Chapter 6: Private pension wealth, Wealth in Great Britain, 2012 to 2014

• In July 2012 to June 2014, 35% of adults aged 16 and over contributed to a private pension. The percentage varied by sex, with 37% of men making contributions compared with 32% of women

• In July 2012 to June 2014, a much higher proportion of employees in the public sector (84% with median wealth of £61,600) belonged to a current occupational pension scheme than their counterparts in the private sector (42% with median wealth of £24,000)

• In July 2012 to June 2014, median wealth held in private pensions from which individuals had not yet drawn an income (that is, current and retained pensions) was much higher in defined benefit (DB) pensions (£63,400) than in defined contribution (DC) pensions (£15,000)

• In July 2012 to June 2014, 19% of individuals aged 16 and over received income from a private pension. The median wealth held in pensions that were already being paid (pensions in payment) was £116,300. A higher proportion of men (22%) than women (17%) received such income. The median level of wealth held by men (£162,400) in pensions in payment was more than double that of women (£73,900)

• In July 2012 to June 2014, around a quarter (24%) of all households in Great Britain had no private pension wealth, the same proportion as in July 2010 to June 2012

• Aggregate private pension wealth in Great Britain increased from £3.5 trillion in the period July 2010 to June 2012 to £4.5 trillion in the period July 2012 to June 2014 (figures not adjusted for inflation). This was mainly explained by an increase in pensions in payment pension wealth driven by changes in annuity rates used to value such wealth

• In July 2012 to June 2014, of those who had any private pension wealth, the 10% of households with the highest total pension wealth had almost half of the all pension wealth in Great Britain (47%). This was almost six times the total private pension wealth of the 50% of households that had the lowest (8%)

This chapter looks at estimates of private (non-state) pension wealth in Great Britain from the Wealth and Assets Survey (WAS). It presents new data from the survey for the period July 2012 to June 2014 alongside revised estimates from the July 2010 to June 2012 period.

Unlike the other forms of wealth presented in this report, pension wealth is not immediately accessible for most individuals. The earliest age at which it was possible to receive an income from a registered private pension increased to 55 from April 2010 as a result of the Finance Act 2004. The first 3 data collection periods of the survey took place before automatic enrolment was introduced in October 2012. The July 2012 to June 2014 data collection period was predominantly following October 2012 so this period should reflect some changes in pension membership and wealth due to automatic enrolment. The statistics for all data collection periods predates the pension flexibilities reform that was introduced in April 2015.

The figures in this chapter relate to private pension wealth only, which means state pension wealth is excluded from the analysis. The latter part of the chapter will show that wealth from private pensions is not very evenly distributed, as many individuals and households have zero or very low private pension wealth. As state pension wealth is more evenly distributed, the distribution of total pension wealth (state plus private) will be less skewed.

The July 2012 to June 2014 period methodology is generally consistent with that used in the July 2010 to June 2012 period. Data from the period July 2012 to June 2014 has been used to improve the imputation process in the previous period (July 2010 to June 2012). Compared to the previous release of the July 2010 to June 2012 data, the changes and improvements through imputation have resulted in some changes to estimates of pension wealth in different areas but predominantly unchanged estimates of membership levels. Overall, the estimates of aggregate pension wealth in this release are 21% lower for the July 2010 to June 2012 period compared to the July 2012 to June 2014 period. The increase in aggregate pension wealth from the July 2010 to June 2012 period to the July 2012 to June 2014 period can largely be attributed to changes in financial parameters used to value defined benefit pensions and pensions in payment wealth (see quality assurance section for further detail).

The chapter begins by looking at the membership of, and level of wealth held in, current pensions, defined as pensions to which individuals were contributing during the reference period. This information is presented for the different types of pension: occupational Defined Benefit (DB); occupational Defined Contribution (DC); and personal pensions, which include group personal and group stakeholder pensions – see Concepts and definitions. In addition, estimates of current occupational pension wealth (DB and DC occupational pension wealth combined) are presented by whether an employee was working in the private or public sector.

The chapter also provides estimates of pension wealth held in retained pensions. These are pensions to which individuals have stopped contributing but from which they are not yet drawing an income. This is followed by analysis comparing, for pensions that have not yet been drawn (current plus retained), the level of wealth held in DB and DC types of pensions. Estimates of wealth held in pensions from which individuals were receiving an income (pensions in payment) are also considered.

The chapter closes by bringing the different forms of private pension wealth together to look at wealth in all private pensions, that is, current, retained and pensions in payment. This is shown at the individual level, for households and for Great Britain as a whole.

The data presented in this chapter are in the form of cross-sectional estimates. All estimates of wealth are in nominal terms (values are not adjusted for inflation). However, some of the estimates required modelling, which was done using a method developed by the Institute for Fiscal Studies (IFS)¹.

