Which annuity is the best way to fund your retirement – a level annuity or an escalating annuity?
- A level annuity provides a fixed income that won’t change until the day you die. In real terms though its value is gradually lapped away by inflation’s sand-papery tongue.
- An escalating annuity (also known as an index-linked annuity) will offer a comparatively puny income today. But it will grow over time – its fortunes are hitched to the Retail Prices Index (RPI) or Consumer Prices Index (CPI).1
Jam today
When you compare the two types, it’s hard not to be seduced by the instant riches offered by the level annuity.
For example, I can currently bag a £20,800 level annuity2 for the same price as a £12,500 escalating RPI annuity. That’s over 66% more income for taking the fix.
That’s gotta be worth something, right?
And as it turns out it is.
Compelling research by Professor of Retirement Income, Wade Pfau, suggests that the most efficient asset allocation for retirees may well be a mix of equities and level annuities.3 In this scenario, any surplus cash generated by the level annuity over and above the retiree’s income needs is invested 100% into equities to create some upside.
So, continuing the example above, I can invest £8,3004 of my £20,800 level annuity income into the stock market (in year one of my retirement), given that I had intended getting by on the £12,500 a year offered by the escalating annuity.
As and when inflation erodes the level annuity’s real income below my minimum income floor, then the equity portfolio can be tapped for a top-up or to buy another annuity.
Equally, the equity portfolio can be a source of lifestyle income, emergency funds, or a legacy when the time comes.
Hold the jam, pass the spreadsheets
Let’s say I am 65. I took to my spreadsheets and worked out that I could maintain my income for 44 years until I age 109 using the Pfau strategy.
If I shuffled off at 84 then I would leave an inheritance of £120,000 in today’s money into the bargain. If I lived any longer, then I spend the lot on another level annuity to keep myself going.
£120,000 is a lot of buffer money and I got there assuming historically average levels of inflation and equity growth.5
Indeed the major flaw in my calculations is that I assume inflation and asset growth trot along smoothly at their historically average levels.
In reality that never happens – for better or worse – as the violent swings of the UK’s inflation history shows:
Source: Bank of England
A Monte Carlo sim would give me a better idea of the range of possibilities. In some inflation and growth rate scenarios I’d end up filthy rich. In others, filthy poor.
Hang on, I’m running out of jam
Where things really come unstuck for the level annuity though is when inflation makes like David Banner and bursts out of its corset in a big, green, income smashathon.
You can estimate the damage for yourself using a level vs escalating annuity calculator.
If inflation pootles along at 3%, the escalating annuity only pays out a higher annual income by the time you’re 83. You’ll have to hang on until age 97 for it to pay out a higher income overall.6
From then on you can die happy.
Right now, UK males live on average until 79 and females until 82. So you’ll need exceptionally youthful genes to make an escalating annuity worth your while. (Or the kind of bitterly tenacious grip on life that’s normally reserved for Dickensian crones with scores to settle.)
But, but, the tide can turn against the level annuity very quickly when inflation runs wild. If prices rise by 13% a year then the escalating annuity pays a higher income within five years. And it pays a higher total income in just eight – that’s shockingly fast.
So when did inflation last average 13% a year in the UK?
In the 1970s, peaking at 25% in 1975.
Ultimately, a level annuity offers more flexibility, growth, and value for money, but it does not offer certainty, security, or safety.
An escalating annuity is the superior product if those are your retirement goals, and frankly who doesn’t want some of that in their retirement?
