A big but oft-overlooked question for aspiring retirees is: “Will my spending decline in retirement?” If the answer is yes, then you could retire sooner than you think, or you could spend more money in the early years after your own freedom day.
As it happens, there’s a large stack of research that suggests people really do see their spending decline in retirement. At least on average.
And if this turns out to be you, then the amount you need to retire should be less daunting than previously advertised. You can even afford to decrease your ‘how much should I put in my pension’ target figure.
That’s because sustainable withdrawal rate (SWR) strategies allow for constant inflation-adjusted spending in retirement.
They assume you want to maintain your purchasing power for the remainder of your days. That sounds reasonable.
But the evidence suggests the majority of retirees actually keep saving as their spending needs fall. The net effect is they beat inflation and do better financially than predicted by overly-cautious retirement planners.
So what’s the evidence for the spending decline in retirement?
That’s a 64 million dollar question. (Okay, more like £6.40 in my case).
UK evidence that spending declines in retirement
I’m going to focus on a UK research paper called Understanding retirement journeys: expectations vs reality.
It was authored by Cesira Urzi Brancati, Brian Beach, Ben Franklin, and Matthew Jones for the International Longevity Centre UK (ILCUK).
The ILCUK is a think tank concerned with the social impact of an aging population.
This paper is based on longitudinal data of households in the UK and England. Brancati and co find the majority of these households experience a real decline in retirement spending:
Our findings suggest that typical consumption in retirement does not follow a U-shaped path – consumption does not dramatically rise at the start of retirement or pick up towards the end of life to meet long-term care-related expenditures.
[…] the inescapable truth is that, regardless of the period analysed, lifestyle or income level, older people in the UK spend less than their younger counterparts, with discretionary spending on life’s luxuries all but disappearing from age 75 and almost all cohorts progressively saving more of their income as they become older.
The findings apply on average to broad populations. Please bear this in mind from here. That way I won’t need to clog up every sentence with words like ‘average’ and ‘typically’.
Also, I don’t want to ruin your day by making you read methodology spoilers. A full discussion on data sources and research parameters is available in each linked paper.
The retiree spending decline headlines
The paper paints a vivid picture of decline:
- Eighty-something year-old households spend an average 43% less than fifty-something households.
- Spending on holidays, eating out, and recreation declines along with other non-essentials.
- Essential items such as food, health, and housing eat up more of the budget.
- Savings increase as the drop in discretionary spending dominates stable non-discretionary outlays.
The research also analyses behavioural surveys. This helps to connect the dots revealed by the expenditure data:
• Time at home alone increases by age, while time spent with family and friends falls. By age 90+, watching television and spending time at home alone are the most common daily activities.
• The age group 70-74 appears to be a tipping point. From this age, the average amount of time spent at home alone increases markedly, while the time spent with family and/or friends falls.
We’ll explore the obvious and not-so-obvious reasons why this is so, shortly.
An immediate question is: does the fall in non-essential spending apply mainly to the affluent?
This graph shows that the retirement spending decline affects low-earners and top-earners alike:
Bottom-earners (blue line) spend proportionally more than their income past age 65, but then their savings rate accelerates.
Top-earners (red line) always have the capacity to save. But again, their spending arc bends increasingly downwards.
Brancati notes:
This pattern is common to both high and low income groups, is robust to the inclusion of factors other than age, and is not simply the result of the time period in which the data was collected. Subsequently, households make substantial savings in later life.
Choose your retirement tribe
Obviously there isn’t a generic lump of retirees that all behave alike. But Brancati identifies five sub-groups, and all see the same pattern:
Find out who these people are on pages 29 to 33 of Brancati’s paper.
In short:
Transport Lovers and Extravagant Couples are described as high income. Both spend freely on non-essentials. Extravagant Couples even go into hock during their early retirement.
Prudent Families have ‘relatively high income’ but are predisposed to save.
Frugal Foodies are low income, spend little on non-essentials, and are diligent about saving. I estimate their annual household income to be around £19,000 at today’s prices (based on the report’s 2013 numbers).
Just Getting By are also low income and are unlikely to own their own home or investments. They’re disproportionately affected by rising rent and energy prices. They start to save from age 75 nonetheless.
What happens to retirees?
The go-to rationale is that fading health cages older retirees. However that’s only part of the answer.
And the notion of people cutting back because of dwindling financial resources is confounded by this graph:
A majority feel more secure about their finances as they age, not less.