Due to the complexity of the data, for example, the use of imputed values and complex weighting, no formal significance testing has been undertaken.

The survey sampled private individuals and households in Great Britain. This means that people in residential institutions, such as retirement homes, nursing homes, prisons, barracks or university halls of residence, and also homeless people, are excluded from the scope of the analysis presented in this chapter.

Further information about the survey is included in Chapter 1: Introduction and Demographics and in Chapter 8: Technical Details (335.5 Kb Pdf).

Concepts and definitions

Private pensions

All pensions that are not provided by the state. They comprise occupational and personal pensions, and include pensions of public sector workers.

Defined benefit (DB) pensions

Pensions in which the rules of the scheme specify the rate of benefits to be paid. The most common DB scheme is a “final salary” scheme in which the benefits are based on the number of years of pensionable service, the accrual rate, and either the final salary, the average of selected years’ salaries, or the best year’s salary within a specified period before retirement. Other types of DB scheme include career average re-valued earnings (CARE) schemes.

Defined contribution (DC) pensions

Pensions in which the benefits are determined by the contributions paid into the pension, the investment return on the contributions (which are normally invested in the stock market), and the type of annuity or drawdown product purchased upon retirement². An annuity is a contract between an insurance company and an individual under which the individual pays all or part of their pension fund to the insurance company in return for an agreed regular income for the remainder of their life. DC pensions are also known as money purchase pensions. They can be either occupational or personal.

Personal pensions include stakeholder and self invested personal pensions, both of which are forms of DC pension. Personal pensions can be sponsored by an employer (referred to as group personal pensions) or arranged on an individual basis. In this chapter, the definition of occupational does not include any personal pensions, while the definition of workplace pensions includes occupational pensions, group personal pensions and group stakeholder pensions.

Medians and means

The median is the preferred measure of central tendency in this chapter because many of the data distributions are not symmetrical. This is because a small proportion of individuals have high values of wealth with a larger proportion of individuals having very low wealth by comparison.

In such unequal distributions, the mean is likely to be influenced by high values, so it does not reflect the experience of most individuals.

Pension wealth

The calculation of pension wealth is complicated. Private pension wealth was split into nine categories and a slightly different valuation method was applied to each. The nine categories of private pension wealth are:

• DB occupational pensions to which interviewees were currently contributing

• DC occupational pensions to which interviewees were currently contributing

• personal pensions (all DC) to which interviewees were contributing or could have contributed at the time of the interview, including group personal and group stakeholder pensions offered by employers

• additional voluntary contributions (AVCs) to personal pensions (all DC) made by people with DB pensions

• retained rights in DB pensions

• retained rights in DC pensions

• remaining value of pension funds from which people were drawing an income through ‘income drawdown’ (where people take income from the fund but the fund remains invested)

• pensions expected in the future based on the pension contributions of a former spouse or partner; Pensions already being paid out (‘pensions in payment’)

The pension wealth figures presented here represent a person’s future pension in retirement, expressed as an equivalent ‘pot of money’. The estimates include only the pension rights accumulated to date. For people who are still working, they do not include rights which may be built up in future.

Wealth from DB pensions (current and retained) and pensions in payment are calculated using financial parameters (annuity factors and or discount rates) which change over time. Wealth from DC pensions is calculated from the reported value of the fund. This was explained in more detail in an annex on pension wealth methodology contained in the report for the period July 2008 to June 2010. Chapter 8 of this Wealth and Assets report also contains relevant information. The change in these financial parameters has affected the estimates of wealth from July 2010 to June 2012 period to July 2012 to June 2014 period and the effect has been investigated in the quality assurance section of this chapter.

Notes for introduction

1. As highlighted in the ONS Wealth in Britain Wave 2.

2. There have been recent changes in options available for drawdown introduced with pension flexibilities in April 2015.

This section explores membership of, and amount of wealth held in, private pensions to which individuals in Great Britain were currently contributing. It, therefore, does not include the pensions that an individual may have contributed to in the past but was no longer contributing to, or pensions from which an individual was receiving an income.

Membership of current pensions

The first part of this section focuses on the proportion of individuals aged 16 and over that were contributing to a pension or to more than one type of pension. It is followed by more detailed analysis of the amounts of wealth held in each pension type and the percentage of individuals in Great Britain, by age group, who were and were not contributing to specific types of private pensions in July 2012 to June 2014.

Table 6.1 shows that the majority of adults (65%, 31.8 million) in Great Britain were not contributing to a private pension in the period July 2012 to June 2014; with a higher proportion of women (68%) than men (63%) not contributing to a private pension. In the period July 2012 to June 2014, 16.9 million adults were contributing to a private pension with occupational defined benefit being the most common type (17%).

