Following a period of consultation, the financial assumptions used in the calculation of some parts of pension wealth were updated at wave two to reflect market developments. Since these assumptions have a significant impact on the valuation of wealth, the change has made it difficult to interpret movements in wealth between waves.
This is because changes in wealth resulting from updating financial assumptions represent revaluations, rather than real changes in the amount of benefits accrued by members.
This means that, in practice, individuals would not have ‘felt’ any wealthier (or poorer) in terms of their pension (apart from the effect of having contributed for up to two more years). The rights that they had accrued would still deliver the same expected income as defined in the scheme rules.The following sections explain the methodology used and the impact of changing the assumptions.
Nine separate components of private pension wealth were calculated based on the WAS survey responses. There were four categories of pension to which respondents were making contributions at the time of the survey (referred to as current pensions):
1. Occupational defined benefit (DB)
2. Additional Voluntary Contributions (AVCs)
3. Occupational defined contribution (DC)
4. Personal pensions (including group personal/stakeholder pensions)
The distinction between occupational DC pensions and personal pensions is as reported by the respondent. So, for example, if an individual had a personal pension facilitated by their employer and chose to report that as an ‘employer-provided/occupational scheme’, this would be counted as category 3. Conversely, if an individual reported this as a personal pension, it would be included in category 4. In addition to the four categories of current pension wealth, five other types of pension wealth were also calculated:
5. Retained rights in DB-type schemes
6. Retained rights in DC-type schemes
7. Pension funds from which the individual is taking income drawdown
8. Pensions expected in future from a former spouse/partner
9. Pensions in payment (or receipt)
The way in which pension wealth was calculated in categories one to seven was dictated by whether the pension was defined benefit (DB) or defined contribution (DC). (Definitions of DB and DC pensions are given in the Concepts and definitions section in the chapter on pension wealth.)
For DB type pensions (categories one and five) modelling was required to estimate wealth, as the future flow of income promised in these pensions needed to be converted into a present day stock value. Using methodology developed by the Institute for Fiscal Studies (IFS), the modelling approach taken was to estimate a market value for the benefits accrued. In other words, the modelling attempted to establish the pot of money that would have to be given to members at the date of the interview in order for them to buy the same benefits as those accrued in the pension.
In contrast, the wealth held in DC type pensions (categories two, three, four, six and seven) was much easier to ascertain as no modelling was required. The calculation of DC type pension wealth depended on the information collected about the value of the pension fund built up at the time of the interview.
For the remaining categories (eight and nine) modelling was also required to estimate wealth. The future income flows promised in these pensions were modelled in a similar manner to DB type pensions – i.e. a present value estimate was established based on market values.
Following a review, the underlying financial assumptions used in the modelled categories (one, five, eight and nine) have been updated at wave 2. Although this has made it harder to interpret changes in wealth between waves, it has maintained the comparability of DB and DC type pension wealth, as both are based on market values.
The Methodology section gives more detail on the method used for calculating each of the nine types of wealth identified above. The final section shows the impact of changing the financial assumptions for the relevant categories and on total pension wealth (individual level).
The rationale behind discounting is that a pound in the future is worth less than a pound in the present, not least because there is an opportunity to invest the money and earn a return. It should be noted that the discounting in the model only applies to the period between a member’s current and retirement age.
Discounting is not needed for the flow of income from retirement age onwards because the annuity factors (inverse of annuity rates) used in the model provide an equivalent to discounting for income flows in retirement. Using annuities in the valuation of current DB wealth allows for comparisons to be made between wealth held in current DB and DC pensions, as benefits paid out in retirement in the latter are also related to the price of annuities available at retirement.
The financial assumptions underlying the model in waves one and two are as follows:
Real discount rate: 2.5% (wave 1) and 1.8% (wave 2),
Annuity factors: the second best quoted RPI index-linked single life annuity rates as quoted on 19 December 2009 (wave 1) and 17 December 2011 (wave 2). (Annuity factors vary by age and sex and are therefore not presented here but are available on request).
Individuals could report up to two current defined benefit pensions. The wealth in each of these schemes was calculated separately (as described below) and then summed together to get total wealth in current defined benefit (DB) occupational schemes.
The calculation of current DB pension wealth consisted of the following three steps:
The annual pension income current DB members could expect to receive when they retired was calculated using a benefit formula that used reported information on (i) the number of years built up in the pension scheme (length of service), (ii) current gross salary and (iii) the scheme’s accrual rate (the fraction of salary accrued by the employee for each year of service that will form the basis of the annual pension at retirement).
The calculated flow of expected pension income in retirement was then converted into a stock value based on the pot of money that would be needed at retirement to purchase an annuity providing an equivalent income stream in retirement.
Finally, the pot of money needed at retirement to purchase the annuity (step 2) and any additional lump sums expected at retirement were discounted (see Discounting and financial assumptions) to show current DB wealth in present value terms at the individual’s current age (age at time of interview).
