in this publication in relation to the scale of undersaving.
By pre-retirement earnings
17. The pre-retirement earnings bands are based on the bands in the 2004 Turner Commission in 2023 prices, for more information see section 9. Figure 2[footnote 9] shows the proportion of people in each pre-retirement earnings band projected to not meet their TRR. Around 13% in the lowest earnings band (less than £15,900 gross pre-retirement earnings per year) are undersaving, compared to 48% in the top earnings band (more than £67,000 gross pre-retirement earnings per year). Earners in the higher income bands are more likely to be undersaving due to a higher reliance on their private pension to contribute to their target pension income.
Figure 2: The proportion of working-age people projected to not meet their TRR in each pre-retirement earnings band, before housing costs and after housing costs
18. Figure 3 shows the proportion of working age people projected to not meet the PLSA Minimum, Moderate and Comfortable RLS in each pre-retirement earnings band. This shows the opposite pattern to replacement rates, where those in the lower earnings bands are more likely to fall short of the minimum RLS. Around 47% of people in the lowest earnings band are estimated to fall below the PLSA Minimum RLS, compared with only 4% in the highest earnings band.
19. Measuring adequacy using TRRs suggests that higher earners are more likely to have inadequate income, however measuring against PLSA RLS suggests that low earners are more likely to have inadequate income. Given the two measures tell different stories, as TRRs are personalised to pre-retirement income and the PLSA measures are a fixed threshold for everyone, it is important to consider multiple measures when trying to understand the adequacy of future pension incomes.
Figure 3: The proportion of working-age people projected to not meet each PLSA RLS in each pre-retirement earnings band, after housing costs
4. Undersaving by individual characteristics
Cohort
20. Table 2 shows the proportion of people projected to not meet their TRR or PLSA RLS by cohort; defined as the decade of reaching SPa.
21. Those reaching SPa in the 2030s and 2040s are more likely to be undersaving, with 45% and 46% undersaving respectively, compared with 40% of those retiring in 2060s. These older cohorts will not have benefitted from a full career auto-enrolled in a workplace pension and are less likely to be in a DB scheme compared to those approaching retirement in the 2020s.
22. Over time, more people are projected to meet the PLSA minimum RLS. This is partially driven by the assumption of the State Pension being uprated by the triple lock and more individuals having additional years of workplace pension saving due to AE. Although more people are also likely to meet the PLSA Moderate RLS by the 2060s, there is still a large proportion (72%) of people who are not projected to meet it.
Table 2: The proportion of working age people projected to not meet their TRR or PLSA RLS by decade of reaching SPa
Cohort | TRR BHC | TRR AHC | PLSA Minimum | PLSA Moderate | PLSA Comfortable |
---|---|---|---|---|---|
2020s | 43% | 44% | 17% | 73% | 90% |
2030s | 45% | 47% | 15% | 71% | 89% |
2040s | 46% | 49% | 15% | 76% | 91% |
2050s | 43% | 45% | 11% | 73% | 91% |
2060s | 40% | 43% | 10% | 72% | 91% |
All Individuals | 43% | 46% | 13% | 73% | 91% |
Partner status
23. There is a 2-percentage point difference between singles and couples meeting their TRR BHC, with 45% of singles not meeting their TRR BHC and 43% of couples. The difference is larger, 3-percentage points, when considering TRR AHC, as couples can take advantage of sharing housing costs.
24.When measuring undersaving against the PLSA RLS, 21% of single pensioners are projected to not meet the PLSA minimum RLS compared with only 9% of pensioners in a couple. This pattern is replicated on other PLSA measures, with a clear gap between singles and couples.
Table 3: The proportion of working-age people projected to not meet their TRR or PLSA RLS by partnership status
Partnership Status | TRR BHC | TRR AHC | PLSA Minimum | PLSA Moderate | PLSA Comfortable |
---|---|---|---|---|---|
Single | 45% | 48% | 21% | 82% | 93% |
Couple | 43% | 45% | 9% | 68% | 89% |
All Individuals | 43% | 46% | 13% | 73% | 91% |
Housing Tenure
25. Table 4 shows that when housing costs are considered, renters [footnote 10] are 8-percentage points more likely to not meet their TRR than non-renters (owner occupiers). 27% of renters are projected to not meet the PLSA minimum RLS compared with 9% of non-renters. This highlights the role of housing costs in the adequacy of pension incomes and the substantially lower costs faced by homeowners. The modelling currently assumes home ownership continues at similar levels in retirement to existing levels (around 80% homeownership). If this was to be lower in the future, as analysis by the Pensions Policy Institute suggests[footnote 11], and more people in retirement had rental payments, it is likely undersaving (AHC) would increase.
Table 4: The proportion of working-age people projected to not meet their TRR or PLSA RLS by housing tenure.