Take it steady,
The Accumulator
- You decide as part of selecting your escalating annuity which inflation index to track, or you can choose a fixed number say 3% or 5% a year. [↩]
- 100% conventional: no dependents, no guarantees, no bells, whistles and big bass drums. [↩]
- Level annuities are known as fixed Single-Premium Immediate Annuities or “fixed SPIAs” in the US. [↩]
- I can actually invest £6,640 after tax. [↩]
- I assumed a consistent inflation rate of 3% p.a. and a real equity return of 5% p.a. The personal allowance is £10,000 (2014-15 rate) and uprated by 3% inflation p.a. My minimum required income was £12,000 after tax. My level annuity income was £20,800. When inflation forced my level annuity income below my minimum required income (in real terms) then I used the accumulated equity portfolio to buy another level annuity. This worked until age 104 when my final £5,000 went into cash and kept my head above water until I was 109. Even then I wasn’t out of money. I was just forced to live on less than £12,000 a year as the level annuity continued to grind down. Perhaps I’d sell my house at that stage. Or an antique kidney. [↩]
- I’m still using the £12,500 escalating annuity vs the £20,800 level annuity example. [↩]
I think this just illustrates that the “insurance” offered by annuities its just overpriced
On the way in, anyone on this site avoided pension fund fees by self managing
Why not take this to the logical conclusion by going into income drawdown on the way out?
Still I bet annuity rates will be looking a lot better in a couple of weeks after the correction in the gilt market, I was looking at the DMO website and it was showing yield to maturity of 2.8% on gilts maturing in September 2023
Thats up 1% on a month/two months ago maybe
Does an escalating RPI annuity normally have no cap? (I thought they commonly were RPI up to a maximum of 5%)
“its value is gradually lapped away by inflation’s sand-papery tongue”
ooo errrr! Fifty shades of price-rising dismay!
It’s also worth pointing out that it’s the rate of price increases of what you actually buy that’s important, not just “inflation” in general.
E.g. food and heating fuel have been racing ahead of everything else and they are a major proportion of most pensioners’ budgets. With an increasing world population, diminishing resources and unpredictable weather, might this trend continue? If it does you could already be half way to your 13% inflation figure.
And if the UK government loses control of the spiraling public debt problem then the pound will fall further, pushing up the price of these imported necessities.
Wait! you’ve made a common mistake! (It doesn’t affect your argument though.)
“Right now, UK males live on average until 79 and females until 82.”
This is not the right stat to use. This is the life expectancy _at birth_. People who are about to retire have managed to not die all the way up until 65, and should therefore live longer. (Otherwise, one’s 79th birthday would be rather sombre!)
Really you need to work out the (projected) probability of dying at each age and calculate your expectancy from that. (The ONS provide these stats: see the bottom of this comment.) You can also calculate the probability of reaching various ages.
However, a quick Google throws up this document which gives the numbers you should be using.
“UK life expectancy estimates at the age of 65 are 85.6 for women and 83 for men”
Possibly a relevant important stat is:
“At age 65, men in the UK can expect to live on average another 10.1 years in good health. Women can expect to live 11.6 years in good health. For both sexes, this constitutes 56.8% of their expected remaining life span.”
Note that life expectancy for old people is increasing rapidly:
“Projections for 2025 for life expectancy at 65 are 88.3 for women and 86.1 for men”
Note that one would expect an annuity that is guaranteed to keep up to give less on average; paying to reduce risk is the whole point of insurance!
See page 6 of:
http://www.ageuk.org.uk/Documents/EN-GB/Factsheets/Later_Life_UK_factsheet.pdf?dtrk=true
ONS data. Open the top spreadsheet:
http://www.ons.gov.uk/ons/publications/re-reference-tables.html?edition=tcm%3A77-257453
“Right now, UK males live on average until 79 and females until 82”: that’s misleading. What we want to know is life expectancy at annuity-buying ages, say age 65 and age 75 for illustration.
@dearieme:
My comment awaiting moderation goes into this. The number of years you would expect to live for if you are this age now are:
65: 23.3 (F) and 21.1 (M)
75: 13.3 (F) and 11.8 (M)
My first reaction was that the number of females dying between 65 and 75 is approximately zero. My second reaction was that there is perhaps a typo there?
@ Greg – I walked right into that one! Thanks for setting me straight and for the useful link. Really annoyed with myself. Will update the piece when I get a mo.
It does at least mean that the index-linked annuity is more likely to ‘pay off’ but it was kind of a flippant point anyway (on my part). I don’t believe you buy an escalating annuity for a higher total income. You buy it to protect yourself against galloping inflation.
