Many DIY investors have given up on bonds. They’ve thrown their lot in with money market funds instead. I think that’s a mistake.
The evidence suggests that replacing bonds with money market holdings is liable to suppress portfolio returns and leave you under-diversified in the face of future stock market crashes.
Let’s see why.
Money market vs gilts: five-year returns
Our first comparison pits a money market ETF versus an intermediate gilts ETF in a cumulative nominal return head-to-head:
Investing returns sidebar – All ETF returns quoted are nominal, GBP total returns (including interest and fees). All asset class index returns are annual, inflation-adjusted, GBP total returns (including interest but not fees). ETF returns data and charts come from JustETF. Gilt annual returns are from JST Macrohistory1 and FTSE Russell. Money market annual returns are from JST Macrohistory and the Heriot-Watt / Institute and Faculty of Actuaries / ESCoE British Government Securities Database. UK inflation statistics are from A Millennium of Macroeconomic Data for the UK and the ONS. May 2025.
Strewth, intermediate gilts lost 28.4% in the past five years! And that’s without trowelling on extra misery from inflation, which the data provider doesn’t incorporate into its graphs.
The real terms loss is more like 38%.2
So much for bonds’ reputation as a ‘safe’ asset.
The money market also inflicted a 9.6% real-terms loss too – but that’s only a quarter of the kicking meted out by bonds. One in the nuts rather than four-times in the nuts.
I find it easier to compare real annualised returns when assessing investments, so I’ll translate the ETF results into that format as we go. (I’ll use inflation-adjusted annual index returns to continue the match-up all the way back to 1870.)
Here’s the real annualised returns for the past five years:
- Money market: -2%
- Intermediate gilts (All stocks): -9.2%
Money market wins!
Money market vs gilts: ten-year returns
We’re supposed to care more about the long term, right? Our investing horizons ought to be counted in decades not a handful of years.
Let’s zoom out to the past ten years, the maximum time frame offered by most data houses:
Do I hear: “So you’re telling me that gilts lost money over the last ten years? I’m out.”
Meanwhile, money market funds – popularly billed as ‘cash’ – are up 15% in nominal terms.
(Incidentally, money market funds are ‘cash’ in the same respect that bonds are ‘safe’. Read that article for more.)
Real annualised ten-year returns:
- Money market: -1.5%
- Intermediate gilts (All stocks): -3.6%
Money market wins!
If you can call a loss winning.
Money market vs gilts: 15-year returns
Let’s keep going. If money market funds are the superior product then they should dominate beyond the last decade. Ten years is nothing much. We overweight its importance due to recency bias.
Well, this complicates the picture.
If you held both ETFs in equal measure for the past 15 years then your money did better in gilts – despite the enormous bond crash of 2022.
Real annualised 15-year returns:
- Money market: -1.8%
- Intermediate gilts (All stocks): -0.9%
Gilts win!
On this view, money market funds were twice as bad as gilts over the last 15 years.
Mind you, gilts still turned in a decade and a half of negative returns. Nobody comes out of this looking good.
Money market vs gilts: 18-year maximum ETF timeframe
The easily-accessible ETF data runs out around the 18-year mark. Money market funds are only lagging further behind at this stage:
Gilts returned 64% more than money market funds over the entire period that both asset classes became accessible via ETFs.
Real annualised 18-year returns:
- Money market: -1.4%
- Intermediate gilts (All stocks): -0.1%
Gilts win again!
True, 18 years worth of negative returns for both asset classes is a poor show. There’s no denying that.
Over the longer run though, they still both offer the expectation of a real-terms gain, which is why they have a place on our list of useful defensive diversifiers.
Thrive or dive
Money market returns were undone over the 18-year view by the period of near-zero interest rates triggered by the Global Financial Crisis (GFC).
Meanwhile gilts were scuppered by the abrupt return to interest rate ‘normality’ as central banks fought post-Covid inflation.
Lost decades happen. That’s the nature of risk.
We’ve documented such wilderness years for equities and gold:
- The Japanese stock market crash
- European stock market returns – check out the French! On second thoughts don’t. It’s too scary.
- UK and US bear market recovery times
- Is gold a good investment – including gold’s biggest market crash. Grim reading for gold bugs.
Nothing is ‘safe’. Every asset class can destroy wealth. That’s why the likes of shares offer you potential returns high enough to beat cash in the bank.
Because we can’t know which risks will materialise in the future, we diversify our portfolios by holdings assets that respond differently to varying conditions.
Not diversifying tempts fate like a farming monoculture. It works until it doesn’t and then failure can be catastrophic.
Keeping hold of what you have
It’s especially important to diversify your defensive, non-stock assets as your pot grows to a significant size. Preservation becomes as desirable as growth, psychologically, once you cross a certain threshold.
The growth side can still be adequately diversified by a single global tracker fund.
However defensive asset allocation is trickier, and neglected because it is complicated to execute, suffers from industry over-simplification, and is less well understood by the public at large.
To be fair, it’s not an easy problem to solve. I guess that’s why many people are throwing up their hands and dumping everything in money market funds.
But I digress.
Money market vs gilts: 125-year returns
Let’s finish off our money market versus gilts drag race. We don’t need to stop after 18 years. We can keep comparing bonds and money market returns all the way back to 1870.