Examining the types of pension that people did contribute to, Table 6.1 shows that, in July 2010 to June 2012 and July 2012 to June 2014, a higher proportion of men than women contributed to either a Defined Contribution (DC) scheme, a personal pension or to more than one type of pension. However, in both periods a higher proportion of women than men contributed to a Defined Benefit (DB) scheme. This may be because a larger proportion of public sector workers are women and pension provision in the public sector predominantly consists of DB schemes.

Current occupational Defined Benefit (DB) pension wealth

Some employers offer their employees the opportunity to join a DB pension scheme. The “Concepts and Definitions” section of this chapter has a description of this type of scheme. Table 6.2 shows the proportion of individuals in Great Britain that belonged to current DB schemes and the wealth individuals held in these schemes.

Table 6.2 shows that in the period July 2012 to June 2014, the overall median for current DB pension wealth was £64,900, which was £21,600 more than the July 2010 to June 2012 equivalent. The main reason for the large increase in current DB wealth is thought to be due to the change in financial parameters used to value DB pensions between the periods of July 2010 to June 2012 and July 2012 to June 2014 (see quality assurance section).

The change in the value of a DB pension (as valued in WAS) can be equated to an individual’s experience of changing house prices. The value of a property or in this case a DB pension can change considerably without an individual changing their behaviour. A property or DB pension can be thought of as an asset. Although the asset remains largely unchanged the value placed upon that asset can change with the market. The financial parameters used to value DB pensions have changed between the period July 2010 to June 2012 and the period July 2012 to June 2014. The consequence of this is that a fixed DB pension would be valued higher on average in the July 2012 to June 2014 period than in the July 2010 to June 2012 period. DB wealth in WAS can be thought of as the equivalent DC pension pot that would be needed to buy the regular income expected from the DB pension. The estimate is valued at the date the individual responded to the survey (based on current entitlement) using financial market data (see quality assurance section for further detail).

In both periods, the overall median current occupational DB pension wealth for men was much higher than that of women. This lower level of wealth among women reflects the combination of fewer years of pension membership and the generally lower salaries of women in employment compared with men³. In the July 2012 to June 2014 period, the median for women was 60% of the median for men at £52,500 and £87,700, respectively (see reference tables).

Current occupational Defined Contribution (DC) pension wealth

Some employers offer their employees the opportunity to join a Defined Contribution (DC) pension scheme. In these types of scheme, the income an individual will receive in retirement usually depends on the contributions that have been paid in, the investment return received on those contributions and the financial products available at retirement. Table 6.3 shows the proportion of individuals in Great Britain that belonged to DC schemes and the wealth those individuals held in these schemes.

Table 6.3 shows, between the period July 2010 to June 2012 and the period July 2012 to June 2014, median occupational DC pension wealth had decreased from £10,500 to £9,000. This may be explained by the introduction of automatic enrolment. There are now more people with current occupational DC pension wealth, but as those automatically enrolled have started contributing relatively recently at the statutory minimum2, their pension wealth is likely to be lower than people who have been contributing to pension schemes for several years.

The median current occupational DC pension wealth for men was larger than for women in the period July 2012 to June 2014, the median for women (£5,500) was just under half of that for men (£12,000).

Current personal pension wealth

All individuals are eligible to make contributions to personal pensions should they choose to do so. Contributors to personal pensions include individuals not eligible for workplace pensions such as the self-employed, those not currently working, those currently not offered a pension scheme by their employer and those contributing on top of their occupational pension.

As explained in the “Concepts and Definitions” section, personal pensions can be purchased from an insurance company by an individual. However, an employer may facilitate the purchase of personal pensions for its employees (known as a group personal or group stakeholder pension). Self-invested personal pensions (on an individual or group basis) are also included in Table 6.4.

As all personal pensions are DC schemes, the income an individual will receive in retirement from such pensions depends on the contributions that have been paid in, the investment return received on those contributions and the financial products available at retirement. Table 6.4 shows the proportion of individuals belonging to personal pensions and the wealth those individuals held in these pensions.

Table 6.4 shows that between the periods July 2010 to June 2012 and July 2012 to June 2014, the overall median current personal pension wealth increased from £15,700 to £18,600. The table shows that the proportion of individuals with wealth in current personal pensions between the periods has stayed the same at 9%. In both periods, the proportion of men that were contributing to a personal pension was about twice the proportion of women. The median current personal pension wealth of all men in July 2012 to June 2014 was almost twice that of women, £24,000 compared with £12,400.

Total current pension wealth

Tables 6.2 to 6.4 explored the value of current pensions held in occupational DB, occupational DC schemes or in personal pensions. The following table and associated text examines the value of total wealth held in all types of current private pensions, that is, the combined value of all DB, DC and personal pensions to which individuals contributed.

