Duvet day here at Monevator as we update our asset allocation quilt with another year’s worth of returns.
The resulting patchwork reveals the fluctuating fortunes of the major asset classes across a decade, and invites a question…
Could you predict the winners and losers from one year to the next?
Asset allocation quilt 2024
The asset allocation quilt ranks the main equity, bond, and commodity sub-asset classes for each year from 2015 to 2024 from the perspective of a UK investor who puts Great British Pounds (GBP) to work.
We’ve also squeezed in money market funds this year. These can be thought of as cash-like, if not quite as safe as money in the bank.
Here’s what you need to know to read the chart:
- We’ve sourced annual returns from publicly available ETFs that represent each sub-asset class.
- The data is courtesy of justETF – an excellent ETF portfolio building service.
- Returns are nominal1. To obtain real annualised returns, subtract the average UK inflation rate of approximately 3% from the nominal figures quoted in the final column of the chart.
- Returns take into account the Ongoing Charge Figure (OCF), dividends or interest earned, and are reported in pounds.
- Again, these are GBP results. If our numbers differ from yours, check that you’re not looking at USD returns. (It’s either that or our minds have been obliterated from staring too long at the crazy pixel explosion above.)
Shady business
While our chart may look like the worst pullover pattern ever, it does offer some useful narrative threads.
For starters we can see investing success is not as simple as piling into last year’s winner. The number one asset in one year typically plunges down the rankings the next. A reigning asset class has only once held onto its crown for two consecutive years – broad commodities achieving the feat across 2021 and 2022.
Long periods of dominance are possible – see US equities. S&P 500 returns have only dropped into the bottom half of the table once in the past decade (in 2022), and stand head and shoulders above the rest in the ten-year return column. If you started investing after the Global Financial Crisis then you have US stocks to thank for the bulk of your growth.
The danger is such patterns gull us into thinking it will always be thus. Whereas in reality, the asset allocation quilt for, say, 1999 to 2008 would have looked very different. US stocks lost 4% per annum over that ten-year stretch.
I suspect S&P 500 tracker funds were a touch less popular back then!
Indeed, US stocks have fallen behind the rest of the world many times over the last century.
And credible voices warn we can’t expect US large caps to rule forever. Albeit such commentators simultaneously acknowledge that they cannot predict when regime change may come.
(We’ve written more about this problem and what you might do about it.)
The golden thread
Gold looks attractive as the leading non-equity diversifier in our chart. Its ten-year return of 10.2% is incredible for an asset class that theorists claim has no intrinsic value.
It’s volatile stuff though. When we first created this asset allocation quilt in 2021, gold’s ten-year return stood at zero after inflation
I remain personally ambivalent about gold.
If you’re a young accumulator you don’t really need it. However aging wealth-preservers may be grateful for gold’s ability to improve risk-adjusted portfolio returns.
And the yellow metal may mitigate sequence of return risk as part of a portfolio designed to cushion the downside.
A chequered past
It’s notable how a truly awful few years can completely contaminate our perceptions about an asset class.
Bond’s ten-year returns were perfectly satisfactory back in 2021. But they have taken a drubbing since.
Now UK government bonds (gilts) look like a liability by the light of the last ten years.
Yet higher bond yields are almost certain to deliver better returns from bonds over the next decade, provided inflation is tamed and the global political outlook doesn’t go from bad to worse.
Over the long run, ditching a key diversifier like bonds is likely to prove a mistake. Splitting your defensive measures between nominal bonds, index-linked bonds, cash, commodities, and gold does make sense though.
Getting defensive
A major Monevator theme over the past couple of years has been to improve our coverage of the defensive asset classes – delving deeper into how they work, when they work, and what the risks are.
Take a look at:
- Why bonds aren’t enough
- Defensive asset allocations
- Cash versus bonds
- Money market funds
- Inflation hedging
- Is gold a good investment
- Why commodities belong in your portfolio
- Decumulation portfolios: The No Cat Food portfolio and the All-Weather portfolio
I appreciate that’s a lot of links. But the more you know, the less the disco dance floor of asset returns in our chart above will cause you a headache.
The colour of money
The bond crash has caused many investors to simply replace bonds with cash.
We think of cash as an asset class like any other and so we’ve introduced it to the table, using a money market ETF as a proxy.
More than any other asset class, cash (here our money market ETF) lurks in the lower half of the table.
That’s no surprise. The job of cash is to be liquid and stable, not to make lurching advances and retreats like the more temperamental asset classes.
On the ten-year measure, cash looks okay. But over the long-term it’s delivered only about half the return of longer bonds.
Material matters
Commodities have crept up the ten-year rankings every year since we began the asset allocation quilt. Now they’re up to fourth place and stand in line with their expected real return of about 3%.
Commodities present a fascinating dilemma.
They’re the one asset class that positively thrives when inflation melts bonds and equities. Commodities are also a tremendous diversifier due to their lack of correlation with equities, bonds, and cash.
But you’ll need testicular fortitude to live with the volatility of raw materials.
Commodities have inflicted losses for five out of the last ten years, but redeemed themselves with spectacular 30%+ gains on three occasions – most critically when inflation lifted off in 2021 and 2022.
Commodities had a surprisingly quiet year in 2024, delivering a decent 7% return thanks to a late comeback in the final quarter.
Our asset allocation quilt suggests they’re rarely so moderate. Most years you’ll love or loathe them.
The missing link
Inflation-linked bonds still make sense despite their desperate showing in 2022.
We’d been warning for years before that mid to long duration UK linker funds were badly flawed. But even our preferred short-duration inflation-linked funds haven’t kept pace with inflation, due to the massive hike in yields that accompanied the 2022 bond rout.
One solution is to hedge rising prices with individual index-linked gilts which – if bought on today’s positive real yields and held to maturity – will protect your purchasing power against headline inflation.
We’ve recently written about how to do that:
- See the Using a rolling linker ladder to hedge unexpected inflation section in our post about deciding whether or not you need such a ladder.
