Should I buy an annuity? September 2025 edition
- Noel Watson CFPᵀᴹ - Chartered Wealth Manager
- 3 days ago
- 10 min read
Introduction
With annuity rates reaching levels not seen in many years, some approaching retirement may be wondering whether an annuity should form part of their retirement income, helping to protect against sequence and longevity risk. We evaluate the sustainability impact of using an annuity as part of your retirement income, as well as the key considerations to take into account.
Disclaimer
Retirement planning is tailored to each individual and involves determining whether to purchase an annuity.
In this blog, we deliberately ignore many real-world variables, including health issues, which could affect the annuity rate offered.
Overview
One ray of light in the continuing gloom in UK PLC finances, in particular soaring gilt yields, is that annuity rates have been increasing in recent years. For example, for a 60 year old purchasing an inflation-linked (RPI) annuity, the rate has increased by 18% from £3,852 to £4,547 since February 2024, according to Hargreaves Lansdown research.
*Hargreaves Lansdown is one of many "best buy" tables that are available, and it's not our recommendation to use this particular one! Seek advice about your specific circumstances.
Data from the SharingPensions website suggests that for a 65-year-old purchasing a level (non inflation-adjusted) annuity, rates haven't been this high since 2008!
How might the increase in annuity rates change the decision-making process of those approaching retirement? Let's introduce our case study.
Billy and Jean King
Billy and Jean King are both soon turning 60 and are considering retiring from work in the near future. They have a total retirement pot of £1,150,000 (£575,000 each in pensions) to fund their retirement. They want to enjoy a comfortable retirement, and they estimate this will cost them £60,600 (net of taxes) per year (which coincidentally aligns with research from Pension UK on retirement living standards!).
Iteration Zero: Calibration
Iteration Zero is effectively a Timeline calibration exercise to ensure the software is set up correctly. It also helps readers relate this article to others on the site, for example, the 4% rule analysis. The following parameters for this iteration are therefore closely related to other research and differ from subsequent iterations (which reflect the Kings' scenario.
A £1,000,000 starting pot.
30-year retirement horizon.
Portfolio with 50% developed market (Timeline displays this as Global) equities and 50% global bonds with zero fees.
Expenditure increases in line with inflation.
No other income sources.
No taxation.
For this baseline scenario, we can see that our safe withdrawal rate (SWR) is 3.35% which aligns with our previous research.
Iteration One: Real world
Now that we've confirmed our baseline is correct, we can adjust the model to reflect real-world retirements better.
Both Billy and Jean will receive a full state pension.
Their expenditure will increase by 1% less than the rate of inflation each year.
Total costs are assumed to be 2% a year.
Billy and Jean are taking on the "nastiest, hardest problem in finance" on a DIY basis. The 2% annual cost reflects the drag on gross investment returns due to fund and platform fees, as well as potential investor misbehaviour, which the latest Morningstar "Mind The Gap" research estimates to be 1.2% per year (measured over the last 10 years). Given that markets have been relatively benign for the last decade, we wonder if the Morningstar number is artificially low.
The portfolio equity allocation is increased to 60%.
The plan length is increased to 40 years, which corresponds to a survival probability (the chance of either Billy or Jean being alive) of 10% or less (approximately age 100).
We can see that for this iteration, the plan is successful in 86% of scenarios.
Let's now examine the chance of the King's outliving their money. The odds are far more favourable than the 86% might suggest (which considers only the portfolio outcomes and doesn't incorporate longevity). For example, at age 100, the chance of one of the King's being alive and the portfolio running out is only 2%.
The Kings face a large range of potential outcomes in retirement:
Worst case (red line): The money runs out at age 75 (all that remains is their state pension income).
Median case (blue line): The investment balance at age 100 is approximately £540,000 (all numbers in real terms).
Best case (green line): The investment balance at age 100 is approximately £5,200,000.
We can see the cuts in income for the worst-case scenarios. For example, in the worst-case scenario, income is reduced from around £57,000 (recall that expenditure decreases by around 1% per year in real terms) to just under £24,000 (their state pensions) at age 75, as mentioned above.