Table 6.5 shows that in the period July 2012 to June 2014, 35% of individuals aged 16 and above were contributing to a private (non-state) pension. Between the periods July 2010 to June 2012 and July 2012 to June 2014, median wealth in current private pensions increased from £33,900 to £40,000 for men and from £22,500 to £28,000 for women.

Current occupational pension wealth in the public and private sector

Since the data collection period of July 2008 to June 2010, WAS has captured information on whether an individual worked in the public or private sector. Employees were asked whether the firm or organisation that they worked for was a private firm, business or limited company, or some other kind of organisation (such as a university, charity or public limited company). Based on their responses, employees were classified as belonging to either the public sector or private sector, with some employees being classified as “unknown”.

Table 6.6 shows the wealth of employees currently contributing to occupational pensions, but not to other workplace pensions (which include group personal pensions). It is not possible to split the wealth of those contributing to personal pensions in WAS between those contributing to group personal pensions and those contributing to individual personal pensions.

Table 6.6 shows that, in the period July 2012 to June 2014, a higher proportion of employees in all age groups in the public sector (84%) belonged to an occupational pension than those in the private sector (42%). The percentage of people belonging to a private sector occupational pension scheme increased slightly between the periods July 2010 to June 2012 and July 2012 to June 2014 from 39% to 42%.

In general, employees in the public sector have more wealth in current occupational pension schemes than their counterparts in the private sector. In the period July 2012 to June 2014, the overall median wealth of public sector workers was £61,600 compared with £24,000 in the private sector. This difference increased absolute terms between the periods July 2010 to June 2012 and July 2012 to June 2014 and was mainly driven by the valuation of DB pensions which are more common in the public sector.

In the public sector, occupational pensions are predominantly DB schemes, which tend to have greater wealth than DC. In the private sector, there is a mix of DB and DC pensions, with a higher proportion of the latter in recent years3. This results in an overall lower level of pension wealth in the private sector.

Notes for current pension wealth

1. As highlighted in the ONS Annual Survey of Hours and Earnings.

2. The changing landscape of workplace pensions in described in the Department of Work and Pensions Automatic enrolment evaluation report 2015.

3. As highlighted in the ONS Occupational Pension Schemes Survey.

The following section examines the wealth held by individuals in pensions to which they were no longer contributing but from which they were not yet drawing an income. This will typically be the case when individuals have been a member of their employer’s pension and then left that employer before reaching the age at which they were able to draw an income.

There has been a slight decrease in the proportion of individuals with retained pension wealth, from 17% in the period July 2010 to June 2012 to 16% in the July 2012 to June 2014 period. Table 6.7 also shows that the proportion of individuals with retained pension wealth generally increased with age up to the 55 to 64 age group in both periods. Median retained wealth was lower in the 65 and over age group as individuals have reached the age that they were able to cash in their retained rights and draw their pension incomes.

The median wealth held by those with retained pensions increased from £20,900 in the period July 2010 to June 2012, to £21,900 in the July 2012 to June 2014 period. For men, across the periods, median wealth increased from £25,000 to £27,700, while for women, it increased from £16,700 to £17,100.

Changes in retained pension wealth are difficult to interpret as a combination of factors affect the estimates, for example:

• changes in the proportion of people reporting retained pensions

• changes in the composition of retained pension wealth (changing proportions of DB and DC wealth)

• changes in the financial assumptions behind the calculation of retained defined benefit pension wealth across the three periods (see annex on pension wealth methodology)

Pension wealth in the accumulation phase

This section provides a complete picture of the accumulation phase by bringing together private pension wealth held in current and retained pensions, from both occupational and personal pensions. In other words, the section explores the wealth held by individuals in private pensions from which they were not yet drawing an income. As private pensions play a crucial part for many people in the savings made for retirement¹, this gives an indication of the level of non-state pension resources that is currently available to individuals to be put towards retirement.

Table 6.8 shows the amount of wealth held in private pensions that were not in payment, as well as the percentage of individuals who had such wealth, by the two main types of pension in which individuals could have built up wealth.

Table 6.8 shows that in the accumulation phase there was a large difference between DB and DC levels of wealth. In the period July 2012 to June 2014, median wealth held in pensions that were not yet in payment was £63,400 for DB pensions compared with £15,000 for DC pensions. The biggest differences in median wealth between the two types of pension were in the older age groups. This gap has increased for the 55-64 age group from £92,200 in the July 2010 to June 2012 period to £136,600 in the July 2012 to June 2014 period.

In both periods, the proportion of individuals who had wealth held in all pensions from which they were not yet drawing an income was highest for those in the age group 45 to 54. The proportion with wealth was low in the 16 to 24 age group and also in the 65 and over age group where individuals are typically drawing on their pension wealth. This pattern was the same for those with DB and DC wealth not yet in payment.