- How to buy index-linked gilts demystifies how to purchase individual linkers.
- See our step-by-step guide to constructing your own index-linked gilt ladder if you do want to do it yourself.
Note that to get ten years worth of returns, our asset allocation quilt currently tracks Xtracker’s Global Inflation-Linked Bond ETF GBP hedged. This is a problematic mid to long duration fund, as discussed!
Stitch in time
However you weave your response to the challenges of investing, the asset allocation quilt makes it plain that the best way to anticipate the future is to be ready for anything.
Buy your asset classes on the cheap after they’ve taken a kicking, grit your teeth while they’re down, then reap the reward when their day – or year – comes around again.
Finally, as uncertainty abounds, let’s be thankful that if you banked on the default position of global equities then you did just fine.
In fact, more than fine over the last decade. That 11.5% annualised return – 8.5% in real terms – is excellent.
Take it steady,
The Accumulator
- That is to say they are not adjusted for inflation. [↩]
Nice to see a British version of this, and timely with the article here about the 250… https://www.ukdividendstocks.com/blog/is-the-ftse-250-expensive-at-three-times-its-dot-com-peak
I am fortunate that I have ridden the 250 wave over the last 10 years with an unbalanced portfolio weighted towards it, and equally unfortunate in my lack of faith in the S+P.
I can’t complain.
Thanks for the quilt, I hope it becomes a regular yearly feature.
JimJim
Fascinating stuff and I expect I’ll return to it nervously, every time I have some funds that need investing.
I’ll echo the earlier poster in hoping this becomes a January feature – appreciate the work involved.
Nice to see a British version,
That one is interesting as well https://www.bankeronwheels.com/the-long-game-historical-market-returns-2022-expectations/
Great summary!! Which Global Equities ETF you considered in your table?
Much appreciated, likewise.
Intrigued by your comment that “in theory our asset allocation should align to the world economy”. Don’t most indexes, and hence most trackers, align to the world’s stock-market capitalisations? Which is not the same thing as the economy at all, although perhaps in theory it ought to be. I assume that a big chunk of China’s economy is outside quoted stocks, with enterprises owned by state bodies, cities etc. The UK’s National Health Service is a big chunk of GDP, but not investable (yet!). And there are grumbles about increasing chunks of US industry being owned privately where the retail investor can’t get a piece of the action. But I have no idea how much any of this matters.
Thanks for this nice UK variant of the chart; appreciate the work taken to generate it!
Re: “Indeed US equities achieved a podium place in nine out of ten years.” Is it not rather eight from ten?
@ David C – You’re right, geographic market-cap trackers are aligned with stock market capitalisations rather than economic punch. The US and UK are both over-represented by stock market capitalisation. It’s impossible to align with the real economy but there are some interesting articles out there on how you might go about it.
@ Attilio – It’s iShares MSCI ACWI UCITS ETF (Acc)
@ Al Cam – cheers! And you’re spot on about 8 out of 10 years. In my defence, I was going colour blind by the time I finished the table 😉
@ James and Jim Jim – thank you for the links. + Chiny – Yes, think this will be an annual feature. I’ve done the hard work now getting the first 10 years worth of data.
A very useful graphic. Thank you.
17% annualised over 10 years from the S&P 500. Jeez, that’s what overthinking has cost me.
@MrOptimistic — At least you recognise it rather than obfuscating, or prophecizing an 80% crash to make things ‘right’.
That’s a credit to you. 🙂
A fabulous infographic TA and very useful for novices such as myself. I wonder what a Cash category would look like over a similar 10 years?
Thanks Grumpy. Cash would be a nice baseline to have in the table. I thought about doing it using some kind of money market fund or ultra-short bond ETF. I’d guess it’d be above commodities, somewhere around gold.
Interesting that the domestic focused FTSE250 has done so well over ten years.
Beating everything outside of the US despite some large economic shocks.
1% wouldn’t be a completely terrible approximation as a baseline for cash over the last ten years. In most individual years you could have done worse, but annualised over the 10 years almost anything would have been better. Published statistics for “average interest rates” are pretty useless as a comparator (why would you leave your money in an average account?). so I wish I had more accessible records for my “shopping around the building societies, with judicious use of regular savings accounts and fixed rate cash ISAs” approach, but I reckon 2-4% pa for up to 2015 or 2016, and 1-2% since would be in the right ballpark (I think it was “funding for lending” that really killed retail savings interest rates, not the post-GFC base rate cuts). So yes, I agree with “somewhere around gold” (and better than money-market funds – at least, mine has been returning 0% – before platform charges).
Incidentally, on reflection, “in most individual years you could have done worse” seems like a good argument, if you needed another one, for a chunky cash buffer if you’re decumulating.
@TA
Thank you for this. Is there any particular reason why global small cap (or US small cap) as sub-asset class is not shown?
REITs global or UK? thanks
@ new investor – just had to draw a line somewhere 🙂
@ Ka – global reits
Fascinating graphic TA, thank you.
I did wonder how an equivalent graphic in local currency would differ but on reflection perhaps a USD equivalent would be a) possible to generate (e.g. swap VWRL for VWRD etc) and b) more informative. I’ve got in the back of my mind how the post-war GBP/USD has juiced returns for UK investors holding US assets.
I have seen equal weight funds like RSP which stacks up well against SPY (as per https://einvestingforbeginners.com/equal-weight-sp-500-etf-ansh/) but RSP is equal weight companies, which still has a skew of a larger % in technology;
Information Technology 15.04
Industrials 14.45
Financials 13.33
Health Care 13.12
Consumer Discretionary 11.04
Consumer Staples 6.89
Real Estate 6.21
Utilities 5.63
Materials 5.47
Communication Services 4.53
Energy 4.29
Using the quilt numbers, it would be interesting to see the results of an equal weight sub-asset portfolio, rebalanced equally at the end of each year.