Housing Tenure | TRR BHC | TRR AHC | PLSA Minimum | PLSA Moderate | PLSA Comfortable |
---|---|---|---|---|---|
Non-renter | 43% | 44% | 9% | 69% | 89% |
Renter | 46% | 52% | 27% | 86% | 95% |
All Individuals | 43% | 46% | 13% | 73% | 91% |
Private Pension Provision
26. The estimates of undersaving include people aged 22 to SPa. Not all this population will be saving into a private pension. This could be due to several reasons such as, but not limited to, a choice not to save, the affordability of saving in later life or reliance on a partner’s pension saving.
27. In this analysis, pension income for couples is combined and then equivalised (see section 9). To assess the impact of pension type on undersaving, the analysis considers pension type within the benefit unit. For single pensioners, this is the equivalent of whether they have DB, DC, both or neither pension types over their working life. For pensioner couples, the analysis considers whether either individual has these pension types over their working life.
28. This shows:
- 72% (equivalent to 0.5 million) of people without a private pension are projected to not meet their TRR and 81% (equivalent to 1.0 million) are projected to not meet the PLSA minimum RLS
- 36% (equivalent to 7.9 million) of people with both DB and DC pensions are projected to not meet their TRR BHC and only 5% to not meet the PLSA minimum RLS
- those with either DB or DC income or a combination of the two are less likely to be undersaving than those with no private pension. Being without a private pension has the highest undersaving rates seen in this analysis across all the characteristics
Table 5: The proportion of people not meeting their TRR or PLSA RLS by private pension type
Private Pension Type | TRR BHC | TRR AHC | PLSA Minimum | PLSA Moderate | PLSA Comfortable |
---|---|---|---|---|---|
No private pension | 72% | 73% | 81% | 100% | 100% |
DB only | 36% | 38% | 17% | 70% | 87% |
DC only | 59% | 61% | 24% | 88% | 96% |
DB and DC | 36% | 39% | 5% | 65% | 88% |
All Individuals | 43% | 46% | 13% | 73% | 91% |
29. Figure 4 shows the change in private pension type over time. Amongst those due to retire in the 2020s, 16% of individuals live in households that have DC pension income only, by the 2060s this rises to 38%. Across all cohorts, most individuals are from households that have DB and DC pension income. This could be due to individuals saving in multiple schemes or being part of a couple where at least one member has participated in a DB pension to be included in this category. The proportion declines from 71% in the 2020s to 59% in 2060s.
Figure 4: The proportion of working-age people with each private pension type within the household by cohort
5. Depth of undersaving
30. Whether someone is undersaving or not is a binary concept, therefore it is useful to understand how far individuals are from reaching an adequate pension income using TRR measures. Analysis of the depth of undersaving focusses on the TRR (BHC) achieved in the ‘all income’ scenario. The pattern of results is broadly consistent when considering AHC or when considering the ‘lump sum’ scenario, therefore only the ‘all income’ BHC results are presented.
31. Figure 5 shows the proportion of target income reached for the working age population. The lighter bars represent the 19.0 million people saving enough for retirement, and the darker bars represent the 14.6 million people undersaving. There is a wide range of outcomes, but a large number are relatively close to their target income. Of the 14.6 million undersavers, nearly half (47%) meet 80% or more of their target income.
Figure 5: Distribution of the proportion of target income reached for working-age people
32. With specific focus on the undersavers, everyone has been categorised into one of three categories:
- substantial undersavers – below 50% of their target income
- modest undersavers – 50% to 80% of their target income
- mild undersavers – 80% to just below 100% of their target income
33. This categorisation is consistent with previous publications. Figure 6 shows that the largest group of undersavers are in the fourth earnings band (£42,001 to £67,000) where 0.8 million are substantial undersavers, 2.5 million are modest undersavers and 3.0 million are mild undersavers. In the lowest income band (less than £15,900) there are fewer undersavers: 70,000 are substantial undersavers; 30,000 are modest undersavers; and 20,000 are mild undersavers due to State Pension providing a large proportion of their TRR. It is also important to note that the lowest income band is the smallest band in the analysis.
Figure 6: Number of undersavers classed as substantial, modest and mild undersavers in each pre-retirement earnings band
34. Figure 7 shows the median replacement rate achieved within each gross pre-retirement earnings band and how this compares to the TRR for that group. The median replacement rate in all these groups is at least the TRR, meaning that the median person in each group is projected to replace their target income in retirement.
Figure 7: Median replacement rate and TRR in each pre-retirement earnings band
35. The median replacement rate can be helpful, but it does not show the distribution of rates achieved within the earnings band and therefore doesn’t illustrate the range of outcomes. Considering the depth of undersaving within income bands can provide a clearer picture on the variation. Figure 8 shows that there is a large range of distribution in the lowest earnings band, with many low earners meeting their TRR. Figure 9 highlights that there is also a large distribution within the highest earnings band.
Figure 8: Distribution of pre-retirement income replaced in the lowest pre-retirement earnings band
Figure 9: Distribution of pre-retirement income replaced in the highest pre-retirement earnings band
6. Case study modelling
36. DWP’s IPEN model is used to create individual case studies, to show what individual circumstances might look like in retirement. The outputs consider:
- DC pot at retirement
- Annual income from annuitised DC pension and State Pension
- Replacement rate attained (compared to TRR)
37. Table 6 illustrates three different individuals and their retirement outcomes. These individuals are all born in 2002 and are automatically enrolled into a workplace pension from age 22 up to their SPa (in line with legislated on SPa timings). These individuals therefore entered the labour market in 2024 and are assumed to be saving at the AE minimum levels. These are all illustrative examples based on a given set of assumptions.