That’s the good news part of the story.
Meanwhile, spending on essentials (food, housing, clothing, and health) remains relatively stable:
The exception that proves the rule is transport. People spend less on this when they leave the workforce and stop commuting. Recreational travel wanes, too.
Notice how gently the health line (in blue-grey) rises at the end. As if wafted by an ill-wind…
In contrast, non-essential spending goes into steep decline long before that – from 65:
Spending on eating out, hotel stays, and even alcohol take a 50% or greater hit from their peaks.
What about the Lamborghini factor?
Brancati says there’s not much evidence for a post-retirement day blowout:
A closer look at the different categories of non-essential spending reveals that people spend a relatively similar amount of money on recreational goods and services between the age of 50 and 65, and only then do they start spending progressively less.
This seems to contradict the stereotypical image of retirees splurging in the immediate post-retirement phase of life, going on cruises and spending all their hard-earned cash on fun activities.
I do know retirees who’ve gone splurging. I have to remind myself to focus on the overall trend and not my own anecdotal evidence.
Particularly troubling though is the dipping purple recreational curve from age 65 on.
That slump is a warning that the ‘active’ years of retirement may be short.
This graph plots that story, and it bothers me the most:
The steepening curves reveal that age 70-74 tipping point I quoted earlier.
Spending time home alone climbs relentlessly. Brancati puts it in stark terms:
Time at home alone rises from 3.5 hours a day for those aged 70-74 to more than nine hours a day by age 90+.
Conversely, time with family or friends falls. I guess that’s partly because funerals become all-too frequent.
Also notice the big drop in walking and other exercise.
In contrast, time spent on ominous-sounding ‘health-related activities’ marches upwards. This is an ill-defined category, but Brancati says it likely involves visits to the doctor and other medical facilities.
Why then, does the line subside from age 85 to 89?
The decline after this age group may simply relate to longevity factors, i.e. healthier people or those with fewer health issues are the ones who survive to this age.
Explaining the retirement spending decline
Wondering whatever happened to the likely lads (and lasses) to keep them home alone? Here’s the money shot:
From age 50 to 64, the number of people who often or sometimes agree that their health stops them doing what they want hovers in and around 30%.
By 70 to 74 that proportion rises to over half.
And for the over-Nineties, 84.5% of them agree their health limits their lifestyle often or sometimes.
But health isn’t the only reason that time spent at home increases.
Other factors with explanatory power include losing your partner (and not being partnered at all), being male, being part of a small household, and not being a carer for someone.
Inevitably, our advancing infirmity changes us. In Brancati’s words:
The anticipation of ill health and disability may also increase the desire to save in order to help meet potential health and long-term care costs in later life.
She also believes a desire to leave an inheritance contributes to waning spending.
Some other studies conclude the decline is mainly explained by falling work-related expenses, the substitution of time for spending, and involuntary early retirement.
Forced retirement is largely due to health shocks. It especially affects low income groups.
Finally, there’s an interesting snippet in a US piece that claims retirees cannily neglect home maintenance later in life. Such slapdashery enables retirees to pile up savings to offset their uncertainty over life expectancy and future income.
But will my spending decline in retirement?
My two big questions about this research are:
- How reliable is it?
- What practical use can we make of it?
I’ll have to dig into those questions in my part two – otherwise, this article would be unreadably long again. (Oops, too late for that!)
Take it steady,
The Accumulator
Great article as ever TA!
One thing I would add is that the pensions freedom changes only came in just before the paper was published. I wonder if the relaxed rules have seen an uptick in Lamborghini orders since?!
Useful article on an exceptionally interesting report. Good to see a decent analysis that appears independent (although noting the Prudential logo).
As FIRsE (slightly early, 3 years) and in my late 60s, it is a little depressing to forecast the inevitable decline, but not a surprise and already in my (prudent) plans.
I await Part 2 eagerly.
My personal experience two years post-FIRE: is that our household spending has dropped by 10-20%. None of this is due to any conscious belt-tightening – in fact, quite the opposite. Factors include less spending to cheer myself up; having the time/energy to make good food at home vs. going out; doing leisure activities that are take more time vs. pricey ‘quick-fixes’; and more time/inclination to shop around for better deals.
I’m not sure how much COVID was a factor at the beginning, but from a spending perspective I have the feeling it’s more or less dropped out of the equation now (although perhaps long-haul holidays are still a bit less frequent atm).