Perhaps a contrarian buy signal for European equities…
@D — Well you have to eat returns in your currency of choice (or a mix). I know investors who measure their returns blended over a basket of currencies because they live a very peripatetic lifestyle for example. A scant handful try to track returns in special drawing rights. But 90% of Monevator readers are UK based and GBP returns are what matter, and most of the rest are US based, where their own USD quilts abound (one reason we did a UK version).
Local returns in local currencies are interesting academically but they are not much practical help to an investor I’d submit. Even if you think they may tell you something about the future performance, remember these things are correlated with equities. For example currency weakness may boost local share prices in a local currency because it does a lot of exporting and listed exporters become more competitive. That doesn’t mean it’s a bad idea to invest, but you could for example pile in, then see the currency strengthen on economic growth (good when translated back in to GBP) but that in turn then de-rates your local investments because the exporter becomes less competitive again!
Of course this sort of thing happens over the long term (and inflation looms large in the maths/impacts, too, with currency appreciation/depreciation) not over a year or two.
@Trevor — Equal weighted funds usually outperform because they have more small cap exposure I believe.
@Hari — I thought *exactly* the same thing, but decided not to burden @TA with my active speculations…
[Edit: Note, especially to @D who has unfortunately subscribed to comments — my first reply was wrong, I banged out the wrong outcome of the currency/competitiveness roundabout! Have tried to expand more clearly/correctly.]
Great article.
I’d also recommend looking at the US angle that came out quite recently.
https://awealthofcommonsense.com/2023/01/updating-my-favorite-performance-chart-for-2022/
Some observations from my perspective
– REITS are not a great diversifier at all. Worth owning as part of your overall allocation to risk assets, for most people through an index tracker. But I struggle to see any benefit in a oversized holding. In times of stress they seem to behave like super volatile equities. Same happened in 2008. Note very different to owning real estate physically, which comes with its unique set of opportunities and challenges
– Inflation hedge. Well there isn’t a practical one is there from this set? There are asset classes that seem to do better than others – cash, commodities, gold but their long term holding costs versus other risk assets means your essentially paying expensive insurance. Commodities is the worst of the lot.
– Interest vs inflation. The trouble is for many asset classes except those with zero duration in times of inflation the negative impact of interest rate rises to counter inflation swamps any inflationary benefits of assets (e.g INLG, $TIPS, to a lesser extent reits per above).
– Equities are not a good inflation hedge. period. They are also the best chance most people have of beating inflation long term. period.
– 10 years is not long term. S&P 500 could be bottom of the pack quite easily in next 10 years given elevated CAPE etc
– Bonds really should do better this year. But a negative real return for global inflation linked bonds over ten years is not great at all given why most people invest in this class
There’s a lot to be said for investing in a global tracker with cash for liquidity and doing something else with your time!
Just to be more annoying than Hari, my favoured asset sub classes to outperform over the next decade relative to S&P500:
European equity income, Asian and EM equity income, investment grade corporate bonds (US and UK), US junk bonds, EM $ denominated govt bonds.
Gorgeous quilt, although it took awhile for my brain to recontextualise away from Red Bad / Green Good and focus on the placement of the squares instead. Maybe a little graphic on the side showing the Y axis would help. Alternatively, once your 10 year rankings get updated you can then colour grade the different asset classes from best to worst – not as if we’d remember that eg. Gilts were green last year.
Great work!
It was the Callan chart of various US stockmarket investments returns that I found many year’s ago on a Vanguard Diehard/Bogleheads blog that finally convinced me that I knew nothing!
Presumably where you got your Quilt idea from?
Apparently Global Equities and Global Bond (hedged to the pound) index trackers were my only requirements
Leave well alone to compound
Some cash for 2-3 years living expenses and that’s it
Worked-so far!
xxd09
Assuming the global equities ETF includes the US, which would skew it significantly, the quilt shows how much the US tech bubble has affected the last decade. It’s deflated a bit as a result if interest rate rises, but there’s still a lot of air in the bubble.
I’ve no idea what will win in the next decade but bonds look a lot more attractive now than a year or two ago.
Just reinforces my belief that the best thing to do is buy the global equity tracker, then adjust risk by holding FSCS protected cash deposits at the best rates available (or gilts to maturity).
– Simple
-You will never find yourself at the bottom of the chart
– Helps to protect you against FOMO
and you own misguided belief that you can predict what is going to outperform
@Naeclue – agreed, and along with that, limit your main portfolio updates to quarterly or even semiannually. Although my VWRP holdings have been up and down, what they haven’t been is bottom of the table – meanwhile after a nice run up my “play account” for naughty active investing is currently down around 20%. Woe, woe – except it inoculates me from doing anything “creative” with my allocations. I’ve mentioned before I’ve started holding the local equivalent of gilts here in SG. Let’s see if that further stabilises the ship – it seems the forecast is for more stormy weather in 2023.
Ha, yes, I expect hedge funds around the globe are loading up on European equities as we speak.
I’d like to put corporate bonds in the table I think, and multi-factor equities, but I’m running out of colours!
@ Mr jetlag – nice idea about the key. I suppose the 10-yr returns fulfil that function but they’re on the wrong side.
@ Seeking Fire – I broadly agree – there isn’t a worthwhile inflation hedge if you’re an accumulator – assuming inflation is brought under control relatively quickly. As a decumulating retiree the ‘insurance’ premium may well be worth paying to avoid getting ravaged by 1970s scale inflation. The perfect instrument is index-linked certificates – which of course we can’t get anymore – though lucky US readers can still buy I bonds. Failing that it still seems worth looking into ladders of individual index-linked gilts – depending on the price you’d have to pay.
Re: commercial property – I wonder what we’d say if it had just had a great decade rather than a dismal one. When I eyeball property against global equities across the table – it’s definitely doing something a little different. Only once in the decade did that work out well however – 2014 – when property returns were double global equities. The same thing happened in 2012 too – the year that’s just dropped off the quilt.
There’s no inherent reason for property to be so poor – except IIRC valuations were extremely high within a couple of years of the Credit Crunch. And the pandemic hasn’t helped.