@ PC – no, I don’t think a capped RPI is common but certainly some providers do cap it, so you need to be careful and check the terms. Excellent point! Here’s a good piece that discusses which providers cap and how they handle deflation too, which also varies: http://www.investmentsense.co.uk/technical-area/retirement/rpi-annuity-increases-are-they-that-simple-and-are-they-worth-it/
@ AJ – Ha Ha! More risque tolerance than risk tolerance required for this blog, eh?
@ Beat – quite right, personal inflation is important but whaddya gonna do? There isn’t an annuity to cover that. I guess then you get into a world of trying to control your own inflation rate. Eating seasonably, eating own brand, investing in insulation and other energy-saving measures before you stop working, signing petitions in favour of the indigenous fracking industry 😉
@ Neverland – you’re right, drawdown is an option but it entails considerable risk. The annuity angle is a follow up to the minimum income floor strategy covered in the two previous posts. Using this strategy, you secure a basic, inflation-proofed income and then take greater risks with your remaining wealth (emergency fund excepted) to secure some upside e.g. invest in stocks.
Certainly my relative, who set me off on this journey, isn’t interested in the financial management required for drawdown and her pension pot is too small to comfortably take the risk. Not to mention that the first sight of a stock market fall is extremely unnerving for her.
Re:
65: 23.3 (F) and 21.1 (M)
75: 13.3 (F) and 11.8 (M)
@dearieme
“My first reaction was that the number of females dying between 65 and 75 is approximately zero. My second reaction was that there is perhaps a typo there?”
No, you’ve misunderstood.
A woman aged 75 now could expect to live 13.3 years. However, life expectancy of old people is rising so quickly that someone who is currently 65 is in far better shape then the 75 year old was 10 years ago. (Perhaps being born just in time for WW2 and the aftermath is not good for longevity, or perhaps care is improving quickly etc.) For men, the effect is not enough to counteract the risk of dying between 65 and 75, but for women it is! It could theoretically become a net gain!
It’s the same sort of thing as the statistic that a newborn girl is more likely to reach 100 than a 90 year old man. (I have not checked this.)
This is (one of the reasons) why the pension timebomb is such a big deal. People don’t understand the stats. (I admit, they are counter-intuitive.) We have fools like Gideot Osbourne who can’t even see past the next election, never mind this problem.
Of course, it is my generation that will have to pay for it. (See my various whinges in previous articles.) Not only are huge amounts of people about to retire who will demand a pension & healthcare, but there aren’t that many of us workers who will pay for it, particularly as we need to navigate the insane property market. (Any party that suggests that old people should downsize or that house prices should come down, will never get elected.)
It is a looming crisis, but no-one will do anything until it is upon us.
Greg
@Greg,
Do you think that, at some point, the Type 2 Diabetes epidemic will result in average life expectancy decreasing?
Are we making a mistake in assuming that life expectancy will continue to increase at the rate it has done over the last few decades?
Given that a major contribution to increased life expectancy at birth is reduced infant mortality, perhaps life expectancy may plateau in the developed world.
Having raised those points, I’m afraid I’m still inclined to agree with your analysis.
@Greg: “No, you’ve misunderstood”: no, I did eventually ponder that point on my own. However, I still wonder whether you have a typo there. The AgeUK link you gave above doesn’t give the figures you report in your first row.
“UK life expectancy estimates at the age of 65 are 85.6 for women and 83 for men.”
@GOP: “Are we making a mistake in assuming that life expectancy will continue to increase at the rate it has done over the last few decades?” Perhaps; since it involves the future, who knows? When someone suggested a future halt or slowdown in lifespan extension a couple of times in a forum at MSE, it seemed to cause unreasoning rage; God knows why. Here’s a link that suggests that the epidemic of coronary heart disease (CHD) is nearly over: if true, that means that increased lifespan from the reducing incidence of CHD must be about to stop. If the decline in smoking has stopped or slowed (has it?) then there’s another reason for increasing lifespan ending. As for the ‘diseases’ that are partly just a matter of arbitrary and changing definitions, again who knows? Or again, maybe the grossly fat young people we see around will all be dead before 65, so that they’ll not affect life expectancy at 65. Maybe mass-statination will start killing old people, though it seems so far that statins make lots of people ill rather than actually killing them. Anyway, the big potential effect is presumably the rise of antibiotic-resistant bacteria: again MSE readers really wanted to ignore such possibilities and cling to somebody-or-other’s projections of life expectancy. Bonkers.