I do think it’s important not to set your retirement date by budgeting for a ‘flat’ spend over time: there’s no question that one’s ability to spend money falls off after a certain age. You may need eventually to think about residential care costs, but IMHO that shouldn’t impede the priority of enjoying yourself to the max while you have the ability.
Fascinating stuff. Interesting to speculate how health care spending might change over the coming decades. Personally, I’ll be keeping a pot vaguely labelled – radical health intervention as if modern science comes along when I’m in my 70s, and says spend £50K-100K and have another 5 to 10 years of good health and active life outside of a care home, I might well take up that offer. Especially, if it gets me to the point that another miracle of modern science offering the same turns up 5-10 years later.
All, of which, suggests why wait for that – and consider what is an appropriate level of health investment now and what should I spend it on? It’s probably not another bike and accompanying gadgets!
I echo some of the earlier comments of mjcross. I retired 7 years ago and my spending has been approx 25% less than expected, even after a rough adjustment for covid. This surprised me because I had kept an annual budget since I started uni in 1986, so had plenty of data. Probably too many unknowns re discretionary spending to predict accurately?
Nice post – looking forward to part 2.
One of the apparently reasonable criticisms of the body of work that shows spending on average declines in retirement is that the studies are often constrained to just “older” folks. That is, the results – by definition – are not applicable to early retirees. See, e.g. chatter between myself and ERN at https://earlyretirementnow.com/2021/08/18/when-to-worry-when-to-wing-it-swr-series-part-47/
IMO, Figures 3 & 4 in the referenced ILCUK provide some food for thought on this point for UK folks!
@TA – great stuff as ever! And it makes intuitive sense, though it’s great to see UK based research. The majority (i.e. US) fetishises (understandably) about healthcare costs.
At 56 and humming and harring about retiring, filling my time constructively is a worry. Apart from my wife and children, I’m not sure I have any outside interests. Lots of plans, yes, but if they were realistic I’d being doing those activities now.
I notice some of the charts are assessing “household expenditure” – not spending per person. Presumably if you live in a couple, and one person dies, your household expenditure will indeed drop – significantly! Does that help? Perhaps not if your income drops at the same time.
If you are single all along, that part of the reduction and the associated figures is irrelevant to you.
On a per capita basis, a single-person household is more expensive – you still have more or less the same utilities costs, for example. But I do find the use of “household” as yardstick a bit misleading, without factoring in household size.
It’s also not clear how “household” expenditure figures relate to people who go into a care home. Are the care home fees included? What is their “household”?
@Haphazard: “our data is restricted to households only and therefore excludes those actively living in care homes who may be paying for it from their remaining assets.” (p.18 of the linked report) But they then talk about the impact of long-term care costs on households. Which is a bit confusing, but I think they’re saying that they include e.g. couples where one of them is in care or a family paying for a parent in care, but don’t include “single” people in care if they’re “financially disconnected”, as it were. Maybe someone else can decode it better than me. By the way, their figure for average care home costs hides massive regional variation (you can look at their source for that).
I have no issue with the data or conclusions with this study. I am more doubtful, however, about how well this data will translate to those of us reaching retirement age in 20+ years time.
There are substantial demographic changes afoot. It seems to me likely that , in particular, the spending profile for households in their 50s will change. The age at which people buy property is later, leading to a later date for becoming unencumbered. My parents’ generation often disposed of financial responsibility for their children before the age of 45-50. These days that responsibility has a much longer tail. Many of us will not dispose of that until 60-65.
So rather than it being a period of relative comfort leaving more substantial ability to spend and save, it actually becomes a period where fixed spending it still rising. The impact of that will have big implications into their retirement in their late 60s and 70s.
Interesting indeed. I found it difficult to determine the sample size, which after some digging is slightly less than 5000 households. This translates to ~9000 adults and with the 70-74 ages being 3.9% of the population means those numbers are based on ~400 people. 2.2% of people are over 85, so that number drops to ~200.
That seems like a reasonable number of people to ask, and to see a trend from, but is far from definitive…. not that you claim it to be, just a further warning.
@Haphazard The report covered here says it used the 2013 data from the ONS. You can read how the ONS did there report here (where I got my numbers from): https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/methodologies/livingcostsandfoodsurvey
Very thought provoking piece.