I think you’re spot on about property being no diversifier at all in a serious bear market. The two asset classes are highly correlated and there are academic papers out there that show property and equities fall like rope climbers tied together during a crisis.
I still have some hopes for property though as an equity diversifier rather than a portfolio diversifier and perhaps it’ll mean revert after recent pummellings.
@xxd09 – yes, the Callan chart is the first ‘quilt’ I remember seeing. I think they called it the periodic table of investing, which I loved.
Interesting (if naughty) to compare annualised returns for some active funds over the same timescale.
From Trustnet data, the two UK biggies: Fundsmith Equity 15.9%, and Scottish Mortgage investment Trust 17%.
Very naughty. Bad Chesterdog 😉
A great comparison would be if we picked 10 different active funds someone tipped a decade ago and made a quilt from their fortunes – for better or worse.
There is a piece by Greybeard somewhere on the site that includes a list of investment trusts he liked the look of many moons ago.
I looked up some of them last year to see how they’d done since. They were all over the shop. Some big winners in his list but some big losers too.
@Chesterdog @TA — My mum set up an investment trust portfolio with two equal sized subscriptions in 2012 and 2013, which was invested with my guidance into various UK equity income trusts. The idea was to create something with a modest portion of her wealth that might do better at providing an income versus inflation than keeping it all in cash, over the longer-term. As things turned out though, the dividends were just reinvested. I directed a bit of modest trading over the years (maybe 10 swaps from one trust to another over the decade).
Messing around with our calculator suggests a rough and ready CAGR of 8%, which I think is much more representative than picking two of the best performing UK funds of the past decade (for a while it wasn’t even close with SMT, but it’s since retrenched markedly).
Some money was taken out to pay for things blowing up (a car or a boiler or something) a few months ago and it isn’t unitized, so it’ll be harder to stab at the CAGR from here.
Needless to say she would have done much better in a global tracker in hindsight (about a 14% CAGR) but as I say she thought she wanted some sort of income at the start. 🙂
It’s been an educational little portfolio for me I must admit. And lately like most UK stocks its been hugely outperforming global equities, so maybe the reversion is on! 😉
@TI (#32):
Great story – IMO real world stuff is always interesting to read. What caught my eye was that: ‘she thought she wanted some sort of income at the start’. Would you be good enough to say a bit more about this and any other lessons?
@Al Cam — Afternoon 🙂 I might write about it sometime, but don’t want to derail this thread further and I typically have to try to stretch my personal input into posts that reach many rather than replying in depth to individual comments for the sake of having time to work and sleep 😉
Nice work, it’s very interesting to see this broken down in GBP.
If anything, the poor run of commodities during most of the decade should have made them *more* interesting… unless one was working from the active prediction that disinflation will last forever.
Volatility really isn’t a problem either as long as it is uncorrelated with the rest of the portfolio. Aggregate commodities have a low correlation with the stock market.
The problem with commodities is how to invest in practice. Commodity ETFs are a funny business, they depend a lot on the shape of the futures curve and can deliver much lower returns than the index they are supposed to track. I found this all too complicated and settled on a chunk of commodity stocks instead. Using a mix of energy, mining and agriculture sector ETFs. It’s less than ideal and comes with a higher correlation with the overall stock market, but worked ok-ish last year.
Has anyone looked into roll yield optimised commodity ETFs (iShares ROLL and such)?
@ Sparschwein – all excellent points. I’d add the volatility of an uncorrelated asset shouldn’t be a problem except that it may well be *psychologically*. The psychic scream that reverberates around Monevator when a major asset class has a bad few months never mind a bad decade makes me think that only a hardcore could live with an asset that inflicts years of misery before hitting pay dirt.
A losing position grinds people down. Sticking with it requires fortitude and faith. Interestingly, this is how Naseem Taleb used to trade and IIRC he struggled to convince his management that it would work.
I think I might take up your challenge on investigating the next generation of commodity ETFs. I’ve looked into commodities in depth but not written about it on Monevator. I kinda got the impression nobody was interested. Not that that normally stops me!
Three things have kept me out of commodities:
Underlying problems with commodity futures funds as you mention.
The balance of the academic research I read was against including commodities in a diversified portfolio.
I could use gold to do the job: low correlations with equities and bonds, no structural problem with using index trackers to gain exposure.
@TA – thanks, I think it’s worth turning every stone to find better diversification. It’s a real problem for anyone who is concerned about stock market risk. The dispersion of outcomes is huge even over the very long-term, when calculated without the usual biases (this podcast was quite the eye-opener https://rationalreminder.ca/podcast/224 ). Bonds too are less reliable in real terms than commonly thought.
Crude hedging with puts or VIX futures is just too expensive. If I had access to a good tail hedge like Taleb’s Universa, I’d go up to 90% in stocks.
There is probably a way to do this buying deep OTM puts and selling calls – figuring this out would be a full-time project for FIRE times…
“Cash would be a nice baseline to have in the table. I thought about doing it using some kind of money market fund or ultra-short bond ETF.”
You might find it reasonably easy to use the returns on Premium Bonds, either including or excluding the huge prizes.
There would be the complication of the tax-free nature of the returns.
I was about to post some smart arse remark along the lines of just bung it all in a global tracker, but just spotted that I did that in the comments from last year!
It seems odd that you have left out European equities. A bigger market than Japan, UK and global EM. Vanguard FTSE Developed Europe ex-U.K. Equity Index Fund or the ETF equivalent VERX would be suitable candidates. 10y annualised GBP returns 7.93% for the funds, 8.04% for the index, just below gold in your quilt.
ps ERNS for cash? Not the same, but reasonably close. 1.17% 10y annualised.
Interesting update, surely mid/long inflation linked bonds might have had good 10 year returns in 2021 but because of that and the deeply negative real interest rates they were incredibly unattractive, following the large falls of 2022 they are potentially moderately attractive..
Whilst mid/long US$ tips with 2%+ real returns going forward over the next few years may well turn out to be satisfactory.