http://qjmed.oxfordjournals.org/content/105/6/509.full.pdf+html
@ dearieme
Not forgetting the possibility of a well overdue global pandemic such as bird flu.
@ dearieme
Argh! Yes! hold on! You are right! Sorry about that!
Thanks for spotting this.
“UK life expectancy estimates at the age of 65 are 85.6 for women and 83 for men.”
“People currently aged 75 can expect to live an average of 13.3 years (for women) and 11.8 years for men”
I meant to subtract the life expectancy estimate from 65 to get the number of years. However, as you point out, 85.6 – 65 = 20.6. Similarly 83 – 65 = 18.
In my effort to make things “clearer”, I used the values from this stat:
“Projections for 2025 for life expectancy at 65 are 88.3 for women and 86.1 for men” 88.3 – 65 = 23.3, 86.1 – 65 = 21.1. What’s particularly annoying is that I did this twice. 🙁
So, the numbers I should have put up are:
right now, you can expect to live this many years if you are:
65: 20.6 (F), 18.0 (M)
75: 13.3 (F), 11.8 (M)
Note that these numbers are rapidly increasing! (My previous explanation still stands in theory, but the effect clearly isn’t large enough over 10 years.)
This is a good lesson in:
a) Why we cite sources
b) Why we actually check people’s working, numbers and sources
If a mod could amend my original “clarification” to point out the numbers are wrong and corrected further down, please do.
@GOP / dearieme
I don’t know how things will change, but I’m willing to trust the ONS for at least the next 10-20 years. Our pensions / health crisis will hit before any meaningful deviation from projections. The selfish generation (baby-boomers) are starting to retire now.
For medium term life expectancy, I just don’t know. I do agree with the point about obesity etc, but I don’t know if it will outweigh other improvements. Perhaps there will be a period where life expectancy at 65 falls. As I’ve mentioned before, interestingly, the death rate has to rise over the next couple of decades but that’s just because of a bulge in old people getting to the stage of their life where they are much more likely to die. 🙁
As for my long-term gut feeling, I think life expectancy will rise significantly, though not evenly. It might get to the point where a treatment becomes available say, that extends your life by ten years. Then, during those 10 years, another treatment becomes available that extends your life by another 20 years. If that thing keeps going, (a few) people might become effectively immortal!
Oh and having read the comments on a few MSE property articles, I’ve made the crass generalisation that they are all idiots who only listen to people who agree with them and never change their mind.
@Greg / dearieme.
The reaction of rage to the suggestion that life expectancy might not continue to rise indefinitely is an interesting one in so far as it gives an insight into people’s thought processes. Unusual for something other than politics or religion to produce such a reaction. Perhaps the reaction arose because the suggestion touches on their own mortality or simply because it is contrary to conventional wisdom.
Greg probably thinks that we baby-boomers will have free access to life extension procedures but will vote to ensure that they are prohibitively expensive for subsequent generations! (Joke – I’m too old for emoticons!)
One thing I find a little disturbing on many FI/early retirement websites/blogs/forums is the assumption that we must plan as if we are all going to live to 90+. Perhaps we might live better today if we remembered that there are no guarantees. 3 out of 4 of my grandparents made it past 90 and all exceeded 80; so far 2 out of the 4 parents and in-laws are dead (at 56 and 74); and I’d be surprised if either of the others make it to 90. OK that’s anecdote, and clearly in the absence of knowing otherwise, its reasonable to use average life expectancy stats for our retirement planning assumptions. However, I think it would also be helpful to try and look at the figures in another way. At age 45, what are the chances of being dead before reaching age 60/70/80? Not sure if those numbers are available anywhere but my feeling is that it might be a useful way of reframing the issue.
