What’s been happening and what’s on the horizon in the world of pensions
After a lifetime of saving, most Defined Contribution (DC) savers seek both income security and flexibility from their accumulated DC pots when targeting a desired standard of living in retirement. In this edition of Pensions Watch, we look at the considerable risks and challenges the decumulation decision-making process poses and how we might solve for these if more optimal retirement outcomes are to become the norm.
A brief retrospective
The fundamental basis of any pension system is to provide a secure long-term income stream to meet spending needs in retirement. This much is clear from both the OECD and the authoritative annual Mercer CFA Institute Global Pensions Index1. Prior to the introduction of the freedom and choice reforms in April 2015, this objective was largely met with over 90% of UK Defined Contribution (DC) pension pots being, mainly mandatorily, annuitised – usually net of 25% tax-free cash2.
However, for the past nine years there’s been complete flexibility as to how DC pension assets can be accessed3 from age 554, whether as cash; via income drawdown, from which an income (flexi-access drawdown) or a series of lump sums (uncrystallised fund pension lump sums – UFPLS) are drawn and a legacy may be left; via an annuity which will provide a guaranteed level, variable or escalating income for life; or a combination of each of these options. Indeed, unless people mix and match, ultimately they must trade off the income security and certainty of an annuity against the flexibility but uncertainty of taking one or a series of cash lump sums and/or an income via income drawdown.
One succinct way to think about decumulation today is as two tightly-interlinked problems: an investment one – how should I invest my money? – and a planning one – how much should I take from my pot and what pattern of income do I want or expect in the future? In the pre-freedom and choice era, the planning decision was determined in advance and so the investment problem became relatively easy – basically it was only an accumulation problem.
How have DC pots been accessed since April 2015?
Figure 1: Pots accessed for the first time since April 2015
Source: FCA Retirement Income Market Data 2022/23. 16/4/245.
Another visible trend has been the number of those tapping into, free-to-access, Pension Wise guidance9 and/or seeking regulated advice before making a decision. For drawdown purchase, whereas before freedom and choice about 90% of purchases were made with advice, since 2015 only around 60% have sought regulated advice, with just under 10% taking advantage of a Pension Wise call. For annuity purchase, the percentages for both seeking advice and engaging with Pension Wise have regularly been around 30% and for full encashments and UFPLS around 25% and 10%, respectively10.
A number of other points stand out from the data:
- In Q123
- 52% of new drawdown contracts were for amounts of £50K+11,with 70% of drawdown contacts being invested for those aged 55 to 64. 44% of this latter group made an initial drawdown of 8%+ of capital – a rate of withdrawal unlikely to be sustained for too long;
- 33% of new annuity contracts were taken out by those aged 55 to 64, while 47% of annuities were bought from the existing pension provider;
- 69% of encashments were made by those aged 55 to 64, with 10% of encashments – c.21K withdrawals – being for amounts of £50K+. Of these, over 1,500 were for amounts of £100K to £250K and 221 for £250K+. Prima facie, excepting those encashments taken as tax-free cash, a lot of income tax would have, probably unwittingly, been paid on these withdrawals by the latter two cohorts;
- Between April 2022 and March 2023, 8%+ withdrawals from drawdown were highly prevalent for pot sizes up to £100K, prevalent for pots sizes of £100K to £250K but less so for pots of £250K+, for which advice was typically sought. This suggests that many, far from planning a sustainable lifetime income stream, were instead adopting a myopic approach to satisfying their living standards, without giving too much thought to the long term.
Separately, the Association of British Insurers (ABI) announced a strong uptick in annuity sales during 2023, with the number and amount invested increasing significantly on 2022. Although the 72,200 contracts purchased at an average size of c.£72K, during what was a much higher yield environment than had prevailed previously, represents the highest number of annuities purchased since 2016, it still is still only 15% of the 466K peak achieved in 2009 and less than 40% of the number purchased in 2014, albeit with a slightly higher total value than that invested in annuities in 201412.