Thanks for the update @TA, and many thanks in particular for not including (as I’ve seen included in US $ versions of the 10 year quilt) BTC as a reference ‘asset’ in there. Nothing so muck irks me as the hypothetical returns I notionally ‘turned my back on’ because it was insanity squared 😉
There’s no pattern to returns from one year to the next across the asset quilt, but it is fair to say, I think, that over shorter periods (3, 6 or 12 month look back and 1, 3 or 6 month holding periods, with shorter holding periods for longer look backs) different assets do tend to show cross sectional (i.e. relative) momentum; and trend following funds like Winton do make money off of that, with returns often being negatively correlated with equities.
As (IIRC here) @Algernond has suggested in comments on other threads, trend following funds might be (in some sense, and within some limits) a ‘safer’ way to access broad commodity and/or gold exposure than a static allocation to either, albeit that the fees for these types of funds are little short of outrageous, especially as they’re quntative and rules based, and so don’t rely on active management/qualitative skill.
The older comments on posts like this are so interesting to read with the benefit of hindsight. The feeling that US equities can’t go on leading the pack forever seems so reasonable particularly after last year then you’re reminded we felt that way this time last year and every year before that. It’s just a good reminder how bloody difficult this game is! SPX 4,756 // ~19x forward earnings at time of writing is that expensive? I guess it will seem obvious to those reading this when TA publishes the 2024 quilt…
I think Churchill was talking about Americans when he said they’ll always do the right thing after trying everything else but he may as well have been describing my investing.
~75% of the portfolio to a permanent global index holding from here on out for me. Limit myself to a 25% pot to injure myself with
I’m now wondering how hard it would be to write, backtest and adjust the weighting of a Perl script to issue buy/sell instructions to run a DIY trend following fund.
Does anyone have a perspective on how frequently it would need to trade? I’m not desperate to have to write automated trading code.
@Aol — I agree. We’re sometimes asked why we don’t delete all the old comments when we update articles on Monevator — standard practice among many publishers to keep things looking fresh and certainly to reduce confusion — but your comment sums up why I usually leave them in place!
There’s whole bushels of old comments on Monevator that have a second life as a cautionary tale, or offer some other kind of lasting perspective. And a few retrospectively regrettable articles by myself, too, to be completely honest. 😉
I love looking back at old comments and laughing at my mistakes – for instance, my naughty play account sat idle(ish) in US T-bills during one of the best Santa rallies ever! Still up for 2023 though, as most seem to be.
Glad this is an annual feature now too.
@xeny – excellent books by Rob Carver (Leveraged Trading) and Andreas Clenow (Following the Trend) on how to implement this. Rob Carver also has a good website on how to do it also: qoppac dot blogspot dot com.
(…And Clenow also has a book ‘Stocks on the Move’, which is about momentum trading.)
They are both talking about Medium / Long term Trend Following, so automation is not a must. What is essential is reliable daily price historical data.. and sticking to your rules.
@TLI – you remember correctly (mostly). I do use TF funds now, and because of that don’t hold long-only positions in Commodity or Bond funds; I do still have long-only Gold though (and actually the defensive ITs I hold do have long-only bonds of course).
Fees / availability / watered-down leverage (OIEC) are the main reasons I have to continually think about if my ~25% position in the TF funds is sustainable….
Hi,
I’ve been looking to add the Royal London Short Duration Global Index Linked Fund R Acc to my portfolio, but the platform that I’m currently on (iWeb) only has the class M income version of the fund. Not ideal as I know the acc version of a fund is much more hands off if you are in the accumulation phase.
Regards,
Calum
@TA:
As ever, very much appreciate the effort it takes you to produce the UK blanket. Appreciate the addition of MMF (as a cash proxy) too.
Thanks again.
Years ago I saw a finance man say that his firm viewed gold as FX and silver as a commodity.
What part of “foreign” does gold relate to? I suppose to the United Kingdom of Great Britain and Ireland in the latter half of the 19th century. Looked at like that, holding some gold seems pretty reasonable. Mr Gladstone and Dizzy might agree were they available for interview.
Interesting analysis, thanks again.
> Splitting your defensive measures between nominal bonds, index-linked bonds, cash, commodities, and gold
This makes so much sense, and those who followed this had a smoother ride through 2021/22.
I’ve been wondering recently if there is any point in holding nominal UK govt bonds. I can see two jobs for bonds in the portfolio:
1. As counterweight in a stock market cash. US treasuries are better for this esp. when unhedged. UK bonds are more “risk on”, and GBP is more “risk on”; and (according to a Reuters article from Monevator’s weekend reading) this trend has increased recently.
2. As a risk-free store of value, to lock in the decent real interest rates. The risk-free asset for the long term is duration-matched linkers (well except for the possibility of govt default…). Cash and nominal bonds can be decimated by inflation, which is a major risk for retirement.
A chunk of linkers will also help somewhat as ballast in a stock crash.
So I don’t really see a case for nominal bonds – but interested to hear other opinions!
(Assuming a UK-based investor, no leverage, and corporate bonds excluded because they tend to be too correlated with stocks.)
@dearieme – I “did my own research” (I know…) before adding a chunk of gold some years back. Conclusion was that in terms of portfolio building /diversification, gold is its own thing, unlike any other asset class.
When I slice my portfolio by Fx exposure (a costly lesson from Brexit to watch this), then gold counts as a separate currency. Commodity ETFs count as USD.
Thanks for the great update @TA, and especially for putting in the leg work to get the £Stg equivalent returns that one doesn’t see elsewhere, with the USD dominated return quilts which dominate online.
“Splitting your defensive measures between nominal bonds, index-linked bonds, cash, commodities, and gold”: could trend following/ managed futures and global macro HFs also be considered to be legitimate defensives here?
The inclusion of Money Market Funds as a proxy for cash is interesting.