What is not alluded too enough is the cost if healthcare that is free in the UK and the lottery of your eventual demise – care home costs or assisted living costs are ruinous for those who foot the bill.
Also worth considering is the longer term cost of kids – I’ll be in my 60s or 70s before the kids marry so that front loads a lot of potential retirement spending. Likewise with university Costs, house deposits.
Of course, lifetime costs are all front loaded and burden the young – life as a pensioner should be cheap (otherwise you’ll have a shit time trying to survive on the UK state pe Sion alone!)
@beeka (#11):
The study uses a number of referenced sources.
AFAICT, the financial data is [largely] culled from what you correctly identify as the ONS’s living cost and food (LCF) surveys from 2003 to 2013 inclusive and re-flated using RPI. IIRC this means a total of six LCF surveys were used; as the LCF is a biennial repeated cross-sectional survey. It is also worth noting that, as is normal for consumption analyses, investments including mortgage costs are excluded. What is included/excluded in consumption is explained in the Report itself.
Whilst the LCF is a entirely reputable source; it is not without issues. For example, households are known to under-report spending e.g. on alcohol and some issues were noted with internet costs collected via the LCF vs what OFCOMs ‘whole of the market’ work gathered. Such issues are not unique to the LCF. No survey is perfect and currently around half of the households invited to participate in the LCF do so – which probably gives rise to some form of selection bias too. Having said that, “in the land of the blind the one eyed man is king” or similar.
BTW, the ELSA survey data is longitudinal, the ONS is not.
@ David C (#9):
I think the answer is wrapped up in the scope of the LCF survey – which IIRC only uses domestic households across the country, i.e. excludes care homes, hospitals etc.
That the Report then goes on to attempt to quantify average care costs (inc links to the wide regional/quality variations) is IMO entirely laudable.
N.B. the Report is from 2015 so some of the numbers are quite dated now.
This bears on the same subject: hope it’s relevant.
http://www.theretirementcafe.com/search?q=Spending+in+retirement
The study and its findings explain in detail a simple truism voiced to me by a 72 year old retiree in 1969. I was then 25. Taking a long drink from his pint glass he leaned toward me confiding, “For most of us three things in life never coincide at the right moment: time, money and inclination. When you’re young you have time and inclination but little money. In middle age you have money and inclination but little time. In old age there’s more money and time but less inclination.” The study’s findings echo what real life has unfolds and continues to demonstrate. Within a few weeks of retirement spending declined.
Really interesting and highlights a big weakness in the SWR projections that, to me, often assume that you’ll be taking 4% from your savings at 60 and will still be doing so at 90. To spend exactly on what, at 90, I often ask myself? The pat answer seems to be “on health care” but to me that’s just a convenient way to prevent yourself from withdrawing 10% from your savings when you’ve the health to enjoy it!
My tuppence worth
Wife and I now 75-until Covid we were spending the same amount as we had when we were earning -mostly on travelling the world-got out of Argentina on March 14th 2020 with the gates of lockdown clanging shut behind us -didn’t stop till safely home!
Covid stopped all that -spending down 30%+ but creeping up again
Both still fit and ready to go again but now Europe only in our sights at the moment
Lots of our compatriots are starting to fail and if one of the couple is crocked then that’s it
Till then which in the law of probabilities could be any time still feeling 21!
xxd09
I’d argue that neglecting home maintenance in retirement doesn’t apply to many. Ive been retired from employment for 10 years now & whilst personal expenditure has remained fairly stable over that period, spending has definitely increased. Main spending increase has been household repairs & renovations. In run up to retirement I was too busy work-wise (& lazy otherwise) to do much in the way of household improvements. With retirement I have had time to install new bathroom, put in new carpets, fittings, etc, etc. With increasing age there has been a feeling of wanting more than budget ranges, almost a sense that this might be the last time (or next to last time) I have a new bathroom installed, so let’s make it a good one. My peers also report dipping into savings and/or using pension lump sum to fund significant additional home improvements throughout retirement – not just as a one-off. Spending decline in retirement certainly doesn’t apply to us.
Good thread this. It’s worth noting this is one paper from a wide literature that points broadly in the same direction.
There does seem to be a marked difference between the ‘young old’ say 65-75 and older age groups increasingly constrained by health.
The Accumulators had our wings clipped this year by an unexpected health issue. I can quite easily see how stacking a few challenges like that would wipe out much of the budget we’d allocated for ‘experiences’. The downward-sloping lines in most of these charts do imply we should live large while we still can.