Table 6: individual case study outcomes in retirement for individuals born in 2002[footnote 12]
Individual Description | DC pot (2023 terms) | Annual Income | Replacement Rate | Target Replacement Rate | PLSA RLS Achieved |
---|---|---|---|---|---|
Median male earnings – Saving from 22, typical earnings profile and full work history | £165,000 | £24,000 | 64% (TRR not met) | 67% | Minimum |
Higher rate taxpayer – Saving from 22, typical earnings profile and full work history | £215,000 | £27,000 | 54% (TRR not met) | 60% | Minimum |
National Living Wage earner – Saving from 22, full work history | £85,000 | £20,000 | 89% (TRR met) | 70% | Minimum |
38. Under current policy circumstances, AE and State Pension provides the National Living Wage (NLW) earner with their target replacement rate for retirement and gets a median earner most of the way to their target. For the higher earners, the gap between income attained and their target income is larger. All case studies in table 6 reach the PLSA minimum RLS throughout their retirement, but none of them meet the moderate standard.
39. This assumes the individuals are saving at AE default levels (8% on qualifying earnings[footnote 13]pension for the entirety of their working life nor work up to their SPa. DWP’s survey Planning and Preparing for Later Life (PPLL)[footnote 14] finds that 2-in-5 of respondents expected to retire at 65 or younger, and the median ideal retirement age was 60. Table 7 illustrates how individual outcomes might look when they delay their pension saving, have time out of the labour market or retire earlier, at the Normal Minimum Pension Age (NMPA)[footnote 15].
Table 7: retirement outcomes for individuals who have delayed savings, career breaks or early retirement for individuals born in 2002[footnote 16]
Individual Description | DC pot (2023 terms) | Annual Income | Replacement Rate | Target Replacement Rate | PLSA RLS Achieved |
---|---|---|---|---|---|
Median earning male – Delays saving 10 years to 32, typical earnings profile and full work history who retires at 68 | £125,000 | £22,000 | 58% (TRR not met) | 67% | Minimum |
Median earning female – Saving from 22, typical earnings profile, career break for 5 years after first child, followed by 5 years part-time and retires at 60 | £60,000 | £17,000 | 52% (TRR not met) | 67% | Minimum |
Median earning male – Saving from 22, typical earnings profile, retires and purchases annuity at NMPA (57) | £75,000 | £18,000 | 45% (TRR not met) | 67% | Minimum |
National Living Wage earner – Saving from 22, sporadic[footnote 17] career breaks due to unemployment and retires at 68 | £70,000 | £19,000 | 80% (TRR met) | 70% | Minimum |
National Living Wage earner – Delays savings to 35, full work history who retires at 68 | £55,000 | £18,000 | 81% (TRR met)[footnote 18] | 70% | Minimum |
40. Table 7 shows retirement outcomes are sensitive to decisions made through working life. Where individuals delay their retirement saving or choose to retire early, they increase the gap between the income attained and their target replacement rate. For the median earning male case study, choosing to delay savings or to retire early causes them to fall short of their TRR. The median earning female case study has a 5-year career break[footnote 19] at the average age of having a first child (aged 31)[footnote 20], followed by 5 years part-time to help with childcare. She then retires early, to assist with parental care. These spells out of the labour market mean that they are unable to attain their TRR. For lower earners, such as NLW earners, having breaks in their saving can still allow them to meet their TRR, but still results in a lower income in retirement. Across all these cases, the individuals can meet the PLSA minimum standard with their private pension and state pension combined.
41. All case studies considered here have been in AE for their entirety of their working life. Given AE was introduced in 2012, some individuals do not have their entire working life to save into a DC pension, which will lead to smaller DC pots at retirement. Generation X (born 1965-1981) are less likely than previous generations to have DB pension wealth and are more likely than younger generations to have smaller DC pots.
Table 8: Median earner born in 1970 retirement outcome, auto enrolled from 2012[footnote 21]
Individual Description | DC pot (2023 terms) | Annual Income | Replacement Rate | Target Replacement Rate | PLSA RLS Achieved |
---|---|---|---|---|---|
Median earning male born 1970[footnote 22] – typical earnings profile and full work history. Assumed not to have saved into a pension until AE was rolled out. | £75,000 | £19,000 | 50% (TRR not met) | 67% | Minimum |
42. Table 8 shows a median earner in Generation X can attain a 50% replacement rate under current state pension and AE policy circumstances, whereas an individual with the same earnings profile (illustrated in table 6) in Generation Z could attain around a 64% replacement rate. This is driven by the amount of time individuals can save into a DC pension, for example the median earner in Gen X was auto-enrolled in 2012 at age 42 whereas an individual in Gen Z will be auto-enrolled from age 22 giving them an extra 20 years of saving.