Based on my experience the annual returns on MMFs is less than the return on best buy savings accounts from 2015-2021 and from 2022 to 2024 slightly more I would suggest. This observation is based on my average savings rate from best buy savings accounts (this ignores some regular savings accounts I have held so slightly understates the average). Plotting these gives this comparison chart
https://ibb.co/9mnWMWCj
The geometric average over 10 years to 2024 for the MMF is 1.5%pa which compares with my best buy savings 10 year average of 2.2%pa.
Over 20 years my best buy savings accounts have returned 3.5%pa which compares to RPI which has averaged 3.8% (geometric averages). That failure to match inflation is mainly down to the roughly 3 year period from 2021 to 2023 when RPI inflation exceeded best buy savings rates by some margin. Around 2017 to 2019 RPI inflation also exceeded savings rates by a small margin. Those bad years for savings more than cancel out that best buy savings have returned more than inflation otherwise.
@Snowman (#54):
Very interesting details, thanks for sharing them.
FWIW, I was somewhat comforted by the MMF figures (esp. for the last ten years) as they are in the same ballpark as my cash IRR over a similar period. Whilst best buy GBP cash a/c rates are interesting, I keep cash in a variety of a/c’s – some of which have/had [good] fixed rates [for years] and some of them do not pay any interest! Hence my interest in the overall cash GBP IRR.
@The Accumulator Could you provide a copy of the spreadsheet (or at least the text) so fellow number crunchers can have a tinker with hypothetical asset allocations? I’m particularly curious about going equal weight on all of these funds, rebalancing each year.
Could you also provide the links to the funds that you’ve used for reference? Again for possible investment purposes.
Thanks.
@Sparschwein – the case for nominal bonds is they’re better than linkers as a counterweight to falling equities during a crash. Moreover, UK linkers have the rare weakness of no deflationary floor, so nominal bonds are a hedge against the UK “doing a Japan” or suffering a 1920s/193os style depression.
That said, I entirely see where you’re coming from. Nominal bonds now figure much less in my own thinking. And there are strong voices out there that favour just using index-linked bonds. That’s easier advice to follow now real yields are positive and individual linkers can be traded online.
@DH – Cheers! The question of what should be on the menu really interests me. I’d be really interested in your personal definition of legitimate defensives.
As a starter for ten, mine is a balance of:
– Accessible by retail investors
– Long track record / solid evidence base
– Not dependent on information asymmetries
– Doesn’t require constant monitoring / active market participation
– Can be readily understood by anyone on Monevator – even if that takes reading several thousand words of interminable wandering in the weeds 🙂
@Snowman – nice work! I haven’t seen anyone else float a long-term number for Best Buys. Really interesting to see Best Buys outstrip MMF over the period. Would you put a cash ceiling on your achievable Best Buy interest rate? I guess in my own case, the best rates I could achieve after 2010 were in current accounts that capped out at £2 – £5K. I’d need to go back to check how much I was able to squirrel away at a decent rate so it’d be interesting to know if you have a number to hand.
I’d like to write a data-led piece on cash as an asset class soon – so I’d like to find some credible numbers for Best Buy rates to go with proxies like MMF / Treasury bills.
@Trever – sure thing. Here’s a copy along with links to the ETFs:
https://docs.google.com/spreadsheets/d/1FaJIjykv1FhzHVoUr2vm6tdEr4leSJ1t-uJxV6Lr1ko/edit?gid=1706459502#gid=1706459502
@TA:
FWIW, IMO: tracking [just] cash a/c best buy rates is a bit misleading for all the reasons (plus some others) that I gave above at #55. In my case, at least, the ebbs and flows are sometimes every bit as important as the headline rates, hence my preferred use of IRR. For convenience, I do not track every cash movement but definitely any significant ones and overall [suitably adjusted] annual deltas. In some cases, exchange rates might also be a consideration too.
Talking of cash – i tried to post a link but no luck and its not a recommendation of the product.But purely from a historical point of view of who has been the more steady payers and also what rates were on offer over the last Ten years then searching “Family Building society how our market tracker saver accounts works” might be of interest (no pun intended)
The product page has a feature that shows the last 10 years broken into quarters with the top 20 easy access accounts and their rate from each period from a pool of approx 100 – 120 providers.
@klj — Apologies if I’ve spammed away a legit post. I have to trawl the spam folder 4-8 times a day, and sometimes (like today) I do so without my glasses! (On a treadmill at the gym…)
I think my pattern matching is pretty good but it’s possible I missed a real comment that needed salvaging, especially if you also filled in the URL field.
I have an overwhelming temptation to print off your quilt, stick it on my dartboard and see if I can do a prediction for next year’s top 3 …
@ The Investor – no worries all good.It was more that as a “long time listener,first time caller” i could not work the link out (and i had my glasses on) And i was also thinking that it might look like a first post plug with a link so omitted it in the end
@Weenie – haha – let us know what the darts say. Meanwhile, TI is prototyping your idea in his Monevator Merch Lab as we speak.
@Al Cam – Yes, the baseline for cash is money market / bills. And money market funds do solve the “loyal customer” problem of being condemned to uncompetitive interest rates because you’re not a rate tart. But still, it would be interesting to have a working number for Best Buy rates. Would let us know how much is potentially up for grabs. Perhaps we can think of it as factor investing for cash? 🙂
Either way, different folks could pick the number that better fitted their own personal habits.
@The Accumulator
I’ve got a record in my finances spreadsheet of almost every savings account I’ve had since about 2004 (with the exception of some regular savers and a few bank accounts which paid good interest). So I extracted the interest rate (AER) and balance at the 1st of every month from that. And hence I could calculate my average interest rate (which is a weighted by balance average of those interest rates).
I’ve also not seen anyone else attempt to quantify a best buy savings account history over 20 years. The swanlowpark.co.uk website has Interest Rates on UK Savings Accounts from 1980 to 2021 but they seem to be lower than my figures so can’t be best buys
Were you asking for the highest average I’ve had? The 1st of month all time high AER was 6.24% pa on 1st October 2008 with Icesave, Kaupthing and AIB being the main contributors to that rate. Icesave collapsed 6 days later!