If money markets really do beat govies then they’ll be back in the lead before long, eh?
| Years | Money market real annualised returns (%) | Gilt real annualised returns (%) |
| 20 | -1 | 0.1 |
| 30 | 0.5 | 2.4 |
| 40 | 1.7 | 3.6 |
| 50 | 1.2 | 4 |
| 100 | 0.4 | 1.5 |
| 125 | 0.4 | 0.8 |
Turns out there is no truly long-run timeframe (beyond the past ten years) over which money markets beat government bonds.
Indeed gilts offer twice the reward of money markets if we take the 125-year average as a yardstick for expected returns, which is a reasonable thing to do.
If we were comparing equity returns, which asset class would you invest in? The one that did better over the last ten years? Or the one that delivered twice the return over the last 125?
Why is it different for money market funds versus bonds?
Why have bonds been cancelled?
The trouble is this happened only yesterday in our cultural memory:
The bond crash of 2022 rendered gilts toxic in the minds of many who lost money in it, or those who see its backwash polluting the trailing return figures.
In contrast, money market funds came good over this short period. (Albeit after delivering 12 years of negative real returns in the previous 13 years.)
There’s a straightforward explanation for this reversal in fortune.
Steep interest rate rises (as per 2022) batter longer duration securities like intermediate gilts.
But they boost money market funds because such vehicles are chock full of short-term instruments that quickly benefit from higher rates.
Short versus long durations
The simplest analogy is fixed-term savings accounts.
If you knew interest rates were about to rise then you’d surely hold very short-term fixed savings accounts beforehand – or better yet, easy access. This way, once interest rates rose, you’d only have to wait a matter of days or weeks to switch your dosh to a bank account offering a plusher rate of return.
But what if interest rates were about to fall and stay down for years?
Then you’d want to lock up your money for as long as you could. You’d know the banks were about to pull their best offers and replace them with stingier ones.
Money market funds are the equivalent to easy access bank accounts in this analogy. They’re the fixed income place to be when interest rates rise, but the place not to be when they fall.
The rub though is that none of us know the trajectory of interest rates. Even the experts fail to predict the future path of interest rates with any reliability.
This is part of the reason why it makes sense to hold both bonds and money market funds. (Or straight spondoolicks instead of money market if you can squirrel enough away into cash ISA boltholes.)
The last five years of fixed income returns are dominated by a nasty sequence of interest rate hikes. Hence money markets won.
But the main event 17 years ago was unprecedented interest rate cuts to near-zero – intended to defibrillate Western economies in the wake of the GFC. Hence money markets lost.
Signal to noise ratio
Trailing returns are shaped by the events that they capture.
The shorter the time frame under review, the more likely it is to reveal only the singular events it records – while telling us little about the mean behaviour of the asset class.
Extraordinary events may not repeat in your future.
I was listening to a podcast recently that claimed business investment was suppressed in the 1950s because people assumed World War Three was all but inevitable given their recent experience.
The important thing about the 125-year record is it contains most of the information we’ve gathered to date on money market funds versus gilts.
Such data covers how each asset performed during two World Wars, two pandemics, one Great Depression, stagflation, the bursting of a tech bubble, plus multiple inflationary shocks, recoveries, go-go years, and interest rate cycles.
This long view tells us that gilts delivered much better average returns across the full spectrum of known economic conditions.
If you ever check past performance figures before investing, then this is the timeframe to care about – because if you’re playing the percentages, then 125 years is the most signal-rich comparison we have.
The underlying rationale
Financial theory helps explain why gilts should eventually reassert their return superiority over money market funds.
It’s that risk-reward trade-off again.
Gilts are the riskier asset in that they’re more volatile. Longer duration bonds can suffer violent reversals such as those seen in 2022. They also frequently deliver double-digit returns, for good or ill.
Double-digit gains and losses are comparatively rare for money market funds. They’re more stable, like cash.
But over time, there’s a price to pay for stability – a lower long-term rate of return. (Also known as cash drag.)
We invest in equities because they’re risky, not because they’re easy to live with. We want to pocket the greater reward that we can reasonably expect for taking this greater risk. Every DIY investor who knows what they’re doing has bought into this.
So why not with bonds?
Diversify your defences
My real argument isn’t pro bonds or money market funds.
I think there’s a case to be made for both.
How much you hold depends on who you are, your financial situation, and your time of life.
For accumulators, the biggest danger is you’re scared out of your wits and the market by a horrendous stock market crash. Intermediate government bonds better protect you against that fate than money markets.
Later in life, especially as a retiree, inflation is likely to be your fiercest foe.
Money market funds against inflation are like high city walls against early cannon. They’re not a good defence but they’re better than nothing. They typically outclass intermediate gilts in that situation.
Meanwhile, gold is an unreliable ally against inflation.
I personally think older investors should seriously consider allocations to individual index-linked gilts and / or commodities and / or gold.
That way you’re defended by multiple layers of fortifications when the inflationary enemy is at the gates.
Take it steady,
The Accumulator
Pah! Equity ETFs behave pretty much like equities. By contrast a gilts ETF has quite different characteristics than a gilt. I’ll use gilts rather than ETFs of them.
As for money market funds I’m too ignorant to have a view. Could they be superior to holding cash in Cash ISAs? If so, how and why? If La Reeves plans a “crackdown” on Cash ISAs doesn’t that suggest that they are rather a good thing for the personal investor?
I am sympathetic to your suggestion that codgers should hold IL gilts and gold. I’m half-persuaded too to hold commodities but, apart from a silver ETF, have yet to summon the spine to buy some. What d’ye think? Tin, copper? Coffee, cocoa, wheat, pork bellies, …? Oil, uranium, …? Would an ETF be a pretty good way to diversify the risks here?