7. Future Pension Incomes
43. This analysis looks at how much income pensioners will have in their first year of retirement, using Pensim3. This is determined by how much private pension wealth they have accrued at State Pension age. The analysis examines the composition of income and wealth and considers how these change for future cohorts of retirees. It replicates work of Pensioner income projections, which was last published in 2015.
44. The analysis looks at change over time under the existing policy framework, including the triple-lock uprating of State Pension and current automatic enrolment policy. Currently, the median weekly income for all pensioner couples is £595 per week, and for singles £282 per week.[footnote 23] In this analysis, it considers State Pension, DC incomes and DB incomes. For the lowest income quintile, Pension Credit has also been added to the State Pension. The figures are in weekly amounts and represent the year in which the individual reaches State Pension age. The projected weekly figures only represent these new retirees reaching State Pension age each year, not the total stock of all pensioners each year[footnote 24].
45. The analysis shows that average total incomes are projected to decline over the next 20 years for future retirees compared to past retirees in real terms. This is due to a lag in DC pension income replacing DB pension income. DB income and coverage is projected to decline for future retirees before people have realised a full career with auto-enrolment. Figures 10 and 11 illustrate that over the next 20 years, both mean and median DB incomes are projected to significantly decline, and DC incomes continue to grow as people retire with larger DC savings pots.
Figure 10: Mean weekly amounts in 23/24 earnings terms of those that reach SPa in each year
Figure 11: Median weekly amounts in 23/24 earnings terms of those that reach SPa in each year
46. The change in income varies across the different income quintiles. Figure 12 shows that new retirees in the highest income quintile sees the largest decline in pension income. New retirees in the two top income quintiles see mean income fall, driven by the decline in DB income. The mid-low income quintiles see some growth in total income. For the highest income quintile, the mean weekly income in 2035 is projected to be £712 and £659 by 2045. For the lowest income quintile, the mean weekly income in 2035 is projected to be £158 and £168 by 2045, supported by the triple lock and gains from auto-enrolment saving.
Figure 12: Mean weekly amounts in 23/24 earnings terms of those that reach SPa in each year by income quintile
47. Figure 13 shows that as time progresses, a smaller proportion of individuals will retire with DB income only, decreasing from 15% in 2033 to 6% in 2063. An increasing proportion will retire with DC pension income, those with DC only will reach 49% by 2063, and those with DC and DB will reach 40%. With time, smaller proportions will be completely reliant on State Pension, falling to 4% by 2063, driven by AE increasing coverage of private pension saving.
Figure 13: Estimated proportions of new pensioners with different pension types
8. Measuring adequacy of future pensioner incomes
48. Any assessment of the adequacy of future pensioner income is subjective. What an individual may need, or want, in retirement is highly dependent on their own individual preference and expectations. Therefore, when measuring the level of undersaving in the current working age population it is important to be aware that estimates can change based on the methodology, the measure used, and the assumptions made.
49. This analysis primarily focuses on what state and private pension income individuals will achieve in retirement. However, DWP’s PPLL survey found many individuals plan to use other sources of income to support in retirement. It is possible that individuals may use housing wealth, other savings or inheritance to support their pension income. For example, DWP’s PPLL 2024[footnote 25] survey found that 56% of 40–75-year-olds who had not yet retired expected to use savings or investments to support their retirement and 24% expected to release equity from their home or downsize.
50. Factors that influence saving by individuals during their working lives will impact on the adequacy of their pension income, for example:
- taking time out of the labour market, including leaving work before SPa
- not saving into a pension whilst in work – though pension participation has been high and opt-outs low under AE – though some groups are not included in AE, such as the self-employed
- not contributing enough into a pension when saving
51. As described above, replacement rates assume a pensioner would want to maintain the same standard of living in retirement that they had during their working life, and so they would need a level of pension income related to their pre-retirement earnings. This is founded from the concept of consumption smoothing. DWP’s Analytical Report[footnote 26] of the 2017 AE review stated the following reasoning behind using this method which still stands today:
Given that the fundamental purpose of pension systems is to replace income in retirement, a starting point for analysis is to look at income to assess the adequacy of saving. Most individuals would not be satisfied with simply receiving a basic minimum level of income, so it is important to estimate what level of income allows people to maintain their same standard of living in retirement.
52. DWP’s Analytical Report of the 2017 AE review also provided an overview of alternative approaches of measuring adequacy and the pensions industry has continued to develop approaches. For example:
The Joseph Rowntree Foundation Minimum Income Standard (JRF MIS) is an expenditure-based measure which determines the amount of expenditure required for a socially acceptable minimum standard of living sometimes referred to as a ‘basket of goods’ measure. However, it may not be considered as adequate for individuals who have enjoyed higher pre-retirement and could result in higher earners having to reduce their living standards in retirement.
53. The Pensions and Lifetime Savings Association (PLSA) Retirement Living Standards published in 2019 go above the JRF MIS and include a ‘Moderate’ and ‘Comfortable’ standard of living for retirees in the UK. The standards help individuals picture what kind of lifestyle they could have or may want in retirement. These standards were most recently updated in February 2024 and are considered within the report.