Happy to send you my monthly data which shows the providers and interest rate (AER) at 1st of each month for the past 20 years, and my (weighted by balance) average savings rate. It’s very interesting to plot these averages together with RPI and bank base rate on a chart. For example it shows how inflation when it got above 10%pa significantly devalued savings in real terms.
I mentioned earlier that over 20 years RPI has averaged 3.8%pa and my average savings rate was 3.5%pa. To add to that base rate averaged 1.8%pa over those 20 years. I did consider a long term expectation of base rate + 1.5% as a proxy for best buy savings, but I don’t think it’s a good proxy as my average savings rate has varied between base rate – 1% (in September 2006) and base rate + 4% (in May 2009 when base rate was 0.5%). At the moment only approximately base rate + 0% is achievable.
@TA: #57: “legitimately defensive”: long and established track record of negative to low correlation with equities. That’s the aim I work back from.
So BTC/ETH/SOL/XRP etc would each fail not on the grounds of their opacity, illiquidity or complexity (although they arguably suffer these) but because Crypto is (at least since 2018) a demonstrably risk on, highly correlated to equities ‘asset’ (correlated especially to the NASDAQ).
It’s also nice if the defensive comes with yield, but that’s a preference not a necessity. GLD will do here provided that it actually does zig when shares zag.
Simplicity is a virtue in asset allocation but, like all virtues, it can become a vice if overdone. So trend following and listed global macro make it onto the list for me provided always that they live up to their billing of giving an at least uncorrelated if not an anti correlated return profile to shares.
@Snowman #62: re: “over 20 years RPI has averaged 3.8% p.a.”: cumulative 73.5% CPI (2.8% p.a.) for 2005-24 from the BoE website. That’s a huge difference over 20 years to RPI. Don’t have sense of cost of living being just over twice as expensive now as in 2005. ~1.7x seems more aligned
Interesting article. The point for me is that there is no way to predict which asset will be top of the league next year! There might be some mileage in picking the bottom of the league this year and investing. Sort of “Dogs of the Dow” type of strategy but it would take nerves of steel. Might be worth a shot with a small proportion of my portfolio but being a risk averse type of chap I would probably back test the strategy first.
My own portfolio actually contains 7 of the assets detailed in the blanket. The main missing assets are Japanese equities, emerging markets and inflation linked gilts. I do have short term gilts (not funds but actual gilts) and long term gilts. The short term gilts mature in 2028 which is the age I get the State Pension. The long term gilts mature in 2054, when I may have reached the ripe old age of 93………..difficult given the state of the NHS and the reliance of us oldies upon it.
@klj: Thank you for the tip about the Family Building Society’s Market Tracker Saver accounts. Never heard about a product like that before.
@TA: As already pointed out, their interest rate data might be useful for your future cash post.
htt ps://w ww.familybuildingsociety .co .uk/savings/easy-access-savings/tracker-savings-account
@klj (59) @Winter Mute (66)
I have a small Cash ISA that is Market Tracker with Family Building Society (formerly National Counties Building Society) more or less by accident. It derives originally from a TESSA, via a TESSA-only ISA to eventually a plain Cash ISA. It has been transferred through numerous institutions and eventually landed as a fixed-rate ISA with NCBS. After a couple of iterations of fixed-rate deals with them it ended up as a Market Tracker. I’m reasonably happy with it – it is currently paying 4.55% AER (was 4.7% before 1 Jan).
@Delta Hedge
I wouldn’t want to digress into a long discussion as to why I use RPI to assess returns rather than CPI or CPIH but I will say a few things. I agree choice of inflation measure is important, it’s just which to use is hard to say.
Firstly personal inflation is very much individual, and very difficult to measure because of fluctuating expenditure and one off increases in expenditure and one off increases in past expenditure (for example a one off switch to buying better quality food) which may not be reflected in the future.
The Royal Statistical Society’s response and discussion into the effective replacement of RPI with CPIH consultation gave some good arguments as to why RPI could be considered a better measure of personal consumer inflation than CPI.
I think most agreed the housing cost element calculation in RPI needs reform and while the CPIH housing cost element through notional rent isn’t relevant to owner occupiers it is still less distorting than the RPI housing cost element that can be significantly affected by changes in house prices.
But the use of the Carli method (as opposed to Jevons) at the first level of aggregation in the calculation of RPI is not remotely as flawed as the ONS suggest when measuring inflation that consumers face rather than macroeconomic inflation. That’s a very technical issue with arguments on both sides.
I sit in the middle ground in terms of whether RPI or CPIH is a better measure for me to make a guess as to how my personal expenditure will increase, which is the purpose of me adjusting returns by inflation. I merely take a pragmatic approach that in the light of uncertainty it’s better to include some margin by using the higher figure, usually RPI (by perhaps 1%pa) in the long term.
1 x thumbs up for Weenie’s comment.
Or it could be an annual Monevator competition (a bit like the FT one of picking 5 stocks each year). The person who gets the top 3 categories in the correct order for 2025 wins an ‘I love passive investing’ T-shirt.
(My stats is a bit rusty but I make the probability of doing that (assuming an equal chance for each category) as 1/2184. Happy to be corrected.)
@Snowman (#68):
Interesting comment – which I largely agree with.
FWIW, after some study I came to this conclusion:
Inflation is a useful macro-economic concept but it has limited explanatory powers – especially at the level of the household/individual where subtle changes can be much more impactful. Also, inflation is essentially a latent variable, that is: you cannot buy an inflation meter, plug it in somewhere and take a reading. That is, inflation is a useful economics concept, whose utility is well understood (pun intended) but it is no more than that!
I am intrigued as to what swayed you towards selecting RPI to represent “how my personal expenditure will increase”, specifically:
a) do you have any quantitative evidence for what you think the “margin” you are building in might be in your case, e.g. might it be more (or less) than the oft quoted 1%PA in the long term,
b) what anecdotal evidence are you using,
c) what will you do after 203o when the current RPI methods, etc will be effectively consigned to history
FWIW, I started out several years ago from a similar position (re RPI) but now prefer CPIH.