Oh, and lest I sound too negative – keep up the good work. You’ve taught me lots, Mr A. Thanks.
@dearieme — Portfolios of gilts behave just like porfolios of gilts, and that is what a gilt ETF is. 🙂
Also I think many people are not comparing apples to apples when they look at gilt funds versus individual gilts. There seems to be some post-hoc belief that the big crashes in gilt values would have been avoided if only they’d owned individual gilts. (Not necessarily in your case, I am speaking generally).
As I’ve flagged many times before, readers should look at this chart of the (now) ten-year UK gilt:
https://www.hl.co.uk/shares/shares-search-results/t/treasury-0.625-31072035-gilt
You’ll notice the price of this supposedly ‘safer’ (I don’t like that word at all in investing contexts 😉 ) individual gilt fell by nearly 40% from 2020 to late 2022.
Holding such individual gilts would not have protected an investor from capital losses from gilts in the bond rout of 2022.
What *would* have helped would have been to hold far shorter-duration gilts, of just a couple of years or less (in retrospect ideally just a few months duration!)
For instance here’s a short duration gilt ETF over the same period:
https://www.hl.co.uk/shares/shares-search-results/a/amundi-uk-government-bond-0-5-yr-ucits-etf
The peak to trough loss in capital terms here was 13%. Much better, because the duration was much lower going into the rout so there wasn’t so much carnage pulled forward from the interest rate reset like there was with the 10+ year gilt.
As TA suggests in his piece, investors would be wise to consider whether basing their forecasts (/prejudices) on the experience of going from near-zero interest rates (never before seen in history) compared to today, when rates are healthy.
The yield-to-maturity on a 10-year gilt in 2020 was near-zero. It as always going to be a rotten investment.
Today the yield-to-maturity on the 10-year gilt is about 4.7%. Very different! 🙂
Another fine (and weirdly timely) article.
Should our appetite for government bonds today be soured by the unwholesome fiscal predicaments of the UK and US governments? Big cheeses in the US administration talk openly about defaulting, and in the UK we are constantly on the cusp of breaking our self-imposed fiscal rules. Are the associated risks appropriately compensated, even at 5 to 5.5% nominal returns (for long term UK government bonds)?
For decumulators, whose arch-nemesis is inflation, aren’t conventional (non-IL) bonds particularly unattractive? If inflation turns up, almost everything in your portfolio takes a hammering (perhaps unless you’ve taken the plunge with commodities), and your defensive asset most of all.
Rightly or wrongly I have a c.20 yr ladder of index linked gilts taking me through to 80. I hope to hold to maturity! The short end (1-3 yrs) looks just fine on a MTM basis. Medium term (4-10yrs) is slightly loss making. The long end is very nasty on a MTM basis. Gains on a 5% holding in gold offset the gilts losses. I also have a decent chunk of equities (30% portfolio) which has performed well. All in all, my portfolio seems to be holding up. What I fear is a real possibility is a “soft default” on gilts.
I think the choice is either divesify or don’t.
For timescales over 20 years (as Fisher Investments noted recently, IIRC) an equity investor would almost always be better (in total return terms) not doing so; but at 20 years or less then diversification increasingly is of benefit.
And if you do diversify then how can you ignore bonds?
The bond matket dwarfs the equity one. It really has to be part of any diversification.
And 10 or 15 year performance tells you little about the next 10 or 15 years.
You need to look at all rolling 10, 15, 20, 25 or 30 year periods over at least a century of data (Dimson and Marsh go back to 1900, Ibbotson to 1928, Siegel to 1802, and McQuarrie to 1793), and over multiple national equity markets.
Take an extreme example here (just to illustrate the point at its most absurd), over the past 15 years BTC has gone from 0.3 cents (the first exchange trade on the now defunct Bitcoinmarkets in March 2010 being $3 for 1,000 BTC) to $110,000 each. But no-one in their right mind would ever expect BTC to do 3.7 bn percent over the next 15 years.
So, by parity of reasoning, why expect bonds (some of which are flat over 15 years) or commodities (down slightly over 15 years) to have no or negative returns?
It makes no sense to anchor expectations on what ‘just happened’ over the last 10 to 15 years.
Now, granted here that there are some pretty worrying developments with the Mad King Donald and the turn away from emulsting Javier Milie (with Musk and DOGE) to cosplaying another Argentine, Juan Peron, with ‘the big beautiful bill’.
And that *might* herald higher rates all round, and possibly poor returns to fixed income (due to interest rate sensitivity and a steepining yield curve/ rising term permium).
But, honestly, if you can precisely predict that one reliably and repeatedly then you’ve struck gold, because that’s a $64 tn question (or, more accurately, a $315 tn one, given that’s the current and rising global debt total – i.e. 3 x worldwide output, with 10% of output globally going on debt servicing now).
I sold my longer duration gilts (vanilla’s) once Reeves talked about a huge £multi-billion black hole whilst at the same time awarding huge public sector pay awards not tied to productivity.
At the time I studied the era from the late 60’s to the early 80’s when yields 10yr treasuries climbed to circa 6% which as a consequence, led to huge drops in equities that took up to 20 years to recover (peak-trough-peak) and this was not until the 10yr treasury yields pulled way back from the 6%.
