In this blog, Jamie Funnell, chair of the CMI Annuities Committee, considers the effect of pension freedoms on annuities
It seems a lifetime ago that George Osborne, then Chancellor, announced the introduction of new ‘pension freedoms’. For anyone approaching their retirement, the end of compulsory annuitisation opened up access to a whole world of other options with which to flexibly use their accrued defined contribution pension wealth. Notwithstanding the potential for individuals to splurge on expensive cars in lieu of a well-thought-out drawdown strategy, this decision was bound to have implications for the individual annuity industry.
The CMI Annuities Committee has recently considered the impact of pension freedoms in its first combined analysis of standard pension annuities in payment and enhanced annuities (annuities where an uplift to payments is applied as a result of individual medical or lifestyle underwriting). We now have data for both product types, covering the periods before, during and after the implementation of this new retirement landscape. This allows us to compare data volumes and features, and to consider early indications of the impact of pension freedoms on the mortality experience of individual annuitants.
Prior to the April 2014 budget in which pension freedoms were announced, data volumes of standard individual annuities were falling, but the market for enhanced annuities was increasing year-on-year. Both annuity types saw a marked fall in annuity commencements from the announcement of pension freedoms. This may have been due initially to individuals delaying decisions about what to do with their pension savings until the implementation of the changes. However, low commencements continued after the implementation of freedoms in April 2015. For enhanced annuities, data volumes remained fairly stable in the following years, but commencements have continued to fall for standard annuities.
The age profile of those purchasing annuities is also of interest. Historically, annuity commencements have been heavily concentrated around traditional retirement ages of 60 (particularly for females) and 65 (the most common retirement age for males). Following the introduction of pension freedoms, these ages still seem to be the most popular, but there appears to be some signs of a shift towards annuities commencing at older ages for both enhanced and standard individual annuities. This is particularly noticeable for females, although this may relate to generational changes in retirement ages for women, due to equalisation of retirement ages with men, rather than being purely linked to pension freedoms.
The fall in annuity commencements, and early indications of a trend towards purchasing annuities at older ages, suggests that many individuals are now taking advantage of other options, but that annuities may appear more desirable at older ages. Where individuals do continue to value the security offered by an annuity, the data suggests that they may be more likely to seek an enhanced option now than prior to pension freedoms. Indeed, between 2016 and 2018, enhanced annuities accounted for 31% of individual annuity commencements, whereas between 2011 and 2013 they represented only 19% of commencements. The figures for earlier years may include a higher proportion of individuals who would qualify for an enhanced annuity but didn’t seek this option and the definition of enhanced annuities has changed over the years to encompass a wider range of health conditions. This means that the health of those taking out standard annuities in more recent years may be better than those who took out standard annuities in the past. It will be interesting to see whether this trend continues.
It is still early days to consider how the mortality experience of pension annuities might have been impacted by the change in profile and characteristics of individuals taking out annuities post-freedoms. We can only review post-freedoms mortality for annuities purchased in the last few years, and we know that historically there have been considerable differences between the level of deaths in annuitants who took out their annuity recently compared with those who took out their annuity many years ago. However, we did consider the emerging impact of pension freedoms with two different kinds of analysis.
The first analysis compared actual deaths with those expected under a mortality table for pensions annuitants and found that mortality experience was lower (ie fewer deaths) post-freedoms than pre-freedoms for both standard and enhanced annuities, with the exception of standard annuities for females, where mortality was higher post-freedoms.
However, the results of a simple generalised linear model (GLM) analysis suggested that annuities commencing post-freedoms experienced higher mortality than annuities commencing prior to the introduction of pension freedoms. This appears to contradict the one-way analysis of actual versus expected deaths. One possible explanation may be because the GLM analysis makes a separate allowance for the length of time since annuity purchase, meaning that this is stripped out of the comparison of pre- and post-freedoms mortality.
We are keen to revisit this analysis in the future with more data for annuities that have been in force for several years to better understand the impact of pension freedoms on mortality experience for pensions annuitants.
As time moves on, commencements of individual annuities may continue to be relatively low, compared with the period before pension freedoms, unless financial conditions, regulation or product innovation serve to make them more attractive to consumers. We expect trends, particularly around mortality experience, to become clearer in the coming years. This is an area that the Annuities Committee is keen to consider again in a follow-up analysis.