@vanguardfan — Good reminder. “Always invert” as Charlie Munger says.
‘The Reappearance of an Intergenerational Equity Debate in the UK” Macnicol 2012
Doesn’t affect lifespan, just broadens the blame so those outside the boomer years can feel less aggrieved about their own.
I’m still waiting for the first sighting of someone in early middle age who says “I’m so disgusted with the babyboomers that I shall refuse my inheritance when it comes”.
Oh, on level annuities. I suppose that you could simulate one roughly by putting your S & S ISA into War Loan, (or for a non-taxpayer just by buying War Loan tax-exposed) and that way there would be no tax to pay and some capital left at the end. You’d want to time it well but that’s true of annuities too. In fact, it might be clever to use this notion until you are, say, 75 – when annuities start to become better value because enough people are dying to subsidise the payments to the survivors. It might not be a bad idea for someone who doesn’t have much money in pensions but who inherits a decent sum. If interest rates rise so that the capital value of the War Loan sinks, you could be cheered by the thought that annuity rates would rise too.
@ Monk
I had a quick read of the paper you mentioned. I hate social sciences, they can spend 24 pages pondering without actually saying anything of substance or supplying anything testable. Furthermore, they come out with assertions with a single piece of incidental backing evidence.
The paper to me seems to read:
– “Inter generational strife is a popular topic right now and has been in the past, even as part of a weird right-wing plot to privatise pensions/medicare” p1-12.
– “Baby boomers” isn’t very well defined. (1941-1970? 1946-1965?) Unsuprisingly, this is a diverse range of histories. p12
– “Baby boomers” thought their parents got too much and it’s also their fault for having too many offspring. p12-16
-Waffle about “is a generation made from individuals?” False assertion that one can’t count consumption as a group because the number of them would have an effect.
-People who are working have higher income than those that are retired. (A revelation that!) p17
– More waffle that there are different people in the generation. p18-19
– It’s hard to work out what a fair share is. p20.
– More waffle about the fact there are both men and women in the world.
– Inheritance is unfair.
– There is a diverse range of histories. (Again.) p21.
– Reducing inheritance is unfair. p22
– Tiny changes made to employment and pensions apparently mean they have had things against them. p23
– There is a diverse range of histories. (Again.) p23
Conclusions they give:
– Some people are looking at this
– They are oversimplifying
–
I hope that’s not the pinnacle of social science research. The only numbers were a table of income showing people working earn more.
Ah, I can’t use triangular brackets.
The last conclusion they gave was an unjustified assertion as to the possible motives of people who want to look into this.
@GOP. I’m afraid, we’re probably the last few generations who have to die.
@Vanguardfan. You can calculate those numbers using the ONS sheet I linked earlier.
@dearieme. A lot of people will get their inheritance just as they retire. I’d imagine most people would prefer a smaller inheritance but have house prices 75% cheaper. It would increase social mobility too. Hah – perhaps a massive house price crash would go a long way to solving this problem!
I saw it more as a finger pointing at the moon Greg.
@Greg,
I share your views on the pseudo sciences. The Wikipedia piece on the Sokal affair makes for entertaining reading.
@Greg. Thanks for the pointer to your link. Fascinating. I did indeed calculate my chances of dying at age 60/70/80/90; it seems rumours of my mortality are greatly exagerrated and I will indeed live forever!
Using the appropriate cohort life table (for my year of birth and sex, using the principle projection and calculating the chances of being alive at x years, using survivors at my current age as the denominator), it appears that I have a 1 in 62 chance of dying before I am 60 (so the good news is, I probably get to retire!!); 1 in 22 chance of dying before age 70; about 1 in 10 chance of dying before I am 80 and only about 1 in 5 chance of dying before I’m 90. I must admit, I am a little sceptical of these projections, which of course are based on assumptions of continuing future improvements in mortality rates (which suggest someone born in my year of birth has nearly a 20% chance of surviving beyond 110. I find that very hard to conceptualise). The prospect of living into extreme old age holds very little appeal to me to be honest. However big my nest egg…
But, it does put those low annuity rates into perspective. Longevity insurance doesn’t seem like a bad idea now…and my bet would be for annuity rates to get lower still, regardless of interest rates.