My point is that I can see high 5’s% on 10yr gilts/treasuries being available again soon as the market in general shuns durations of anything longer than just a few years and it will be at this point that we could really enter a bear market in general equities as the smart money locks in close to 6% for the next decade.
In the meantime for me it is 2-3 yr on vanillas and upto 6-7 years on linkers along with a good slug in Vanguards MMF (in a tax wrapper)
I vaguely remembered reading a Trustnet article from last year with the Royal London team about their Royal London MM fund and their views on comparing it with their Short term fixed income fund which maybe of interest.
No at good at links but a search of “Royal London money market team 19th January 2024 Trustnet” seems to work
@klj:
this looks to be the one https://www.trustnet.com/news/13402449/royal-londons-money-market-team-why-the-market-is-wrong-about-rates
Is there a case for active (gasp) management of bond exposure?
Since interest rates are unlikely to drop (much) below 0%, at that stage it makes not really sense to invest in bonds as losses are pretty much guaranteed. The issue is how to diversify at that level…
Although I have no magic ball it looks more likely than not that rates will keep rising and that credit is too tight right now.
Bonds have at least a better tendency to revert to mean than equities. Something to consider when choosing your poison.
@dearieme – This is the commodities ETF I bought into:
https://monevator.com/best-commodities-etf/
Took me a while to pluck up the courage too. The academic papers on the topic extol the virtues of diversifying across futures contracts, so generally it looks like a broad commodities ETF is the way to go.
GBP money market rates pretty much track the Base Rate as do current accounts – except when your friendly high street bank is dangling a sweet introductory offer 🙂
I know what you mean about there being certain behavioural differences between an individual gilt and gilt funds – holding to maturity and all that. But the returns of equivalent portfolios are the same over time. You can compare the yield-to-maturity of any gilt ETF with its equivalent gilt and they’ll be very similar.
For example:
IGLT is duration 7.5, YTM 4.5
https://www.ishares.com/uk/individual/en/products/251806/ishares-uk-gilts-ucits-etf
8-yr gilt is YTM 4.5
https://markets.ft.com/data/bonds
FT data is tracking a day or two ahead of iShares and the gilt’s duration is less than its maturity. If we factored that in then I bet the 7.5 to 8 discrepancy would essentially disappear. Otherwise, arbitrage time!
@CGT101 – Excellent thoughts! My first reaction is that in my lifetime the finances of the British State have more or less always been parlous. Aside from brief periods when it seemed like we were the ones getting the pints in at the G7 during the ’90s and early noughties.
It also seems to me that the bond market is doing an excellent job at reining in the politicians. Truss got offed, Reeves is boxed in, and the one thing that forced Trump to walk back his Liberation Day madness was a sell-off in US Treasuries.
What we do know is that bond yields represent the market’s current view of the risks embedded in British finances. So, as per the stock market, we’re relying on the wisdom of the crowd to be our guide.
But I agree, the bond market may not care as much about inflation as say a retiree. In which case, that retiree may be better off overweighting (or going 100%) index-linked gilts versus nominal bonds. I completely buy that argument for an individual who sees inflation as enemy no. 1.
The recent flows into money market, OTOH, are driven by recency bias. It’s like watching people rush into Bitcoin when the price jumps.
The difficulty of assessing performance of gilts over long time periods to now, is that it includes at least one period where it made no sense at that time to invest in long term gilts at all, namely the recent period where index linked gilts were priced with negative real yields down to as much as negative three percent, and conventional gilts were priced at that same time at very low nominal yields consistent with that negative real return on linkers.
If there are structural reasons for asset classes to be both priced for an individual investor offering no prospect of a reasonable return or over the same period any material diversification benefit, such as gilts have been until recently, then they should simply be avoided. And when gilts didn’t make sense, best buy savings accounts or money market funds were the obvious choice to replace the allocation to gilts that would otherwise be used. Certainly the need for institutional investors (such as pension funds) to hold gilts at seemingly all costs to meet their statutory requirements was a key structural factor involved here.
We are now in a position where gilts and conventional gilts are priced at a level where there is upside and downside potential for long term investment in them. And so it makes sense to invest in this asset class alongside best buy savings accounts or money market funds. Personally I am still avoiding conventional gilts because index linked gilts provide a better match for my spending needs (while inflation is priced in at about 3%pa) but that might be different for others.
I don’t know if that’s active management of bond exposure in the sense that I won’t be selling my index linked gilts if real yields go up or down a few per cent. But there is a subjective element to the allocation to gilts all the same, in that until recently my allocation was 0%. And even if we agree gilts should have been avoided but now look a reasonable choice, at what point did they become a reasonable choice again?
It is easy to take the concepts of passive investment too far. Not every asset class is priced at all times consistently for every investor, taking into account return, risk and diversification potential.
An interesting article although I have a few (highly technical) quibbles
1) ‘All Stocks’ is not an intermediate gilt fund (although it currently has a duration that can be considered intermediate) – it is a fund that includes every single gilt in issue that has a minimum amount outstanding. This means that the weighted maturity is dependent on the term structure of UK government debt (which is dependent on what gilts the treasury want to issue) as well as yields and coupons. I note that the modified duration of ‘all stocks’ peaked around 13 in 2020 (it started the decade in 2010 just under 10) since yields and coupons had been low for at least a decade, while back in the 1980s the duration was below 5 (I calculated gilt index returns and durations going back to 1870 at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4742450 with a link to the download in the paper). Anyone wanting to choose an intermediate gilt fund that will at least not stray into long durations could choose one with a fixed maturity range (e.g., ishares up to 10 year gilt index fund) or hold a combination of two funds (e.g., under 10 years and over 15 years) to obtain the desired ‘target’ duration. Rebalancing between the two is then a periodic activity.