Decumulation decision making is difficult
Deciding how and when to start the transition from accumulating wealth into decumulation and retirement is one of the most challenging financial decisions an individual will make in their lifetime.
Robert Vaudrey, Copia Capital13
Given this, for those seeking both security and flexibility in decumulation, as freedom and choice intended, the decision-making process is incredibly complex, with myriad largely unquantifiable demographic, health, economic and investment risks to manage allied to a paucity of frames of reference, guidance and accessible advice. Consequently, most people do not know what is feasible or realistic at and in retirement. Couple this with retirement no longer being a one-off event with a well-defined destination point, having to contend with often unforeseen changed circumstances – whether family or financial – along the way and the considerable cognitive, or behavioural, impediments to informed decision making and you have the most difficult of life’s financial decisions to make.
The behavioural impediments to optimising decumulation decision making
The behavioural, or cognitive, impediments to optimising decumulation decision making are many and varied and typically compound with age.
Top of the list is the fact that moving from a deep-seated lifetime-of-saving mindset to a spending frame, by drawing on an exhaustible, non-replenishing pot of money, demands a fundamental psychological shift of mindset. Consequently, when given a choice of how to utilise their DC pension pots, many savers are reluctant to draw down their pension assets – many simply withdraw the income generated – thereby compromising their retirement living standards, often dramatically so16. Of course, at the other end of the spectrum, there are those who operate a myopic approach to retirement spending, where instant gratification trumps adopting a more measured, sustainable approach to consumption resulting in this cohort outliving their retirement savings, sometimes very prematurely17.So, whether outliving retirement savings or living in penury in fear of doing so, the risk of individuals sleepwalking into a suboptimal retirement is ever-present. Moreover, this decision making comes at a time in many people’s lives when financial literacy and cognitive ability typically starts to decline and is often accompanied by increased risk and loss aversion 18.Quite the problem then and unquestionably a growing policy issue.
Then there’s the degree to which saver choices are sensitive to the way in which options are presented to them. Framing, word choice and the use of numbers really do matter. Indeed, a seminal FCA study suggested that options worded in an income/consumption frame resulted in annuities being predominantly chosen. However, when worded in an investment frame, i.e. referencing likely investment returns with no mention of income/consumption, drawdown was mostly selected. Moreover, low/median income households, typically those who do not seek advice, were more sensitive to framing than higher income19.Never was Jean-Paul Sartre’s “words are loaded pistols” quote more prescient. As to numbers, single easily relatable numbers trump all else and are an incredible powerful tool to help anchor those who are metaphorically out at sea20.
Of course, the potentially dire consequences of making a wrong decision or suffering decision paralysis will compound over time as Defined Benefit (DB) pensions disappear, people start receiving their state pension ever later in life and increasingly become solely reliant on their DC pensions pots to support their desired standard of living in retirement which, for many, may extend to 30-plus years.
That said, it is also evident from the data, and not unreasonable to expect, that people want control over and ownership of their money. Indeed, it is well documented that people don’t like being told what to do, not least by those perceived as strangers (read policymakers), and especially when it comes to their hard-earned savings. That’s where cannily employing behavioural science comes in. More on that shortly.
Spending patterns in retirement
Before looking at how we might solve for the difficult decision making that comes with freedom and choice for those seeking income security and flexibility, let’s consider retirement spending patterns and how much income people need in retirement. First spending patterns.