The median total wealth held in pensions that are not yet in payment was higher in the period July 2012 to June 2014 than it was in the period July 2010 to June 2012 (£39,000 and £32,500, respectively). There was no change in median DC wealth between the periods, which remained at £15,000, while the DB median pension wealth rose from £39,400 to £63,400, mainly due to valuation change, driving the increase in median total wealth between these periods.

Notes for retained pension wealth

1. As highlighted in the DWP estimates of the number of people facing inadequate retirement incomes.

The following section complements the retained pension wealth section by looking at wealth held by individuals in private pensions that were in payment. This means that the chapter moves on from showing private pension wealth in the accumulation phase to showing private pension wealth in the decumulation phase. However, it should be noted that it is possible that an individual could be receiving an income from one private pension while still accumulating pension wealth in another.

In WAS, wealth derived from pensions in payment was calculated by asking people how much private pension income they receive and then working out how much would be needed to purchase this pension, in the form of an annuity, for the remainder of their life. Those in older age groups, with fewer years of life remaining, have lower levels of wealth than those in younger age groups.

Table 6.9 shows estimates for the proportion of individuals who were receiving any income from a private pension and the value of this wealth. This includes private pensions received from a former spouse. Since very few people under the age of 50 received any income from private pensions, the age categories shown in Table 6.9 are different from those shown in previous tables in order to focus on the distribution of pensions in payment wealth within the older population.

Table 6.9 shows that in the period July 2012 to June 2014 the overall median wealth held in pensions in payment was £116,300, compared with £91,100 in the July 2010 to June 2012 period. The increase in pension in payment wealth is thought to be mainly down to the way such wealth is valued. Financial parameters (annuity rates) used to value pension in payment wealth have changed from between the periods and as a consequence such pensions have been valued higher in the period July 2012 to June 2014 (see quality assurance section).

For all individuals, the proportion receiving income from private pensions has been fairly similar in each period (19% in July 2012 to June 2014 and July 2010 to June 2012). For the three oldest age groups (65-69, 70-74 and 75+), the percentages were also fairly constant (in the range of 62% to 64% in all periods), increasing slightly over time. For men, the median wealth held in pensions in payment (£162,400) in the period July 2012 to June 2014 was more than double that for women (£73,900) and the proportion who received income from pensions in payment was also higher for men (22%, compared to 17% for women). In the groups for ages over 65, but below 75, the proportions of men receiving income were much higher than women: around three quarters in each male age group compared with about half of women. This may be because women in these age groups had fewer opportunities to contribute to pensions when they were of working age.

Table 6.10 shows the percentage of individuals with any private pension wealth by age and sex. It also includes details of the median private pension wealth for these individuals.

As Table 6.10 shows, 59% of individuals had some private pension wealth in the period July 2012 to June 2014, which is about the same as in July 2010 to June 2012 period (58%). The median value for all private pension wealth (excluding those with zero private pension wealth) has increased with each successive period, from £46,600 in the period July 2010 to June 2012 and £57,000 in the July 2012 to June 2014 period. For individuals as a whole, including those with zero private pension wealth, the median wealth in private pensions was £7,500 in the July 2012 to June 2014 period and £7,000 in July 2010 to June 2012 period (see download for private pensions with zero wealth included).

In both periods, a higher proportion of men than women had some pension wealth. In the period July 2012 to June 2014, for example, the proportions were 64% and 55% for men and women respectively. Overall, the median value of men’s total pension wealth was nearly twice as high as women’s in the period July 2012 to June 2014, £76,900 compared with £41,700 (excluding those with zero pension wealth).

Membership and, to a lesser extent, wealth is broadly comparable for men and women in younger age groups (under 45 years old). However, sex differences for median pension wealth and for the percentages with pension wealth become more pronounced with age, with larger values for men in both cases. There are a number of possible reasons for the greater equality in younger people. Firstly, a much lower proportion of people in this age group have some pension wealth and those that do are more likely to be in similar employment situations1, regardless of sex. Secondly, many men and women may have differing career paths as their working lives progress, influenced, for example, by caring responsibilities.

Notes for total wealth held in private pensions (individual level)

1. As highlighted in the ONS Labour Force Survey publication, Women in the Workplace.

In the following section, all sources of private pension wealth, current (including additional voluntary contributions or AVCs), retained and pensions in payment, are drawn together. This allows us to summarise the level of total private pension wealth among households in Great Britain by certain main characteristics.

The remainder of this section presents estimates of wealth held in private pensions at the household rather than at the individual level. In the following tables, household wealth has been calculated as the sum of private pension wealth across all adults within the household.

In many cases, couples may have made joint provision for retirement. For example, if one partner worked and the other did not, the working partner may have contributed additional amounts to his or her private pension to ensure the final income would be sufficient to support both partners during retirement. Therefore, it makes sense to examine total private pension wealth held by all members of the household to supplement the individual picture presented elsewhere in this chapter. This approach also allows us to make comparisons at the household level between pension wealth and the other types of wealth which are reported at household level (see Total Wealth chapter).