54. The analysis highlights there is no perfect measure of adequacy of future pension incomes and provides support for there being multiple measures which help to identify key groups who are most at risk of undersaving.
55. Other measures of adequacy have emerged in recent years. An example is the Living Pension[footnote 27], created by the Living Wage Foundation. The Living Pension standard is a voluntary savings target for employers that would help build a pension pot to provide enough income to meet basic everyday needs in retirement. The standard sets out a minimum annual contribution required through an average working life, which equates to 12% full-time living wage salary.
9. About these statistics
56. The analysis used in this report assesses the adequacy of pension incomes by modelling TRRs for simulated individuals and comparing pension income to the PLSA RLSs. All estimates can be found in the published data spreadsheet.
57. There are 33.6 million individuals included in the analysis when measuring adequacy using TRRs and 34.8 million when using PLSA RLS. This will be less than the current 22-SPa population as some individuals are excluded from the analysis. The 22-SPa population includes individuals that are modelled as not reaching SPa (either through emigration or death) or have no earnings between ages 50 and SPa. Individuals whose average income between 50 and SPa is below the Guaranteed Credit level for Pension Credit were excluded when measuring against the TRR’s as it is assumed their in-work earnings would be 100% replaced by Pension Credit.
58. These individuals are included when measuring against the PLSA RLS. The analysis may be underestimating these individuals’ total income as the analysis excludes means-tested benefits, like Pension Credit and Housing Benefit. However, this is not expected to make a material difference to who is meeting each retirement living standard.
59. Replacement rates measure income in retirement as a percentage of income in working life. In the analysis, pre-retirement earnings are calculated as the average earnings from age 50 until SPa over the period they are employed or self-employed. Pension income is calculated as the income from State Pension age onwards averaged over the whole of retirement. The same definition of pension income is used when comparing to the PLSA RLS.
Pre-retirement earnings
60. Pre-retirement earnings methodology:
- income is put in constant earnings terms before averaging over 50-SPa
- income is gross of tax, and of pension contributions
- only income from earnings (including self-employment) is included – income from benefits, tax credits, and pensions received before SPa are excluded
- when comparing against TRRs, the methodology is effectively looking at whether pension income will replace earnings
- when comparing against the PLSA RLS, the methodology is looking at whether pension income will meet each standard
- only years with positive earnings are included in the calculation of pre-retirement earnings. So, someone who stops work at 55 will have the same average pre-retirement earnings as someone earning the same amount who continues to work up to SPa
- for couples, average pre-retirement earnings are calculated separately for both members, summed, then equivalised see income equivalisation
Pension income
61. Pension income methodology:
- income is in constant price terms before averaging over whole of retirement
- income is gross of tax
- income includes State Pension and private pension income
- income from means tested benefits such as Pension Credit and disability benefits is excluded
- State Pension is uprated using the triple lock
- while financial wealth (financial assets), is not modelled directly, an amount is imputed and then annuitised based on Wealth and Assets Survey data. Other forms of individual or household wealth (for example, property wealth) are not included
- only income received after SPa is considered for the analysis
- for couples, their pension income is equivalised on a year-by-year basis and then averaged across retirement for the members of the couple separately. This means that the members of a couple can have different replacement rates as they have different lengths of retirement. They could also achieve different standards of living as measured by PLSA RLS
- Income is annuitised (either in the All Income or Income after Lump Sum scenario) so the analysis does not take into account pension freedoms but however shows what people could have if they used their pot to purchase an annuity
62. For replacement rates, each individual’s replacement rate is compared to a benchmark to determine whether it is adequate. The benchmark from the 2004 Pensions Commission report is used, with the earnings thresholds which they apply adjusted for earnings growth, see table 9.
63. Replacement rates can also be calculated before or after housing costs have been deducted from both pre-retirement earnings and pension income.
64. For the AHC measure, the Pensions Commission TRRs have been adjusted to reflect the impact of different estimated housing costs between working life and retirement. Using the Family Resources Survey, for each earnings band, average housing costs (net of housing benefits) are deducted during working life and retirement and the replacement rate is adjusted accordingly to obtain an AHC set of target replacement rates. Effectively, this means that those who rent during retirement will be required to reach a higher replacement rate than someone with equivalent income who is an owner occupier, as the renter will need to meet higher housing costs.
Table 9: Target Replacement Rate and Pre-retirement earnings bands
Gross Pre-retirement (50-SPa) Earnings | Pension Commission (2004) | 2023 Earnings Terms | BHC Target RR | AHC earnings | AHC Target RR |
---|---|---|---|---|---|
Band 1 | Less than £9,500 | <£15,900 | 80% | <£13,500 | 84% |
Band 2 | £9,500-£17,500 | £15,900-£29,000 | 70% | £13,501-£26,000 | 75% |
Band 3 | £17,500-£25,000 | £29,001-£42,000 | 67% | £26,001-£38,000 | 71% |
Band 4 | £25,000-£40,000 | £42,001-£67,000 | 60% | £38,001-£62,000 | 63% |
Band 5 | Over £40,000 | >£67,000 | 50% | >£62,000 | 53% |
65. The PLSA RLS have last been updated in February 2024, based on research conducted throughout 2023. These standards are used throughout the analysis. Although the PLSA RLS have separate standards for London and outside of London, our modelling assesses at a national level. The analysis therefore assesses individuals against the outside of London standards. As the present analysis uses gross pension income, the PLSA standards have been adjusted to allow for comparison. The analysis uses the single pensioner total income pre-tax as our benchmark Retirement Living Standards.