2) IIRC, the returns for gilts in the JST database are for consols (i.e., very long duration) prior to 1962, 20 year gilts until 1980(?), then 15 year gilts, and since 2016(?) ‘all stocks’.
3) Have you modelled STMMF returns before the existence of the actual fund on returns for 3-month T-bills or the base rate? These are not necessarily the same as each other or the SONIA rate.
@TA (#10), @dearieme (#1)
I looked and found (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5166182) little difference over the long-term between bond funds and rolling bond ladders (i.e., where coupons and maturing bonds are reinvested, typically, at the highest maturity) for the same maturity range (e.g., where the fund and the ladder both held gilts in the range 0 to 5 years). Over 30-year rolling periods the difference in annualised returns varied from about -10 to 25 bp (+ve means ladder was best) depending on whether yields tended to rise or fall over the period. Outside a SIPP or ISA, there are capital gains advantages to individual gilts and no fund fee that will favour the ladder.
Of course, collapsing ladders (i.e., where coupons and maturing bonds are spent) are a different beast.
@Wannabe Retiree (#9). Fixed income duration (but not the overall allocation) is the one area where I allow myself a bit of activity for two reasons a) it satisfies my urge to tinker, and b) the effects are relatively small (so mistakes are unlikely to be costly).
@xeny
Thanks that was the article and it would seem they were right and the Short Term Fixed fund edged a win over the last year or so but with a smidge higher fee it might not be worth the worry.
The enhanced version has done better but the 2022 performance showed the risk compared to the other 2 funds
The 800 lbs gorilla that is the bond market is the biggest beast in the jungle.
Mightier than any politician, even the mad MAGA king on full caps.
I think markets are right to demand term premium.
Institutional bond buyers get that governments need to run a deficit when the economy is running below capacity (demand deficit).
But running structural deficits above the rate of nominal GDP growth (such that the deficit increases as a share of GDP) when the economy has less supply capacity than demand (inflation or stagflation) makes no sense, and, not unreasonably, is a no no.
In fairness, this is not party political, as both left and right are spending like a drunken sailor on shore leave.
Before 2020 the pattern seemed to be that governments underspent their way out of the GFC, which also made no sense as the credit destruction of reduced bank lending to the real economy more than covered the money creation of QE.
After COVID, with economies at full capacity and the price level rising, governments did the opposite, and failed to throttle back expenditure.
It’s like a driver under and then over steering.
@Snowman (#11)
As you say, institutional investors have to invest in gilts in order to satisfy their liability matching obligations. One outcome was that funding ratios became low and at least one open DB pension upped contribution rates to compensate. This was also the reason some were trying to juice their returns with the liability-driven investments that came undone during the Truss episode.
For the individual investor, constructing a collapsing ladder of inflation linked gilts even with real yields as low as -2% could still have marginally improved portfolio longevity (I modelled this at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4256534) – mainly by removing sequence of return and inflation risks early on in a retirement.
@Snowman, Alan S
I think you will find there were no statutory requirements for DB schemes to purchase said linkers. However, IMO they were coerced/bullied by the TPR who pushed [leveraged] LDI relentlessly and IMO entirely ill-advisedly too. AFAICT, some employer DB funds resisted the pressure, but not many. OOI, local government pension schemes by and large avoided the LDI scam!
@JPGR (#4):
I do not know how long ago you set up your ladder*, or if you are already drawing from it. Could you possibly say a few more words about this. And, if you are drawing from it, could you indicate how accurate your estimated income need has turned out to be in reality. The reason I ask is, in my case my estimated income need was very inaccurate and over just eight years that has had a not insignificant opportunity cost.
That is, the accuracy of your estimated income need [from the ladder] is IMO another crucial parameter (albeit rarely discussed) in this calculus.
*and whether it was set up in advance of pulling the plug or …
Thought provoking article@TA, thanks as always for keeping the site live!
I think it’s always worth looking at yields at the start of the period under examination when looking at investment returns on bonds.
Prices are inversely proportional to yields, and IF you hold to maturity you’ll earn that yield*. This is one point of difference between a portfolio of bonds and a Bond ETF – as the ETF manager may not hold the bonds to maturity.
I think you’ll find that if yields are high when you buy the bonds the subsequent investment performance will generally be better. Conversely if the starting yield is low you are picking up pennies in front of the steam-roller.
I don’t generally go for active management but blindly investing a passive allocation in bonds without knowing what yield you are locking into or what the implications are for your expected returns doesn’t seem the right answer either.
*to first order – there is still uncertainty around the reinvestment of coupons.
@Alan S – money market data mostly from Macrohistory:
1870-2015 Money market rate
2015-2020 BOE deposit rate
Thereafter I use the British Government Securities Database. I’ve updated the article citation because I forgot to credit money market returns to Macrohistory.
@Snowman and Prospector – I hear what you’re saying about the limits of passive investing. At the same time, a passive investor admits they don’t know better than the market. You may well decide that QE represents a market distortion that requires an active response. That could be analogous to finding yourself in the middle of a stock market mania and deciding it’s best to head for the door before everyone else.
But you can be wrong for a very long time. Perhaps if a pandemic hadn’t come along, bonds would still be bossing money market even over the last 5 years. We’ll never know.