The model is summarised by this equation:
Where:
A_{R} is the age- and sex-specific annuity factor at retirement age, R, based on (single life) inflation-linked annuity rates. In other words, it is the pot of money required to deliver a price indexed income stream of £1 per year every year from age R to death.
Y_{i} ^{P} is annual pension income defined as Y_{i} ^{P} = x_{i}n_{i}s_{i} _{ }
x_{i} is the accrual rate in the individual’s scheme
n_{i} is the individual’s tenure in the scheme (length of service)
s_{i} is the individual’s gross pay at the time of interview
L_{i} is the lump sum that the individual expects to receive at retirement
i is the individual member
r is the real investment return or discount rate (based on AA corporate bond yields)
R is the earliest retirement age in the pension scheme
a is the individual’s age at time of interview
An important element of the calculation was the age at which pension income started being paid (‘retirement age’). Ideally the normal pension age of the DB pension scheme to which the interviewee belonged should have been used for ‘retirement age’.
Unfortunately, this information was not collected (questions on normal pension age will be asked as from Wave 4 onwards). Nevertheless, information on the earliest age at which a current member could retire and take a pension was collected, and this was used as a substitute.
In a number of cases in both waves, the earliest retirement age reported by an interviewee was lower than their current age. In such cases the earliest retirement age was amended to be equal to the person’s current age in order to avoid negative discounting in the model.
The model only measures the wealth held by members based on the benefits they have accrued in the pension up to the date of the interview. It therefore ignores the possibility that members might continue to make contributions to their scheme and thus accrue more benefits before they reach retirement age.
If all else stays constant, this would mean that members would have higher wealth. In addition to this, the benefit formula in the model makes the assumption that all of the benefits accrued by current DB members were in final salary schemes.
Although the vast majority of DB members are in such schemes, according to the Occupational Pension Schemes Survey (OPSS), a considerable number of people in DB schemes now belong to schemes that calculate pension benefits on the basis of earnings over the whole career, known as Career Average Re-valued Earnings (CARE) schemes.
Because the model provides individual not household pension wealth measures and due to the complexity of the calculations and the information that would have been required from respondents, survivor benefits (benefits to dependents on death of a pension scheme member) are not taken into account.
In practice this would lead to underreporting of pension wealth, particularly for women, because on average women live longer than men. These survivor benefits will be sometime in the future for most women, so their omission will have only a small effect on the calculations.
Individuals who reported being members of an occupational DB scheme were asked whether they had made any AVCs and, if so, what the value at the time of interview of their AVC fund was. Current AVC wealth is simply defined as the fund value reported by the respondent at the time of the interview.
Current occupational DC pension wealth is simply defined as the fund value reported by the respondent at the time of the interview. Individuals were asked the value of their fund and were encouraged to consult recent statements where available. They could report up to two current occupational defined contribution pensions.
The wealth in each of these schemes was calculated separately and then summed together to get total wealth in current defined contribution (DC) occupational schemes. This procedure was also followed for those who reported that their employer-provided scheme was a hybrid scheme or that they did not know the type of scheme.
Personal pensions may include stakeholder and self-invested personal pensions, whether on a group or individual basis.
Current personal pension wealth is simply defined as the fund value reported by the respondent at the time of the interview. Individuals were asked to report the value of their fund at the time of the interview and were encouraged to consult recent statements where available. They could report up to two current personal pensions in wave 1and four in wave 2. The wealth in each of these schemes was calculated separately and then summed together to get total wealth in personal pensions.
Retained pensions are pensions to which individuals have stopped contributing but from which they are not yet drawing an income. Individuals could report up to three pensions in which rights have been retained in wave 1 and six in wave 2. These could have been either DB or DC schemes. The wealth in each DB retained scheme was calculated separately and then summed together to get total wealth held as retained rights in defined benefit (DB) occupational schemes.
The wealth held in retained DB pensions was modelled in a similar way to current DB pensions. However, in retained DB pensions expected pension income in retirement was calculated using self reported estimates. This is in contrast to the three-component benefit formula used for current DB wealth, where expected pension income was calculated as a combination of current gross salary, length of service and the accrual rate. Also, for retained DB pensions, retirement age is assumed to be 65, or the individual’s current age if over 65.
The model is summarised by this equation:
A_{R} is the age- and sex-specific annuity factor at retirement age, R, based on (single life) inflation-linked annuity rates. In other words, it is the pot of wealth required to deliver a price indexed income stream of £1 per year every year from age R to death.
Y_{i} ^{P} is self-reported expected annual pension
L_{i} is the lump sum that the individual expects to receive at retirement
i is the individual member
r is the real investment return or discount rate (based on AA corporate bond yields)
R is assumed to be 65, or the individual’s current age if he/she was already aged over 65
a is the individual’s age at time of interview
The financial assumptions underlying the model are the same as category one.