66. When measuring undersaving using PLSA’s RLS the analysis compares pension income AHC as the standards assume people are mortgage and rent free.
Table 10: PLSA Retirement Living Standards 2023/2024 (outside of London)[footnote 28]
Minimum | Moderate | Comfortable | |
---|---|---|---|
Single (Expenditure) | £14,400 | £31,300 | £43,100 |
Single (Total income pre-tax) | £14,857 | £35,982 | £50,887 |
Couples (Expenditure) | £22,400 | £43,100 | £59,000 |
Couples (Total income pre-tax) | £23,000 | £47,590 | £67,464 |
Income equivalisation
67. When two adults live together in a couple, they usually benefit from economies of scale in their normal living costs. For example, it is expected that two adults will pay a rent or mortgage that is less than twice as much what each of them would pay if living separately. The same applies to other normal living costs (transport, utilities etc).
68. Income equivalisation is a technique that recognises these economies of scale and adjusts a couple’s income accordingly. For example, in the analysis without housing costs, each individual in a couple is assigned an income equal to 67% of the whole couple’s income. When housing costs are subtracted, each individual in a couple is assigned an income equal to 58% of the whole couple’s income[footnote 29].
69. In both cases, the total income assigned to the couple exceeds 100% of the actual nominal income, reflecting the fact that when in a couple, each pound ‘goes further’ due to economies of scale.
Key assumptions in Pensim3
70. The analysis is shaped by methodological choices and the capacity of the modelling tools. Assumptions about inflation, earnings growth and the labour market are consistent with assumptions from the OBR.
71. Workplace participation is set in line with data from the Annual Survey of Hours and Earnings (ASHE 2022). It is dependent on factors such as AE eligibility, sector and income. Of those eligible for AE, 86% of private sector and 92% of public sector employees are saving into a pension. Personal pension participation is based on the Wealth and Assets survey. In the model, 5% of employees and 17% of self-employed pay into a personal pension each year.
72. Based on data from the Annual Survey of Hours and Earnings (ASHE) 2022 dataset, individuals saving into a workplace DC pension are firstly assigned to pay contributions on either qualifying earnings or the full value of their pay. They are then assigned a total contribution (employee plus employer) of between 8% and 30%. The contribution rate assigned, and whether this is paid on full pay or qualifying earnings, is dependent on factors such as income.
73. Annual management charges are derived from a probability distribution. For workplace schemes these are capped at 0.75%, based on the charge cap.
74. It is assumed 86% of annuities purchased are flat rate, the remaining 14% are linked to the Retail Price Index (RPI), using FCA Retirement Income Data between March 2017 and April 2021.[footnote 30] The annuity rates an individual receives depends on the type of annuity they purchase, their age and the year the annuity is taken. It is assumed that as life expectancy increases, annuity rates decrease.
75.DC schemes have an assumed range of investment fund growth based on Black Rock equities and bonds returns. The modelling assumes that throughout most working life, there is an 88% / 12% split of equities to bonds, with an average return of 6.15% / 4.43% respectively. In the modelling, individuals getting closer to retirement age enter ‘lifestyling’ whereby their pension fund investments are changed to be less risky in these years and start to increase the share invested in bonds and reduce those invested in equities.
76. In Pensim3, housing tenure type is assigned when an individual turns 60. This is done using an alignment to ensure that the proportion getting assigned to each housing tenure type matches the proportion in the population.
Pension Income Projections
77. The analysis considers individual income rather than household, which is different from the undersaving analysis where household income is equivalised. The weekly figures only relate to these individuals reaching State Pension age, not the total stock of all pensioners each year. The analysis has some key assumptions which include:
- Everyone is assumed to retire at State Pension age.
- There are no economic shocks.
- The analysis assumes funds are annuitised. There are two scenarios consistent with the undersaving analysis.
a. All Income Scenario where individuals with DC pensions are modelled as converting their entire pension pot into an annuity.
b. Income after Lump Sum where individuals with DC pensions withdraw 25% of their pot as a lump sum before converting their pot into an annuity.