But we definitely didn’t know that bonds were about to turn that ugly in 2022. You could guess it. People had been making that guess ever since 2008. They were wrong for 14 years.
Meanwhile, if I’d held money markets instead of gilts for 15 years then I’d be worse off. Maybe I’d be better off if I made the right guesses ahead of time but I’m not confident in my ability to do that. As the old saying goes, “you have to be right twice when trying to beat the market.”
@Al Cam (#17)
Happy to share my story. In brief:
I retired on 31 March 2024, so about 14 months ago. I had been looking at index linked gilts for a good number of years but couldn’t bring myself to invest at negative real yields (I’m risk averse and came close).
The Truss/Kwarteng Budget and consequent increase in real yields gave me my opportunity. Over the next 18 months or so I rebalanced my portfolio away from equities and cash, and into index linked gilts (leaving c.30% in equities, some cash and some gold(!)). I will be drawing on my ladder at the beginning of next year (2026) I would expect.
My estimated day to day expenses have been reasonably accurate, although predictably I’ve spent a bit more than expected. I set aside a separate pot for early years travelling – and that is certainly being depleted more speedily than expected. (Wow – travelling is expensive these days.)
So, all in all, not entirely sure how smart this has been, and how it will work out in the long run, but seems to be working reasonably well 14 months in.
Happy to answer any further queries as best I can.
@TA (#10)
I take what you say about this not being the first that UK government finances were in poor shape! But today they seem in particularly poor shape (debt relative to GDP is at its highest since the 60s – when it was at least falling – and debt interest accounts for a historically high % of public spending.) And in the past we had economic growth and a younger demographic.
On the bond market’s assessment of UK public finances – I guess the fact that the same market five years ago judged it appropriate to lend to the UK government at negative real rates makes me wonder about its wisdom. Or whether there remain distortions like the “artificial” demand from pension funds that others have mentioned. As your article shows, the real achieved yield over the very long term has been less than 1%. Does that seem like adequate compensation for the risks of lending to the UK government?
@TA #19 you are absolutely right about it taking a long time. I experienced this first hand having taken out a 10-year fixed rate mortgage in 2010 expecting BoE base rates to revert to “normal” in a few years. As it was I ended up paying way more interest than if I’d gone with a tracker. And a fixed rate mortgage is economically being short bonds.
On the flip side I ended up with a way higher equity allocation than was wise as I felt TINA to equities. Thus participating in a bull market that continues. At least with the normalisation of rates I’m able to now switch into bonds with a clear conscience. But I recognise I’ve been lucky and had we had “the Big one” and equities had crashed in that period I would likely be sitting 5 years further away from FIRE.
@JPGR (#20):
Thank you very much for your additional details – very illuminating. Real UK data/experiences over the initial retired period is quite rare and IMO is far more helpful than theoretical spreadsheet-like calculations.
Nice to hear from somebody who can estimate their retired spending [and/or drawdown needs] fairly accurately. With the benefit of 20/20 hindsight, my estimate was pretty woeful! I think my biggest lesson is that if your estimates are by design conservative you run the risk of the errors combining such that overall you end up a long way off*. It could, of course, go the other way – nobody knows the exact path in advance! Are there any estimating tips/pointers you wish to share?
My estimated opportunity cost is based on funds not being largely invested in equities. I note you still have some 30% in equities**. OOI, do you recall roughly what your portfolio allocation was prior to re-shaping?
Lastly, I note your comment about 14 months vs the long run. FWIW I had identified my likely pathology within that period; although it took me a good few years to really believe what was happening. My notes at the end of year 1 state “an exceptional start, that is very unlikely to be repeated in subsequent years”.
All of which is just my experience and YMMV.
Thanks again.
*Which could diverge more the longer the period under consideration. IMO twenty years is a long time – having said that an Annuity is for your lifetime (at least) and irreversible.
**which have done very well over the period under consideration, and may not, of course, continue to do so
@CGT101
In answer to your question I would say they don’t and as a result am looking very closely at High Yield / Junk bonds. They are relatively short duration and offer a decent yield (both helping with unexpected inflation). Sure there is downside risk in a recession but this tends to be more contained than equities and they have historically come back decently – note they won’t give any gains like a Gilt could/should in a downturn. Would be interested to hear what others think.
@CGT101 – I do agree there are reasons to be concerned and the debt level is high. But it’s been higher, much higher in the past.
Your point about secular stagnation and demographic overhang is also well made. But there are other countries with worse demographics apparently handling larger debt burdens. I don’t think we’re in great shape but we’re no basket case either. If the market believed we were in danger of defaulting then gilt yields would be much higher.
I do think we’re in a tight spot. I don’t think this time is different. (As in I don’t think gilts are too risky now.)
When I started investing, the expected real yield of high-quality government bonds was around 1%. Everyone’s not just the UK’s. We’re a bit under that, which isn’t great. But some of that may be an artefact of our RPI history – we were late to CPI.
Essentially, 1% real yield is about the going rate for a government-backed asset that diversifies an equity portfolio against major demand-shock recessions. If you want / need that, then bonds it is.
Some savvy commenters on here (including @Snowman) have noted that we can buy linkers for a real yield of 1%+. That does seem like a good idea to me, especially if you’re seasoned enough not to worry so much about the crash protection.
FWIW, I’ve shifted my view (and money) over the years to a much more diversified defensive allocation.