Table 6.11 allows us to compare the median total pension wealth of all households in each period in Great Britain (including those with no private pension wealth) with the median of only those households that have some private pension wealth. Similar comparisons of upper and lower quartile values are also included and, for each type of pension wealth, median and quartile data are presented (excluding those with no pension wealth of the type in question).

About a quarter of households in Great Britain had no private pension wealth in July 2012 to June 2014 and in the two earlier periods, highlighting an unequal distribution of private pension wealth, which is explored further in Figure 6.13.

The median private pension wealth of all households was higher in July 2012 to June 2014 than it was in the period July 2010 to June 2012 (£47,100 compared to £39,700). The median private pension wealth in households excluding those with no such wealth was higher in the period July 2012 to June 2014 than in the period July 2010 to June 2012 (£97,300 compared to £80,500). The driver behind this increase is changing financial parameters used to value pension in payment wealth and DB wealth, both current and retained (see quality assurance section).

In July 2012 to June 2014, 28% of households in Great Britain had wealth in current occupational DB pensions (median wealth £85,500) with a similar proportion (30%) having pensions in payment (median wealth £146,900).

Aggregate household private pension wealth

Table 6.12 presents a breakdown of aggregate private pension wealth of households in Great Britain by the overall components discussed in the previous sections.

Table 6.12 shows that aggregate private pension wealth in Great Britain increased from £2.9 trillion in the period July 2006 to June 2008 to £4.5 trillion in the July 2012 to June 2014 period. The increase is mainly explained by the increases in pension in payment wealth and current pension wealth. In July 2012 to June 2014, 12% of the aggregate private pension wealth related to retained pension wealth with 49% due to pension in payment wealth and 39% due to current pension wealth.

The equivalent percentages in the period July 2010 to June 2012 were, respectively, 13%, 47% and 40%. Following the finding from Table 6.11, that 76% of households had some pension wealth in July 2012 to June 2014, it is interesting to examine further the distribution of aggregate pension wealth among those 76%. This is illustrated in Figure 6.13 below.

Figure 6.13 shows aggregate private pension wealth, for those households with some private pension wealth, broken down into deciles, each of which represents 10% of the households that have private pension wealth. These households are sorted in ascending order of aggregate private pension wealth.

In the period July 2012 to June 2014, out of all households with private pension wealth, the decile with the most wealth (tenth decile) had almost half (47%) of the total aggregate pension wealth, while the five deciles with the lowest aggregates had 8% of the total pension wealth. The total aggregate wealth of the top decile was £2,094 billion compared to less than £6 billion for the bottom decile.

The composition of aggregate pension wealth among households with any private pension wealth varied by decile, with pension in payment wealth making up an increasingly higher proportion as the deciles move from the first towards the tenth. The percentage contributions of current and retained pension wealth to aggregate private pension wealth decreased in the higher deciles. For households in the top decile in the period July 2012 to June 2014, over half (55%) of private pension wealth was pension in payment wealth, 34% was current pension wealth and 10% was retained pension wealth. The equivalent figures for households in the lowest decile were 18% pension in payment wealth, 50% current pension wealth and 33% retained pension wealth.

Household characteristics

Distribution of pension wealth by household income

Figure 6.14 shows aggregate private pension wealth, of all households, broken down into deciles of equivalised household income, each of which represents 10% of the households in Great Britain. These households are sorted in ascending order of equivalised household income.

In the period July 2012 to June 2014, the tenth equivalised income decile had 31% of the aggregate pension wealth, while the five income deciles with the lowest aggregate private pension’s wealth had less than a fifth (17%) of the total aggregate pension wealth. The total aggregate private pension wealth of the top income decile was £1,376 billion compared to less than £70 billion for the bottom decile.

The composition of aggregate pension wealth among households with some private pension wealth varied by income decile, with current pension wealth making up an increasingly higher proportion in the deciles with higher income. The percentage contributions of the other two pension wealth types (pensions in payment and retained) to total aggregate private pension wealth both fell with increasing income deciles. For households in the top income decile in the period July 2012 to June 2014, about half (49%) of private pension wealth was pension in payment wealth, 41% was current pension in payment wealth and 10% was retained pension wealth. The equivalent figures for households in the lowest decile were 22% pension in payment wealth, 18% current pension wealth and 60% retained pension wealth.

Distribution of private pension wealth by region

Table 6.15 includes information about the proportions of households with private pension wealth across the regions and nations of Great Britain and measures of the median wealth of the households that have such wealth.