Key assumptions in IPEN
78. The following key assumptions have been made when calculating these case studies. The modelling is sensitive to assumptions made about future economic determinants and the methodology used.
a. Each individual saves into a defined contribution scheme with an annual management charge of 0.3% and a contribution charge of 1.8%.
b. Each individual contributes 5% and their employer contributes 3%.
c. Each individual’s fund is invested in 88% equities and 12% bonds with real fund growth of 4.07% and 2.39% respectively.
d. Each individual retires at State Pension age of 68.
e. The automatic enrolment earnings trigger, lower earnings limit and upper earnings limit are frozen until 2028 and then increase in line with earnings over the long-term.
f. Each individual does not opt-out of pension saving.
g. For National Minimum/Living Wage (NLW) earners the individual continues to make pension contributions whilst working part time, even though their earnings drop below the £10,000 earnings trigger. This is in line with the AE framework, assuming they continue to work for the same employer.
h. Full time employees work 37 hours per week whilst part-time employees work 18.5 hours per week.
i. Each individual’s earnings increase in-line with average earnings growth. The final pension pot size is reported in 2024/25 earnings using the Average Weekly Earnings growth deflator.
j. Where earnings have been reported, these are the current values. For the NLW/NMW earner aged 40 years, earnings in previous years were decreased in-line with average earnings. This means that income in previous years may not match the NLW/NMW figure for that year.
k. Earnings in the first-year employment are increased in line with average earnings growth.
l. The median annual salary for a female working full-time is £31,672 (Source: ASHE 2023)[footnote 31].
m. The median annual salary for a male working full-time is £37,382 (Source: ASHE 2023).
n. Salaries for individuals on NMW/NLW are calculated by hourly rate multiplied by weekly hours worked multiplied by 52.
9. Statement of compliance with the Code of Practice for Statistics
The Code of Practice for Statistics (the Code) is built around 3 pillars:
- trustworthiness – is about having confidence in the people and organisations that publish statistics
- quality – is about using data and methods that produce statistics
- value – is about publishing statistics that support society’s needs
The following explains how these pillars of the code have been applied in a proportionate way for this analysis.
Trustworthiness
These figures have been published to provide an updated estimate of the number and proportion of people undersaving for retirement. They are being released now to ensure equal access to the analysis. The analysis now includes case study analysis, to provide more estimates to the public and ensure transparency.
Quality
Future pension incomes are modelled using DWP’s long-term dynamic microsimulation model, now referred to as Pensim3. The model has been through extensive development to ensure it is using the latest available data to underpin its assumptions. The methodology and calculations in this analysis have been quality assured by DWP analysts to ensure they are robust. The IPEN model has been updated to reflect the latest available data and has been extensively quality assured by DWP analysts.
Value
Releasing these estimates provides the public and user organisations with up-to-date estimates on the adequacy of future pension incomes, which were last published in 2023. Making this information accessible helps reduce the requirement for Parliamentary Questions, Freedom of Information requests and ad hoc queries to be asked about pension adequacy through transparency of information and analysis.
Authors
Sophie Phillips
Annex
A1 Key Modelling Changes from 2023 Publication[footnote 32]
- Changes to the labour market methodology. The model now has updated assumptions using more recent data to estimate what types of jobs individuals have throughout the model (+1 ppt impact on undersaving).
- Using more recent data on outturns on employment and saving levels, with amendments to missing data (+3 ppts impact).
- Changes to the investment returns methodology. The previous analysis used RPI+ to estimate investment returns - this is now based on Capital Market assumptions and split by equities and bonds determined by Corporate Advisor Pensions Average (CAPA) data. This lowers investment returns (-3 ppt impact).
- Updates to Office for Budget Responsibility (OBR) economic assumptions. Office for National Statistics and OBR have revised their population projections and economic assumptions respectively (-2 ppts impact).
A2 Creating personas using IPEN
To bring alive some of the challenges and problems faced when saving for retirement, personas (hypothetical case studies) can be used. This can illustrate, that for certain circumstances, the likely scenario that individuals may face in their retirement. These personas are generated using DWP’s IPEN model[footnote 33] and are enhanced through narrative to deliver a more relatable example.
Low earner
Sam is a national living wage earner. They have worked in the hospitality industry since they were 18, but didn’t start private pension saving until they were auto enrolled at age 22. Sam works in hospitality, earning the national living wage throughout their whole career. They continue to work at the national living wage until age 68, as they didn’t wish to draw on any private saving before they could receive their State Pension. Like around 77%[footnote 34] of those 40-who haven’t accessed their DC pot, they don’t have a clear plan for accessing their DC pension. Sam has a pot of £85,000 at age 68. With pension freedoms, there are several options available to them, the three they consider are:
- taking an annuity- resulting in an average annual retirement income of £20,000, with around £15,000 coming from State Pension and £5,000 from their annuity
- taking a lump sum (25% of their pot) and entering drawdown at a rate of around 6% per annum, giving them around £4,000 a year income on top of their £15,000 State Pension
- cashing the entire pension pot out - resulting in average annual retirement income of £15,000, solely from the State Pension
Average earner
Jack is a median earner working in legal administration services. He started his job as a graduate at age 21 but is auto enrolled from age 22. Jack works in his role, with a typical earnings profile, throughout his entire career. He chooses to retire at his State Pension age, which is 68, as he feels supported in his firm. Like around 77%[footnote 35] of those 40-who haven’t accessed their DC pot, he doesn’t have a clear plan for accessing his DC pension. Jack has a pot of £165,000 at age 68. With pension freedoms, he has several options available to him, and he considers:
- taking an annuity- resulting in an average annual retirement income of £25,000, with around £15,000 coming from State Pension and around £10,000 from their annuity
- taking a lump sum (25% of his pot) and entering drawdown at a rate of around 6% per annum, gives him around £7,500 a year from private pension income
- cashing the entire pension pot out – would leave them with around £15,000 annual retirement income solely from the State Pension
Gaps in work history
Jade is a median earner, working in financial services. She started her job as a graduate at age 21 but was auto enrolled from age 22. Jade, like 4-in-5 women in the UK has a child, and takes a 5-year career break in her early 30s to support the child through their early years. She returns to work part-time for 5 years, like around 40% of mothers.