@ Vanguardfan – Good stuff, very interesting to come at it from a different perspective. What would you do if you discovered you had a 3 in 5 chance of dying before 90? I ask because those middling numbers are the worst. They offer no real guidance. That’s why I think the minimum income floor strategy makes so much sense. You’re prepared for the worst but are positioned for upside if you’ve got enough left over to create a risk portfolio.
Given the maxim ‘prepare for the worst’ what’s the worst in this scenario? Dying early or extremely late 😉
It’s not just life expectancy it’s at what point you become physically or mentally incapacitated.
The worst happened to an old friend of mine a few years ago. After a fairly dissolute life, he gave up beer, fags and fast women, had a mild stroke a year later followed by a fatal car accident with no-one else involved. He was 62 and financially secure.
The lesson I take from this is ‘Enjoy it while you can’; you can overdo deferred gratification. Plan to spend more in the earlier part of your retirement whilst you can still enjoy it.
@G.O.P. — Good advice, that I for one need to take to heart. My father who inspired this blog got about three years out of his annuity (my mum still gets a reduced payout from it) before he had an incapacitating stroke that kept him chair-bound until he died last year.
We used to have spirited arguments quite a lot, but his argument to me that it was best for him to get the level annuity versus my urging of an an inflation-linked one was an argument I’m glad he won.
@beattheseasons – yes that is a very good point. Unfortunately we don’t (yet) have life tables for ‘healthy life expectancy’ but I am aware of research in this area that suggests that many/most of the years we are adding to life are in fact not healthy but affected by some kind of disability (which of course can be a long slow process from full capacity to very limited quality of life).
Don’t know what the answer is from a planning point of view (good care in old age is pretty expensive) but if I could put my request in, it would be for about 6 months notice before my date of death and full health before that 😉
Yeah, as long as spending more in the earlier part of your retirement doesn’t lead to subsisting on cat food if, by some stroke of fortune, you don’t die young and end up trucking well into your 90s.
Loss of mental faculty just increases the argument for a simple plan that pays out rather than complex financial management that may prove beyond us later in life.
My father had an annuity that escalated at a fixed 3% , regardless of inflation – worked out very well as it turned out.
Is there a simple way to calculate a fair price for these alternatives?
It seems we might not be living as long as we thought:
http://www.bbc.co.uk/news/magazine-23126814
I rather buy a level annuity as well and keep some cash to invest in order to improve my overall portfolio return. I believe fees incurred in indexed-annuities are too high to buy the piece of mind of avoiding inflation risk. A good dividend portfolio can do the job while you receive a steady income from your level annuity.
Remember two critical things about annuities: (a) they’re not for everyone or every need, and (b) they [by implication] nevertheless probably have a useful role in most anyone’s portfolio–even if they CAN answer ‘yes’ to such questions as “I can do that myself” or “yes, I have a trusted advisor.” Hey “guarantee” is “guaranteed” and as any fellow securities-licensed person can tell you, that’s a word that’s used very sparingly. And the Suzy Orman’s of the world insist on finessing that point as they obsess on the commissions we, rightly, earn by selling great products.
Yeah, I know the Dow is at 16,000+. For now. I had as much luck as foresight when I moved the bulk of our funds into an indexed annuity just before the ’07 melt-down. I’d be in the same place now if I would have stayed in the market, but without the peace of mind I had from a great annuity product. As I’ve never worried about “maxing out my return” I’ve ended up, at 68, in a great place.
For the UK, you need to factor in the State Pension with the triple lock, or perhaps a future double lock, as an escalating annuity. It is too politically sensitive to renege on beyond that. So, in these terms, a combination of level annuity, drawdown AND 25% tax free cash, at the moment, squirrelled away in ISAs as a reserve, would seem a good diversified pension portfolio…..