Table 6.15 shows that, in the period July 2012 to June 2014, 76% of households in England had at least some private pension wealth, a slightly lower proportion than in Wales (77%) and slightly higher than Scotland (73%). The median private pension wealth of these households during the period was lower in England (£95,700) than in Wales (£99,400) or Scotland (£111,600). The median in England was higher than in the other two nations in the previous years (see reference tables).

In the periods July 2010 to June 2012 and July 2012 to June 2014, the region with the highest proportion of households with private pension wealth was the South East (83% in July 2012 to June 2014), in the same period proportion the South West proportion was also 83%. London had the lowest proportion of households with any private pension wealth (67% in July 2012 to June 2014), followed by the North East (73%).

The South East and the South West were the regions with the highest median household pension wealth in the period July 2012 to June 2014, with the East Midlands having the lowest median household wealth. In each period, the household medians varied less between regions than between different pensions types (see Table 6.11) or between household types (see Table 6.16).

Distribution of household private pension wealth by household type

Table 6.16 shows the distribution of household private pension wealth for different types of household and the percentage of households in each category with at least some private pension wealth.

As Table 6.16 shows, in the period July 2012 to June 2014 the types of household with the highest proportions having wealth in private pensions were those with a married or cohabiting couple and nondependent children only (91%) and those with a married or cohabiting couple, one of which was over and one of which was under 60/651, with no children (91%).

The percentages for the household types with married or cohabiting individuals had collectively the highest number of people with pension wealth (79 – 91%). This was a similar pattern across the periods. The only household type where less than half of households had some pension wealth, in each period, was those with a lone parent and one or more dependent children (47% in July 2012 to June 2014).

Excluding households with no private pension wealth, the highest median private pension wealth in both periods were in households with couples with no children in which one individual was aged under and the other over 60/65. Households with a lone parent and one or more dependent children had the lowest median private pension wealth in each period, with values in each case that were less than 10% of the highest median.

Notes for total wealth held in private pensions (households)

1. 60/65 refers to women up to the age of 60 and men up to the age of 65.

The following section compares some of the estimates of pension membership in WAS with those available from other data sources. Sensitivity analysis is presented that allows a better understanding of a main contributing factor to the increases seen in some aspects of pension wealth: annuity rates.

Comparison with other data sources

Data sources such as the Annual Survey of Hours and Earnings (ASHE) and the Occupational Pension Schemes Survey (OPSS) have displayed large increases in current pension membership of employees for the year 2014. The most likely explanation is the gradual roll out of automatic enrolment that began in October 2012. It may therefore, be expected that similar increases in membership of current pensions should also be observed in WAS. However, only modest increases in pension membership were observed from the period July 2010 to June 2012, to the period July 2012 to June 2014 in WAS. This may be explained by two main differences between the data sources. Firstly, WAS covers a 2 year period and from July to June and data is collected in each month of that period. The most recent WAS data from the period July 2012 to June 2014 contains 3 months of data prior to the introduction of automatic enrolment. Furthermore, the effect of the gradual roll out is spread over the rest of the data collection period. ASHE and OPSS are conducted annually and refer to particular reference dates in April of the year of collection. Therefore data from ASHE and OPSS in 2013 fully capture the first 6 months of automatic enrolment.

The second difference is that WAS contains information on all people whether they are employees or not, for instance, self-employed individuals. ASHE and OPSS only capture those who have a pension scheme through their employment hence only captures data on workplace pensions of employees. Considering that Automatic Enrolment is targeted at employees, this may explain the larger increases in pension participation of the relevant populations between WAS (all adults) and ASHE and OPSS (employees).

The Family Resources Survey (FRS) is a more comparable data source because, like WAS, the FRS includes pension membership of the self-employed and the economically inactive. Comparison of the change in percentage of individuals with current pensions in WAS from the period July 2010 to June 2012, to the period July 2012 to June 2014 with the changes in the FRS over a similar period found that the sources were in broad agreement. The comparison is complicated by other differences between the sources, for example, the FRS data is collected every year and presented on a financial year basis, whilst a data collection for WAS is over a two year period. Table 6.17 shows current pension membership as estimated for WAS and the FRS.

The data from the FRS showed that pension membership increased following the introduction of automatic enrolment (2012-13). There was also an increase between the period July 2010 to June 2012 and the period July 2012 and June 2014 in the Wealth and Assets Survey but it was less pronounced. As the FRS is conducted annually, the increase in pension membership is more clearly seen than for WAS where an increase is averaged over 2 years. Overall, the FRS data indicates a similar pattern of pension membership to the one found within the Wealth in Great Britain analysis.

In this section the period July 2010 to June 2012 will be referred to as Wave 3 and the period July 2012 to June 2014 is referred to as Wave 4. The use of this notation simplifies the technical discussion in the following section.