Since the 1950s, economic theory has assumed that an individual’s spending remains reasonably constant throughout their lifetime, even when making the transition between work and retirement21.By contrast, the commonly held view of many in the pensions industry is that spending in retirement is ‘U-shaped’. That is, retirement is assumed to start with a spending spree in the ‘go-go’ years, decelerate in the ‘slow-go’ years, when the novelty of one’s new-found freedom dissipates and morbidity begins to materialise, only for it to skyrocket again in the ‘no-go’ years as the costs of long-term care surface22. Or so the story goes. Empirically, neither economic theory nor the U-shaped spending pattern stand up in practice, with studies around the world suggesting spending in retirement typically follows a downward trajectory in real terms regardless of income level, lifestyle and the period analysed with almost all cohorts progressively saving more of their income as they become older23. Intuitively, this reduced consumption and increased saving is likely to be driven by a combination of declining health, a desire to provide bequests and much changed personal preferences for goods and services in old age. Needless to say, as retirees enter the no-go years, long-term care needs often surface, though the idiom ‘long-term care’ is perhaps a misnomer, as the life expectancy of those who remain outside of care homes far exceeds that of those within24.
But here’s the thing. Most contemporary retirement spending studies have been conducted on those either born or who grew up during the Second World War, commonly referred to as the make do and mend and never let anything go to waste generation. By contrast, the generation below, the second wave baby boomers of the mid-1950s to mid-1960s and early Generation Xers of the mid-1960s to early-1970s, today’s 50 to 60-somethings, adopt a more aspirational approach to life. They also have greater disposable income than their parents, often have considerable DB pension rights and possess the ability to accelerate access to this stream of income via freedom and choice. Crucially, this “sandwich generation” – the first to be simultaneously caring for ageing parents while helping out financially dependent adult children (hence sandwich) and be faced with the rapidly declining economics of social care – may well be the generation whose retirement spending doesn’t decline in real terms but continually increases. Moreover, the sheer size of this demographic bulge presents a potential decumulation tsunami25.One thing looks increasingly certain though; the U-shaped retirement spending pattern will remain atypical.
So how much income do people need in retirement?
Now to income requirements and where better to turn than to the Pensions and Lifetime Savings Association’s (PLSA’s) Retirement Living Standards (RLS)26.Introduced in 2019, the PLSA’s RLS helpfully illustrate, via single relatable numbers, what life in retirement might look like for both a single retiree and a couple, either living in or outside of London, at three different levels of spending: minimum, moderate and comfortable.
Minimum
- £14,400 for a single person living outside of London (London £15,700)
- £22,400 for a couple living outside of London (London £24,500)
Moderate
- £31,300 for a single person living outside of London (London £32,800)
- £43,100 for a couple living outside of London (London £44,900)
Comfortable
- £43,100 for single person living outside of London (London £45,000)
- £59,000 for a couple living outside of London (London £61,200)
So how do we solve a problem like decumulation?
Recognising that decision-making at and in retirement has become more complex with considerable risks and scope for error, a greater focus on optimising decumulation outcomes is evident from both policymakers and regulators. Indeed, DWP has proposed placing trust-based schemes under a legal obligation to provide, or work with a partner in providing, decumulation products and services for members, while The Pensions Regulator (TPR) and the FCA have each set out their decumulation objectives and begun working in earnest with myriad stakeholders to accomplish these objectives29.
Something’s gotta change
Is the solution to provide more and better guidance?…
So to guidance, which is integral to TPR’s decumulation objectives in giving agency to savers to make their own decisions, via a guided journey that funnels people to tailored and personalised solutions.
Indeed, placing simple, accessible, trusted, timely, well signposted and behaviourally robust guidance – think rules of thumb, on-line expenditure calculators and drawdown modellers – at the centre of decumulation policy and delivering this in a consistent way, to support individual decision making, can lead to desired behaviours, such as overcoming the reluctance to draw down pension assets sustainably to support a realistic standard of living in retirement and addressing myopic behaviour at the other end of the spectrum. However, if guidance is to materially improve outcomes through regular engagement, the imperative must be to first define what a good outcome looks like and – given life’s twists and turns – to ensure the guidance is regularly revisited. Otherwise, an optimal retirement path is unlikely to be followed.