She then works until aged 60, where she retires early to support with caring for her elderly parents. She is retiring earlier than she thought but she has concerns about the cost of social care otherwise. Jade has a pot of £60,000 at age 60. Like around 77%[footnote 36] of those 40-75 who haven’t accessed their DC pot, she does not have a clear plan for accessing her DC pension plan. She considers:
- taking an annuity- resulting in an average annual income of £17,000 where £14,000 is from State Pension and £3,000 is from the annuity
- taking a lump sum (25% of her pot) and entering drawdown- resulting in an average annual income of £16,000, where £2,000 a year comes from drawing their pension down at a rate of around 4% per year[footnote 37]
- cashing the entire pension pot out- resulting in average annual income of around £14,000 and being completely reliant on State Pension
-
Workplace pension participation and savings trends of eligible employees: 2009 to 2023 - GOV.UK ↩
-
Based on Turner Commission 2004 Income Bands for Target Replacement Rates, in 2023 prices ↩
-
Workplace pension participation and savings trends of eligible employees: 2009 to 2023 - GOV.UK ↩
-
Rounded to the nearest 10,000 ↩
-
The aggregate figures are different for analysis of TRRs compared to analysis of PLSA measures. This is because individuals whose average income between 50 and SPa is below the Guaranteed Credit level were excluded when measuring against the TRR’s as we assume their in-work earnings would be 100% replaced by Pension Credit ↩
-
20231018-ppi-technical-report-for-plsa-adequacy-modelling-update-final.pdf ↩
-
The income bands here refer to both BHC and AHC earnings bands. Band 1 is the lowest band and band 5 is the highest. Table 9 reports these bands in £ earnings ↩
-
This includes both private and social renters ↩
-
The UK Pensions Framework: Renting in Retirement- the Fault Line Below the UK Pensions System ↩
-
DC pot rounded to the nearest £5,000, annual income to the nearest £1,000 ↩
-
Planning and Preparing for Later Life Survey – forthcoming, expected Summer 2025 ↩
-
The NMPA is the minimum age most savers can access their pensions without incurring unauthorised payments tax charge, unless retiring due to ill health. This is set to increase from age 55 to age 57 in April 2028 ↩
-
DC pots are rounded to the nearest £5,000, annual incomes are rounded to the nearest £1,000 ↩
-
Career breaks are: 2 years out of work aged 22 and 23, 3 years out of work at age 30, 31 and 32 and 5 years out of work at 45, 46, 47, 48 and 49. A total of 10 years out of the labour market ↩
-
In the absence of private pension income, the replacement rate achieved through State Pension income would be 66% ↩
-
How women’s employment changes after having a child - Understanding Society ↩
-
Birth characteristics in England and Wales - Office for National Statistics ↩
-
DC pot rounded to the nearest £5,000, annual income to the nearest £1,000 ↩
-
For this individual case study, the individual was automatically enrolled into DC saving in 2012 ↩
-
Pensioners’ Incomes: financial years ending 1995 to 2024 - GOV.UK ↩
-
This is calculated as a five-year moving average, e.g. 2021 is the average of 2019-2023 ↩
-
Planning and Preparing for Later Life- forthcoming, expected Summer 2025 ↩
-
The equivalisation assumes all individuals in the household benefit equally from combined income. Equivalence scales typically take an adult couple with no children as 1, the process relatively increases the income of a single person household. This method is consistent with other Departmental publications such as HBAI. ↩
-
Retirement income market interactive analysis 2023 to 2024 FCA. Recent data further reflects that flat rate annuities make up most annuities purchased, giving further evidence to the modelling assumptions ↩
-
Annual Survey of Hours and Earnings time series of selected estimates - Office for National Statistics ↩
-
All percentage point impacts refer to the TRR BHC estimate ↩
-
Figures for DC pots are rounded to the nearest £5,000, and figures for annual income are rounded to the nearest £1,000 ↩
-
Planning and Preparing for Later Life Survey – forthcoming, expected Summer 2025 ↩
-
Planning and Preparing for Later Life Survey – forthcoming, expected Summer 2025 ↩
-
Planning and Preparing for Later Life Survey – forthcoming, expected Summer 2025 ↩
-
Jade is assumed to drawdown at a lower level than the other personas presented in the annex due to her earlier retirement ↩