The Wealth and Assets Survey is unique in that it allows comparison of wealth in Defined Contribution (DC) pensions with wealth in Defined Benefit (DB) pensions. To achieve this comparison, a methodology has been developed to convert DB pension wealth into a DC pension equivalent (see Chapter 7 Wave 2 for more details).

Historically, the money in a DC pension pot was used to buy an annuity on retirement (an annual income until death). The amount of annual income that can be bought with a certain pot size can be calculated using annuity rates¹. In WAS the size of the DC pension pot (wealth) is provided directly by the respondent.

For DB pensions, WAS collects information on the annual income that an individual can expect upon retiring and receiving their pension. Using this expected income value, we can estimate the equivalent DC wealth that an individual would need to buy an equivalent income using annuity rates and other data such as sex and age at retirement.

The annuity rates that are used in the calculation of DB wealth are collected from financial markets and change over time. The higher the annuity rate the greater the annual retirement income that can be bought with a fixed amount of DC pension wealth (pension pot). If annuity rates increase then the DB wealth calculated using those annuity rates will decrease. The opposite is also true, if annuity rates decrease, then the DB wealth calculated using those annuity rates will increase. Annuity rates have generally decreased between Wave 3 and Wave 4 (see Table 6.19) and as a consequence we would expect to see DB wealth increase, which was observed. The methodology was designed to reflect that, with a fall in annuity rates, the relative value of a DB pension would increase compared to equivalent DC wealth.

It is of interest to try to isolate the effect of annuity rates and other financial assumptions so that underlying trends in the data may be observed. To get an estimate of the change in DB pension wealth, excluding the effect of changing annuity rates and discount rates, a sensitivity analysis was conducted. In the calculations of DB pension wealth, annuity rates are converted to annuity factors (the inverse of the annuity rate). In the analysis, annuity factors and discount rates were fixed at average Wave 3 values. These average annuity factors and discount rates were applied to Wave 3 and Wave 4 data at record level. DB wealth was then recalculated for Wave 3 and Wave 4. Table 6.18 shows the median current DB wealth with fixed annuity factors and as presented in Table 6.2.

Table 6.18 shows that the change observed between Wave 3 and 4 with fixed annuity factors (change B) is small compared to the change in wealth observed when the original annuity factors (change A). Compared to original analysis, fixing the annuity rates at average levels for Wave 3 also led to a drop in DB pension wealth (change C).

To estimate the change caused by varying annuity rates between Wave 3 and Wave 4 we can subtract the change in DB wealth with fixed annuity rates (change B and change C) from the original increase in DB wealth (change A).

Table 6.18 indicates that the overall median of DB wealth attributed to change in the annuity factors (rather than a change in respondent data) was £14,400. The estimate of DB wealth attributable to changes in annuity factors and discount rates was larger for men than for women (£21,900 and £12,900, respectively).

DB pension wealth is not the only type of pension wealth that uses annuity factors in its calculation. Pension in payment wealth (see section pension in payment) is also calculated using annuity factors². It should be noted that the annuity rates used to calculate the Pensions in Payment are assigned differently from those used to calculate DB wealth. As annuity rates have decreased (meaning annuity factors have increased) from Wave 3 to Wave 4, we would expect to see an increase in the valuation of pensions in payment wealth that was observed.

Table 6.19 shows the change in average annuity factors from Wave 3 to Wave 4 for those who have some of the corresponding pension wealth. There has been an increase in average annuity factors for both pensions in payment and DB wealth. The increase in annuity factors between Wave 3 and Wave 4 was larger for men than women in both Pensions in Payment (3.8 compared to 1.7) and DB wealth (5.4 compared to 4.0). This may help explain why the increases in pension wealth were larger for men than women in Pensions in Payment and DB wealth (see reference table).

So why have annuity rates changed?

There are likely to be many factors contributing to changes annuity rates. One such factor is likely to be that a change to regulations around annuity rates has occurred since Wave 3. The legal change has meant that men and women must be offered the same rates (unisex rates) which should be reflected in the Wave 4 data. Annuity rates have decreased from Wave 3 to Wave 4 for both men and women, increasing DB and pension in payment wealth as calculated in WAS. The change to unisex annuity rates was predicted to affect men and women differently³ and may help explain why there has not been an equal increase in pensions in payment wealth and DB pension wealth for men and women (see reference table).

Notes for sensitivity analysis and the effect of annuity rates on Defined Benefit and Pension in Payment wealth

1. An annuity rate converts a sum of money to a regular income over a defined period (often until death). An individual with a pension pot of £100,000 offered an annuity rate of 0.05 could get £5000 a year as an annual income for the duration of the agreed period.

2. Discussion of pension methodology is not warranted here, for an overview of how pension in payment and DB wealth are calculated see Wealth in Britain, Wave 2 Chapter 5.

3. An analysis of unisex annuity rates, pension policy institute, 2004.