That CPIH has drifted towards RPI over the last year or so (so much so that Dec ’24 monthly values are the same) is an interesting development?
@ WinterMute – I remember having a similar product with Nationwide about 20 years ago but guess they phased it out? I only noticed this version last week.
@DavidV – Thanks for your feedback on using it and good to know you seem happy with it and the society
As well as swanlowpark (already mentioned – do check the linked sources, because these are not necessarily ‘best rates’) there are a number of other sources for rates on cash accounts
The Building Society Association yearbook (https://bsa.lansdownepublishing.com) has data going back to 1939 (although somewhat sparse before about 1955)
The BSA website has data for a range of accounts going back to 1999 (which I think are at https://www.bsa.org.uk/statistics/savings-statistics – their site is currently down)
The Barclays Equity Gilt study has Building Society account rates going back to 1946 (although I find it difficult to believe the early entries because they are inconsistent with the BSA handbook and bank rate and closer to the mortgage rates in the handbook)
Moneysaving expert has an archive of the weekly emails (https://www.moneysavingexpert.com/tips/ – the absence of a ‘last’ button is really annoying for that archive – although a search can get your to the beginning) that have best savings rates going back to 2004 (at least that is the earliest date I have extracted).
FWIW, in order to extrapolate to before a 1951 start point, I found using a combination of the above data sources that savings account interest rates, S approximately fitted a model such that S~0.89*Y+1.14 where Y was a combination of the bill rate and long-term bond yield (70% bills and 30% bonds). At some stage I am going to have trawl through old newspapers to properly extend the cash rates back in time since, AFAIK, no archive currently exists. I note that annualised real returns from the extrapolated savings account data (so 1871 to 2022) were about 1.9% compared to 1.7% for long-bonds and 1.1% for bills.
@Snowman – it’d be great to see your data when I start researching the piece. Thank you! That’s really helpful.
My question around Best Buys is connected to the ZIRP period after the Financial Crisis. The best rates I could find at the time were pretty much all current accounts / regular savers that strictly limited the amount you could save at higher rates. So the more I saved, the lower my average savings rate as the Best Buy opportunities thinned out.
Which complicates the analysis for that period i.e. the average savings rate achievable with Best Buys was higher if you had, say, £10,000 put away rather than £100,000.
Whereas there’s ostensibly no ceiling on MMF savings.
So I was just putting that out there and wondering what your experience was?
I guess this may only be a problem during the ZIRP era but I don’t know? It’s not a problem now at any rate. The max amounts on most of today’s Best Buy savings accounts are north of £500K!
@Klj and Winter Mute – thank you for the Family Building Society tip!
@Alan S – that’s fantastic. Thank you!
@Alex. Close but no cigar I believe. You don’t care about the order of the 3 chosen so apply a 3! multiple gets you to 1/364.
As you say, good idea though.
@A1 Cam
I don’t really have anything insightful to say in reply to your questions. But using your lettering
a) In terms of the 1% difference I’m going off comparing historical data myself for CPIH and RPI, or as quantified here with a breakdown of factors in figure 1 for example https://www.ons.gov.uk/economy/inflationandpriceindices/articles/shortcomingsoftheretailpricesindexasameasureofinflation/2018-03-08
b) I don’t have any real anecdotal evidence or good feel for my own personal inflation, I’ve just using RPI to provide a margin against over-estimating real returns
c) Post 2030 I will just use the new RPI (i.e. CPIH in reality). If I’m buying an index linked gilt say I will just require a slightly higher real return than I might have required had it been index linked to the old RPI methodology
@The Accumulator
From a quick look I can’t see any of my average savings figures were particularly affected by high rates on restricted balance accounts. The few regular savers in there haven’t been obviously material at any point, and I don’t think there are any high interest current accounts recorded in there as I only really kept records of savings accounts.
I did have some sizeable amounts in some 5 year fixed rate accounts with Northern Rock and Newcastle Building Society taken out around 2009-2011 at rates of 5% or just under. These accounts had an option to withdraw penalty free with 90 days notice. But rates came down and so I didn’t withdraw and they were paying at above what easy access accounts were subsequently paying and so inflated my average savings rates up to about 2016.
Thanks, Snowman. All good food for thought. I need to go back through my records and see what my own experience was.
@Snowman (#76):
Thank you for taking the time to reply.
Despite now believing that trying to determine your own personal/household inflation rate is really a folly (as there are just so many idiosyncrasies and subtleties at the individual/household level) I have been running an experiment along these lines since I retired*. The results of said experiment might just imply our household inflation [CPI & CPIH] is slightly lower than the official inflation measures.
As I say the the signal is weak, but if it is present then this would imply that were I to use RPI instead of CPIH our “margin” would most likely be a bit larger than the oft quoted 1%PA.
*in many ways conducting said experiment for the last eight years has firmed up my views on the impracticality of measuring the personal/household inflation rate; albeit a helpful/attractive concept
@Alan S (#73):
Interesting formula and notable that long term savings account data (1871 to 2022) topped long-bonds and bills.
Did you ever get a chance to look at the retirement risks and opportunities post that I linked to a few weeks ago, see: https://monevator.com/weekend-reading-politics-yields-bad-news-for-bonds/#comment-1856670
Any chance next time you could please include Global government bonds (hedged to £GBP) in the quilt ? It is an alternative to or augments UK intermediate gilts in the defensive portfolio. Suggest the JP Morgan GBI Global Index as the representative index.
Yes, would be interesting to see. Maybe I should consolidate the FTSE 100 and 250 into a single FTSE All-Share block and then introduce global bonds.
Oh, if you’re accepting suggestions, could I proffer the MSCI categories:
Europe (or EMEA) ex UK
Asia Pacific ex Japan
No offence if not of course, it is your quilt after all.
Running a quick calc of my equity and bond allocations by the latest returns on your quilt does match my return for the year up about £250k. That’s a lot faster than I’m spending at present but I am casually looking over house adverts.