While we said this example was "real world", there are still many simplifications/assumptions versus a real-world client scenario, including:
Taxation optimisation is ignored.
There is no consideration around gifting or leaving a legacy.
We are not planning for any inheritances.
We are ignoring potential care home fees.
We are ignoring the possibility of potential downsizing
There are no "big ticket" items planned.
Iteration Two: Annuity to the rescue?
The Kings are wondering how an annuity might impact the range of outcomes. As mentioned above, a single life, RPI-linked annuity pays £4,547 per £100,000. However, if they purchased two of these policies (one each) and one of them died, one of the annuity incomes would also cease. They would prefer joint policies with a 50% survivor annuity (meaning the household annuity income would only reduce by 25% if one of them died). We assume the joint life policy reduces the annuity by around 5% to approximately £4,300 per £100,000 compared to a single life policy. Billy and Jean each put £287,500 (50% of their pension) towards purchasing annuities, which gives them each an annuity income of £12,400. The remaining portfolio is invested identically to iteration one (a 60/40 portfolio).
At a high level, this iteration improves the outcome, with a 10% increase in success rate to 96%.
If we examine the range of outcomes, the worst-case scenario reduces by one year to 74. The reduction in income, should the worst-case scenarios occur, is relatively muted versus iteration one. In the worst case, the income falls from around £57,000 to £44,000 at age 74.
The median case is slightly better (£595,000 vs £540,000), while the best case is much reduced (£2,840,000 vs £5,190,000).
Iteration Three: Do we need to fear inflation?
The level annuity option appeals to the Kings, given its higher starting payout and the fact that they plan to reduce their expenditure in real terms as they age. They are each able to obtain a level annuity of £19,250, around 55% more than for the RPI-linked annuity (£12,400).
The answer is yes, we do need to fear inflation! The overall success rate versus iteration two drops from 96% to 72%.
The spread of outcomes has increased compared to iteration two. The worst case has the money running out two years earlier at 72. While the Kings still have the annuity income to rely on at this point, worst-case outcomes tend to coincide with periods of inflation in early retirement, meaning that the purchasing power of the level annuites may be significantly reduced at this point. For example, in the scenario for the 1973 retiree, the income significantly drops at age 75 from around £54,000 to £32,000.
The final balance for the median case outcome has decreased from £595,000 to £430,000, while the best-case outcome has improved to £3,860,000, up from £2,840,000.
Iteration Four: The advised alternative
There were two potential challenges with scenario one.
The drag on returns is 2% a year.
The equity portion of the portfolio was not particularly diversified, consisting solely of equities from developed markets.
Regarding challenge one, given the size of the King's portfolio (£1,150,000), it should be perfectly possible to hire a financial adviser who will charge a total of around 1% (for advice, funds, and platform). Hiring a financial adviser should, in theory, eliminate investor misbehaviour. There is, of course, no guarantee that this would be the case if Billy and Jean encounter some severe market turbulence in retirement (which, as pointed out above, we've not had for well over a decade) and decide to sell despite the adviser's best efforts.
For challenge two, we will now construct a more diversified portfolio consisting of 70% equities and 30% bonds, broken down as follows:
42% developed market equities (large and mid-cap).
14% developed small-cap value equities
14% emerging markets
30% bonds
As with our solution for challenge one, there is no guarantee that this more diversified portfolio will provide superior future outcomes; we can only look back. Furthermore, we assume that the client is comfortable with the potential increased volatility and tracking error that this portfolio may experience.
The success rate is now the highest of all scenarios at 97%.
The worst-case outcome has improved dramatically to age 79 (from 75 in scenario one), but it's worth emphasising that, as with iteration one, once the investment pot runs out, the Kings are left with just their state pension. The median outcome has increased to £2,570,000 while the best case is £11,440,000!
Iteration Five: Can the inflation-linked annuity improve iteration four?