Individuals were asked to report the value (at the time of interview) of any retained DC pension funds. The wealth in each of their DC retained pensions was calculated separately and then added together to get total wealth held as retained rights in such pensions.
Drawdown is where people take income from the pension fund but the fund remains invested. Individuals could report that they were already drawing down assets from a retained pension scheme. In these cases, individuals were asked to report what the remaining fund value for their scheme was at the time of interview. Their wealth in each of these schemes was then added together to get their total wealth held in such schemes.
Individuals were asked to report in total how much they expected to get in future from private pensions from a former spouse or partner. Respondents were given the choice to report this either as a lump sum wealth figure, or as an expected annual income. Two slightly different approaches were followed, depending on how the respondent answered.
For those who reported a total lump sum value, this figure was taken as the relevant wealth measure and discounted back to the time of the interview. For those who reported an expected future annual income, wealth was calculated in a similar manner as for DB schemes described in category one.
A_{R} is the age and sex specific annuity factor at retirement age, R (see above)
Y_{i}P is expected annual pension
r is the real investment return (based on AA corporate bond yields)
R is assumed to be 65, or the individual’s current age if he/she was already aged over 65
a is the individual’s age at interview
The financial assumptions underlying the model are the same as category one.
The value of the future flow of income provided by pensions from which individuals are already receiving an income is calculated as the pot of money that would be needed at the time of the interview to buy that future income stream from a pension provider.
A_{a} is the age and sex specific annuity factor based on respondent’s current age, a
Y^{P} is reported current annual private pension income
There is no need for a discount rate in the model because the annuity factors provide an equivalent to discounting for income flows in retirement (see Discounting and financial assumptions). As interviewees are already receiving their pensions, there is no gap between the time of interview and the time when the pension will start to be paid.
For those age groups for whom no market annuity price was available (ages 75 and over), a hypothetical annuity price was predicted based on the information from those ages where annuity prices were available.
The same annuity factors have been used in the model as those in category one.
The discount rate and/or annuity factors used in the calculation of categories that required modelling (one, five, eight and nine) were updated at wave 2 to reflect market developments. Since these assumptions have a significant impact on the valuation of wealth, the changes in assumptions have made it difficult to interpret changes in wealth between waves.
This is because changes in wealth resulting from updating financial assumptions represent revaluations, rather than real changes in the amount of benefits accrued by members.
The impact that changes in financial assumptions have had on the modelled categories is presented in Table 1:
2006/08 | 2008/10 (old assumptions) | 2008/10 (new assumptions) | ||||
---|---|---|---|---|---|---|
Mean | Median | Mean | Median | Mean | Median | |
Current DB pension wealth | 159,900 | 82,100 | 159,700 | 82,700 | 188,700 | 100,400 |
Retained DB pension wealth | 146,300 | 46,400 | 123,900 | 45,800 | 148,900 | 55,000 |
Pensions in payment wealth | 196,800 | 79,400 | 215,200 | 83,000 | 238,300 | 90,500 |
Total pension wealth | 144,300 | 50,400 | 156,700 | 57,800 | 176,900 | 64,400 |
Table 1 shows estimates of the relevant categories for 2008/10 using the new and old (wave 1) financial assumptions, alongside those for 2006/08 (wave 1). (The sample size for category eight - spouse pensions - is too small to produce robust estimates). Estimates for total pension wealth are also shown. These estimates are at individual level. Appendix 1 of Chapter 2 includes estimates at the household level.
Updating the financial assumptions at wave 2 pushed up wealth in all of the modelled categories shown. For example, for current DB wealth, almost all of the increase in wealth observed between 2006/08 and 2008/10 could be explained by the use of new assumptions (96 per cent of the increase in the median wealth estimate). For total pension wealth, 47 per cent of the increase in the median wealth estimate could be explained by the use of the new financial assumptions.
The change in the discount rate, which is linked to AA corporate bond yields, had an upward effect on wealth. The lower discount rate used in the modelling increased wealth as this meant more money needed to be set aside by members in the present to ensure that enough money was available at retirement to purchase their accrued benefits.
The change in annuity factors, which are the inverse of annuity rates, also had upward effect. The higher annuity factors used increased wealth as it meant that a larger pot of money was needed to purchase the same ‘income flow’ in retirement as that which has been accrued.
It should be noted that estimates for categories one and five have been affected by both changes, in contrast to category eight, which was affected only by higher annuity factors.
Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: media.relations@ons.gsi.gov.uk
The United Kingdom Statistics Authority has designated these statistics as National Statistics, in accordance with the Statistics and Registration Service Act 2007 and signifying compliance with the Code of Practice for Official Statistics.
Designation can be broadly interpreted to mean that the statistics:
Once statistics have been designated as National Statistics it is a statutory requirement that the Code of Practice shall continue to be observed.