Guidance in the UK principally comprises two relatively well signposted free-to-access information sources in Pension Wise and the PLSA’s Retirement Living Standards (RLS). Pension Wise, the free-to-access telephone and face-to-face-based generic pensions guidance service for those aged 50+, noted earlier, has, by approaching an ultimately complex decision via logical and well framed questioning within a series of simple steps, helped many achieve better retirement outcomes31. However, one-off generic guidance from Pension Wise, typically at the point of retirement, is unlikely to be sufficient, given that peoples’ circumstances and spending patterns change throughout retirement, sometimes markedly so. Therefore, this aspect of guidance should be applied repeatedly to ensure an optimal path is continually followed. Meanwhile, the PLSA’s RLS, also considered above, helpfully illustrate what life in retirement might look like via simple, accessible and relatable single number anchors. Moreover, later in 2026 this guidance will be supplemented by the long-anticipated launch of the Money and Pensions Service (MaPS) pensions dashboard, which will allow individuals to view all of their pensions information, including the state pension, together in one secure place32.Of course, good guidance must always be simple and easily accessible, as even the smallest number of unnecessary steps or additional effort required by the prospective DC decumulator can derail the whole process. Once again, simple behavioural interventions have a central role to play here.
Employing simple behavioural interventions to guidance
As suggested earlier, framing, word choice and the use of numbers are critical to motivating beneficial behaviours. For instance, using positive words, or priming, and, as noted, employing single relatable numbers encourages positive behaviour. Equally, never underestimate the power of socialising good ideas and behaviours. After all, as social animals, our decisions are rarely made in a detached manner. Instead, we act on the information, opinions and actions of others, especially those we perceive as experts, and like to have our own actions and opinions validated by others. Ultimately, whenever employing simple behavioural interventions, never forget the principles of the EAST framework of making interventions Easy, Attractive, Social and Timely33.
Of course, much more can and should be done – perhaps along the comprehensive lines of Australia’s Retirement Income Covenant and the New Zealand Actuaries Retirement Income Interest Group’s (RIIG) Rules of Thumb34 – in helping those at and in retirement to gauge what is reasonable and realistic. Indeed, in the latter case, the idea is to help retirees to choose withdrawal rates in an informed manner, based on their retirement needs and preferences. That said, many commentators have suggested that no amount of guidance can substitute for advice. Therefore, the question of how to provide simple and affordable advice to those of a certain age, not least to those who are not tech-savvy, still needs to be answered.
Segmenting retirement income policy
As an adjunct to this, there is a strong case for segmenting retirement income policy between low earners, those on a middle income and high earners. In the UK, as the full state pension provides a high replacement rate for those on a below median working life income, any accumulated pension savings are best employed as a capital buffer for contingencies and to meet large one-off expenditures. At the other extreme, those with a high working life income are the least likely to require high replacement rates in retirement and be the most likely to seek regulated advice in exercising freedom and choice. However, those with a middle working life income will likely be the cohort to most rely on their DC pension pots to supplement the state pension if replacement rates, hence retirement living standards, are not to be unduly compromised. Therefore, it is this cohort on which UK decumulation policy should principally focus.
… Or does the answer lie in most being defaulted to a default?
Moving from one end of the choice architecture spectrum – the paternalistic requirement to annuitise DC pension pots pre-2015 – to the other – by introducing libertarian freedom and choice from 2015 – was, for many people, a step too far. Instead, a libertarian-paternalism halfway house which simplifies and guides choices, while still providing people with ownership and control, is perhaps a better way forward for most people. Soft defaults also play a role in TPR’s thinking.
Indeed, the contention of many prominent pension practitioners, commentators and academics is that most people don’t know what they don’t know. As such, they will never truly engage with the complex decisions to be made at and in retirement, nor will they ever have the confidence and capability to select and successfully manage the retirement solution that most closely meets their needs. Indeed, as noted earlier, even the simplest of calculations and financial concepts are unfathomable to many and few appreciate what is feasible and realistic to achieve at and in retirement.