Hope this doesn’t sound argumentative. I’m enjoying the debate 🙂
@Prospector – Absolutely. I think we all have our share of wins and losses along the way. I chose a tracker mortgage on the eve of the Financial Crisis because I had a sense from the news there was “Trouble at t’mill.” But I had no idea 14-years of historically low interest rates were ahead. I got very lucky there. OTOH, I’ve made so many mistakes over the years I shouldn’t be allowed out by myself 🙂
I agree that bonds now look better value than they did.
Thinking back, the bond advice during the ZIRP era (such as it was) was to go short but not to dump ’em. If you didn’t want to go 100% equities then you just had to suck it up.
@Chuckie B – Junk bonds are not a defensive asset class. As you note, they go kaput at the wrong time. So they belong in the equity side of your portfolio. They don’t solve the problem of curtailing portfolio losses when stocks meltdown.
@TA doesn’t sound argumentative at all, at least not unreasonably so! I don’t have settled views on this topic, it’s incredibly useful to explore the different points of view with people who have thought about it.
@TA
Sorry I guess I’m moving away from the safe assets discussion which is what the article is about, but following the theme of are government bonds a good investment.
I would suggest that when you look at the thin real returns from Gilts in a falling interest rate environment and then you compare it with what we face today – I believe there is a significant risk of capital erosion through financial repression. Hence am looking at fixed income alternatives that may not lose me money in real terms over the long term – recognising the need to accept more volatility and them not giving downside protection.
@Al Cam (#23)
Candidly I’m not sufficiently confident in my predicting abilities to share any particular insights. If you want to make God laugh, show him your plans……
Your other question: my portfolio was, roughly, 60% equities, 35% cash/money markets/corporate bonds, 5% gold. And I moved to, roughly, 30% equities, 60% index linked gilts (ladder), 5% cash/money markets and 5% gold. I did it over 18 months (so quite quick) and accepted the tax leakage. My fear in retirement is inflation (not deflation) and I’m hoping this will protect me, albeit not perfectly. I’m content to forgo the expected greater returns that a larger exposure to equities would provide (I don’t like the thought of equity volatility in retirement).
@JPGR:
Thanks.
I assume you factored state pension entitlement(s) into your ladder – which may ameliorate some of your current ” long end” paper [MTM] losses*. OOI, do you have any DB pension entitlements?
Re: “I don’t like the thought of equity volatility in retirement”:
I guess MTM bond volatility is easier to live with provided they are held to maturity with no defaults en route. This post (and in particular its comments) might be of some interest: https://henrytapper.com/2025/05/27/what-the-hell-is-going-on-with-long-dated-gilts/
*IMO, MTM valuation is a whole other topic.
@ChuckieB – cool. So following that track i.e. fixed income assets come with their own risks attached, and to avoid those we’re prepared to accept volatility and downside risk… I’m wondering why the answer to that isn’t equities? As in government bonds diversify the risks of equities and equities diversify the risk of government bonds.
Although, looking back at your first comment I guess your thinking is that junk bonds add an other layer of diversification that’s different from either govies and equities?
For a while, Emerging Market $-denominated sovereign debt was a great example of your high yield bond thesis. Been off the boil for at least a decade now IIRC. I’m interested though in anything that genuinely diversifies a portfolio and is implementable.
Have you come across any long-term studies that support the case for junk bonds? My concern is that ultimately they lower long-term returns without improving risk-adjusted returns.
@ Al Cam (#29)
I did factor in state pension entitlement. I have a slightly unusual workplace arrangement, which provides for payments for the first 10 years of retirement only.
I intend and, famous last words, expect to hold my ILGs to maturity so have (just about) rationalised away concerns about their volatility. Thank you for the link.
@JPGR – your comment made me smile in recognition. Every time I catch sight of my ILGs, I’m also rationalising away the (paper!) losses because I know, er think, I’ll hold them until maturity.
I find the defensive part of my portfolio the most awkward. And I for one am getting twitchy about the vast debt/GDP ratios of the major economies. If you buy a straight-down-the-line G7 govt ETF such as iShares Global Government Bond (hedged to GDP) (IGLH) you’re loading up on the debt of hugely indebted countries. Is this really wise? As an alternative there is iShares Global AAA-AA Government Bond (SAAA) which in effect elevates the AAA debtor to 20%, and caps the AA at a max of 10%. And so (roughly) you get:
Germany 20%
USA 10%
France 10%
UK 10%
Canada 10%
10 or so other AA nations – 40%
You’ll notice in particular the total absence of Japan and Italy, and the comparably small exposure to the US – plus diversification through the inclusion of AA rated nations beyond the G7. Alas it’s unhedged so you’re introducing currency risk – so not necessarily what you want for the steady-eddie part of your portfolio.
It’s all about trade-offs I guess.
@JPGR (#31):
That is indeed an unusual arrangement – and, as a minimum, probably blunts the amelioration I was thinking of when your state pension(s) kick in. An interesting situation that I have never seen before – but every day is a school day, even as we age – which IMO is no bad thing.
OOI, do you currently consider this workplace arrangement (and the SP in due course) as income*? I ask because the MTM approach is almost** entirely responsible for the wildly differing*** cash equivalent transfer values (CETV) offered for the same DB pension today and say just five years ago. What is even more odd is: that whilst seemingly the DB schemes today are in a better position to pay out CETV’s (that are at the lowest they have been for years); and when CETV’s were at their highest then schemes in general were least able**** to pay them. As I said before, MTM is a whole other topic!