I also wonder if many people just care less about living material lives as our time runs down? The paper indicates that as long as people can cover their essentials, then they feel more financially secure. That makes sense when I think about other findings that suggest people’s happiness and self-confidence often increases with age. Nothing to prove. Less need for external validation. Money matters less when you can’t take it with you.
I think it’s interesting that lower income age groups tend to save too, seemingly to ensure they have something for a rainy day.
So I wonder if the trend could be robust even if financial conditions worsen for later generations? If you’re forced to make a choice, then many
still opt to keep a savings buffer, and forgo some other options.
Some of the older age groups showing up in this data must be pre-boomer. Presumably their retirements were/are less lush than post-war age groups.
It’s good to have all the data and charts etc but this does seem intuitively obvious, although I guess that depends on your personal experience.
In my case I can’t think of any elderly relatives who spent huge sums of money on cruises, cars, clothes etc. They mostly just pottered around drinking tea, going to the shops and doing some knitting/gardening.
I’m only 50 and my inclinations have already shifted from driving fast cars around race tracks to watching a movie and eating biscuits and one activity is much cheaper than the other…
As for the younger generations, I agree with ZXSpectrum48k. I think they’ll find retirement much less pleasant. They bought homes much later (if ever) and when they did buy, they paid up to twice as much as earlier generations for the same house. Then we’ve had low inflation (so far) so the mortgage payments aren’t being inflated away as they were in the 80s and 90s. Add all that up and I think a lot of people are going to be paying a lot each month on their mortgage well into their 50s and 60s, leaving little for pension savings.
My parents finished paying for their house and kids by the time they were 50, and I’d say that was pretty normal for their generation. No chance of that for the vast majority of Gen Y and Z (and quite a lot of Gen X too).
Studies based on averages provide false comfort for a retirement that will be uniquely your own family’s journey.
The idea that spending will decline with age is totally dependent on whether someone needs some form of external old age care and what the rules are for other tax payers paying for that care in the location and time where they need it.
Agree on the kids part. The forecasting is vastly simpler for those without kids or the certainty that they’ll be settled into self sustaining careers by point X.
I think also there is a wildcard for parents as well when it comes to those retiring where one or both parents still around may start absorbing time/money/energy in their care need. Even if they don’t represent a financial draw they can curtail or disrupt plans for a carefree retirement with associated spending.
Contemplating on when I pull the trigger with a father who is problematic at present is very difficult – I’d like the early years to give me the time and flexibility to travel etc but that’s fairly pointless if I feel I need to be be in reach to help my mother with things
Our spending shot up for the first few years after we stopped work, but it was all expected and planned for in advance with a budget of cash to cover the increased spending.
When you are not working you have more time to spend. Not having parents, grandkids or other dependants to think about gives us a lot of freedom at present as well, which may not last.
I agree with some of the comments around spending being higher in future due to purchasing property later and kids being less financially secure. However, it’s also the case that many will have substantial inheritances at some point in the future. Certainly in my case, with boomer parents hoarding property wealth, I can be relatively confident of coming into a largeish sum of money some time in my 70s or. 80s. (My hope is that I will be well off enough to pass that straight on to my kids). I think inheritance is something that people have avoided factoring in in the past, but with so much property wealth locked up it sort of becomes impossible to ignore.
Great point about inheritance. Property price growth will increase inheritance value substantially but so will the pension reforms. As the numbers of people purchasing annuities falls you’d expect those passing away with sizable pension pots remaining to increase.
@BBBobbins
‘Contemplating on when I pull the trigger with a father who is problematic at present is very difficult ’
Crikey. Sounds a bit extreme. Is that legal?
I have retired and my spending has fallen off a cliff. I basically need coffee, books, gardens tools and Im happy.
I loathe travelling so avoid that if possible.
I can eek out a month on £200.
It’s not just expensive properties that will be inherited, but now no-one gets DB pensions anymore, there will be large lump sums being inherited from the previous generation’s SIPPs too.
@Mousecatcher007 maybe unfortunate choice of words but also a bit of accidental truth in them as to how my mother feels at times sadly.
As for the inheritance point – I’ve always planned on the basis that it will be nil because of care costs. But I suspect it is probably fair to factor it as one part of a portfolio of possible upsides like equity release or downsizing/location arbitrage that might mitigate against worst case scenarios with your pot.