So where does this leave us? Well, given the success of auto enrolment in the accumulation stage, the idea of auto enrolling people at the point of retirement into an institutionally-managed default that combines the central roles of collectivity, in the provision of income security and longevity insurance, allied to a degree of flexibility and the security of trust-based governance, has, for many years, gained traction. Indeed, for most people, given the behavioural and structural impediments to raising engagement levels, a well thought out default, which offers value for money, with options to finesse the default’s parameters and the provision for those, who are better able and willing to make their own decisions, to opt out with advice, is perhaps the best possible option. Of course, in allowing a degree of flexibility for those remaining in the default, which in itself would require more accessible guidance and low-cost advice, the default could be finessed at set times and within certain parameters to meet individual preferences. To prove sustainable, the extent to which each feature could be flexed would, in some cases, be constrained by the flexing of the other features, the individual’s age and the size of the remaining pot.
This default, which would likely be underpinned by a well-diversified managed fund with strong responsible investment credentials, could be complemented by a guaranteed minimum income underpin, financed by an element of the DC pot being invested in a level or escalating annuity. This could be further complemented by the purchase of one or a series of deferred annuities35,to provide a guaranteed income later in the retirement journey, perhaps in the no-go years when the costs of social care and leaving a bequest come into sharp focus36. Once again, the provision of advice is crucial here37.
Meanwhile, in Australia…
Australia, like the UK, continues to grapple with the thorny issue of how best to optimise retirement outcomes given unfettered access, initially from age 56, now from age 60, to DC superannuation pots. Unlike in the UK, annuitisation of superannuation pots has never been mandatory in Australia.
Could variable annuities work in the UK? Possibly. Having been launched in the UK in 2005, variable annuities steadily gained traction until the fallout from the Global Financial Crisis in 2008 unseated the cost to insurers of providing a lifetime income guarantee. Although their reintroduction to the UK market was widely anticipated on the surprise announcement, in April 2014, of freedom of access to DC pots from age 55, instead full and partial cash withdrawals and income drawdown took centre stage and continue to do so.
Therefore, if variable annuities are to emerge in the UK as a viable post-retirement option, there is work needed by insurers on product design to find that balance between marketability to consumers and ease of ongoing management. Likewise, regulators will need to play a role in creating the right incentives for insurers to offer a broad range of post-retirement solutions that offer genuine choice to consumers.
Why does this matter?
In a world of freedom and choice, the decumulation stage of the DC journey is fraught with difficulty. Currently, those at and in retirement, seeking flexibility and income security, through income drawdown, must successfully navigate myriad largely unquantifiable investment, economic and demographic risks. If not managed well, these risks can add up to an uncomfortable retirement at best or, worst case, lead to the retiree outliving their savings or living in penury in fear of the latter.
Indeed, the potentially dire consequences of making a wrong decision or decision paralysis, through lack of numeracy, financial literacy, relatable guidance and accessible advice, will continue to rise over time as DB pensions disappear, people start receiving their state pension ever later in life and increasingly become reliant on their DC pension pot to support their standard of living in retirement which, for many, might extend to 30-plus years.
If retirement is to be enjoyed and not endured, not least by an aspirational sandwich-generation, then the thought given by the pensions, asset management and adviser communities, with substantive input from regulators, to the design of what can be considered genuinely fit-for-purpose decumulation solutions, must be stepped up.
Indeed, whether a minimum, moderate or comfortable retirement in the UK becomes the norm, is largely contingent on timely and decisive action or continued inaction by both the pensions industry and policymakers. If policymakers, providers and practitioners can agree on appropriate defaults, a suite of simple, accessible and behaviourally robust guidance and low-cost advice more broadly, then it’s reasonable to expect that this could unlock better investment strategies and culminate in better retirement outcomes. However, timely intervention is needed if the potentially significant, if not catastrophic, economic and societal risks that could result from continued inaction are to be avoided. In short, the clock is ticking.