We observed how the outcomes in iteration one improved when RPI-linked annuities were introduced in iteration two. What if we take iteration two and make the same changes (advised solution, more diversified portfolio) as we did from iterations three to four:
The drag on returns is reduced from 2% to 1.5% (I've assumed the overall advice costs are higher, given the smaller investment).
The portfolio is now the same as in iteration four.
At a high level, the success rate is unchanged at 97%. The worst case reduces by two years to 77. The median case balance has broadly halved from £2,570,000 to £1,290,000, and the best-case scenario is significantly reduced at £4,850,000 (from £11,440,000).
Iteration Six: Spending flexibility
If we revert to iteration four, we can see that with a success rate of 97%, the investments are exhausted in only a few historical scenarios.
But what if any failure is unacceptable? We can see that for the worst-case outcome, there were signs of trouble early in retirement, with the 1915 retiree seeing their investment balance falling dramatically at the outset of retirement.
Research indicates that returns in the first decade are a significant driver of plan success, and with this in mind, the Kings wonder what might happen if they were prepared to be flexible with their spending should an unfavourable scenario occur.
The Kings' current planned expenditure is £60,600 per year, increasing at a rate of 1% less than inflation.
If they were prepared to reduce their expenditure to £52,500 per year for the first decade, the worst-case scenario shows the money running out at age 87.
If we examine the chance of the Kings being alive when the money runs out, we can see that it is minimal!
Of course, this potential reduction in expenditure may be too much for the Kings. However, if we assume they are willing to tighten their belts should we have a very unfortunate scenario, there is also the upside that they may be able to increase spending should retirement outcomes be more favourable. For example, if we increase initial spending from £60,600 to £80,000 a year, in the median outcome, the money lasts all the way to age 97.
In the best case, they can start out spending £100,000 a year and still have a significant surplus at the end of the road!
Conclusion
Iterations 1-5 are summarised below. We have excluded iteration zero as it was purely for calibration purposes, and iteration six is more to show options around spending flexibility.
As we discovered in iteration three, for our case study, purchasing level annuties was suboptimal. The outcome for the inflation-linked annuity was more mixed, in our case, being influenced by the fees and asset allocation of the pensions it was "replacing".
When it comes to real-world planning, there are many things to consider:
The size of the portfolio relative to planned expenditure: This will influence when the worst-case outcome (when the investment pot is exhausted) could occur. Of course, given a sufficiently large investment pot, it may not be exhausted, leading to a surplus upon death.
Whether you are prepared to be flexible with your spending in the event of less favourable outcomes, particularly in early retirement. If you were to have a good outcome (see the final balance for the good scenario in iteration four!), there may be the possibility of significant increases in spending!
How much guaranteed income do you require to be comfortable? For example, you may want a guaranteed income to cover household running costs (such as food, lighting, and heating).
The estimated size of drag on investment returns: This could be due to many things;
The likelihood and magnitude of your potential misbehaviour, especially during challenging periods
The fees you are paying, which should be considered whether you are a DIY investor or an advised client,
The asset allocation of the investment portfolio. We observed that the RPI-linked annuity led to better outcomes in scenario two, but didn't improve matters in scenario five.
While our analysis covered the Kings at retirement, the annuity evaluation can be periodically reviewed as circumstances change, for example, if financial literacy declines.
About Pyrford Financial Planning
Pyrford Financial Planning is an Independent Financial Adviser based in Weybridge, Surrey.
We specialise in retirement planning and provide independent financial advice, including pension and investment advice, and inheritance tax planning.
We offer a no-obligation introductory meeting, which will be held over Zoom.
Our office telephone number is 01932 645150.
Our address is No. 5 The Heights, Weybridge, Surrey, KT13 0NY.
Please note: This blog is for general information only and does not constitute advice. The information is aimed at retail clients only.
Although best efforts are made to ensure all information is accurate, you should not rely on this blog for your personal situation or planning.
The value of your investment can go down as well as up, and you may not get back the full amount you invested. Past performance is not a reliable indicator of future performance.
About the author
Noel is passionate about helping clients plan for retirement, preparing and guiding them through this key life transition. He has written a book on retirement planning and regularly publishes retirement research on this blog.
Comments