* ie not MTM
** the passing of time is also involved
*** perhaps 2 : 1
**** on a common fundedness basis
@TA (#32):
Almost an article of faith then?
As reported here many times before my admittedly limited experience is that things change and will almost certainly keep changing too. And that includes changing for the better too!
@TA (#30), @Chuckie B, @Mousecatcher007
Might be worth noting that Life Insurers (LI) prefer corporate debt and other instruments. AIUI more often than not on receipt of a bulk transferred DB scheme the LI immediately offloads most of the gilts* and re-risks! They have been at this game for years and do know a thing or two; as well as how to make money!
FWIW, IMO Alan S’s earlier comment (#15) about gilts and negative returns only works for an individual (vs a DB fund) as no leverage is/was involved.
Principally, this is because individuals cannot actually replicate a leveraged LDI approach. Why have I raised this point? The reason is I suspect it is at least open to question whether a punter could ever successfully replicate a LI’s approach – but IMO it is definitely worth bearing in mind that LI’s do not love gilts, in spite of all the many myths to the contrary!
*including linkers too
@Al Cam (#34)
I do consider the workplace arrangement as income.
Thanks @TA – good article and good comments
Excellent bond chat on Merryn Talks Money this week which is worth a listen in this context: https://podcasts.apple.com/kz/podcast/japans-bond-market-in-focus/id1654809850?i=1000710310582
@JPGR (#36):
Perhaps that is the way to look at your ladder; and therefore “unhook” yourself from all the MTM volatility. I no longer think about my DB in CETV terms*. In part however, that is because the option to transfer to cash lapses once the DB comes into payment.
Another possible psychological benefit of this approach is that movements in your portfolio are more instantly meaningful to you?
Tricky, and what would you consider your portfolio to consist of if you had bought an Annuity?
*In fairness I only ever really thought of my DB in those terms when looking at things like net worth. I did consider cashing it in but always decided against it. Had a partial transfer been available I may well have gone for that, but, in my case, that was not ever an option. Interestingly, had I transferred at or near the CETV peak and conservatively invested the proceeds I could probably today buy an Annuity similar to my DB and still be quids in – but that is life!
@Mousecatcher – really interesting find, thank you! I guess if it catches on then hedged versions will follow. Very much the bond equivalent of the World ex-USA ETF that hit the market recently.
@Al Cam – more instinctive genuflection to JPGR’s “Man plans, God laughs…” clause 🙂
Why is it do you think that life insurers prefer corporate debt?
It’s also worth noting that a recurring investor mistake is to ape institutional strategies that aren’t aligned to their individual objectives, time horizon, or capabilities. The Yale Model comes to mind.
@Andy J – thank you! Japan is such an interesting contrast to our own situation.
@ TA
Answering your last question first, I don’t I’m afraid which might tell us something 🙂 however…
My starting point is that I don’t think Governments will honour their debt obligations and will do this through stealth inflation destroying the value of long term and probably intermediate government bonds. As inflation rises it will lead to a revaluation of equity markets which either causes a slump or a flat real performance for a lengthy period as per US in 1970s/80s (think that’s the right decades). So in order to get some sort of meaningful return I am looking at HYB which as you know can be considered a bit debt and a bit equity, but generate a steady return that should be above inflation (as long as they don’t default!).
Eyeballing the more recent performance of a HY fund (Aegon HY) this seems to have outperformed the lifestrategy 60% equity but has had similar/less volatility (of course noting difficult bond period recently). I have also listened to some recent Howard Marks podcasts which may be influencing my thoughts!
Would welcome any observations/critiques of the thought process noting noone knows the future.
@TA (#39):
Not totally sure, but take a look at the last slide in the annex to this presentation (Chart 7) to get an idea of the level of preference https://www.bankofengland.co.uk/-/media/boe/files/speech/2022/september/charlotte-gerken-whos-concentrating-speech-annex.pdf
This incidentally accords well with Ned Cazalets When Im Sixty Four paper of some ten years ago.
Agree that aping is rarely a good idea unless it is exactly the same problem being solved. There is IMO a very good paper by Blanchet et al called LDI Misapplied from 2017 (that unfortunately has disappeared from the net) that IMO provides some good food for thought for individuals along those lines.
Thanks for this – had me checking my portfolio to reveal that I have 20% more in MMFs than Bond ETFs but I also have a bit in Treasury bills – where do these fit in? A bit like MMFs, as they too are short term?
@ChuckieB – that’s interesting. And I share your concern that repeated bouts of inflation could hurt govies as per the interest rate cycle approx mid 1930s to end of 1970s.
My personal response has been to increasingly diversify beyond nominal bonds into index-linked bonds, gold, commodities, and cash. The drawback is I’ve less protection than I used to in the face of a good old-fashioned Great Recession type crash. Unless gold keeps coming good, which I’m wary of relying upon.
Your thesis that this requires greater diversification on the growth side is worth thinking about. The 1970s were terrible for equities and the 1960s were so-so. 1980s on OTOH – best decade ever!
If a repeat of 1970s stagflation were the issue then I’d think myself covered by commodities and gold.
For junk bonds to come good then I think I have to buy when credit spreads are high i.e. I can buy on sale cos everyone thinks dodgy companies are gonna go to the wall. But then in actuality, things turn out OK and casualties among the infirm and highly leveraged aren’t so bad.
There are three things I’d to check out to stand-up the thesis:
-What’s the track record of junk bonds during bouts of high unexpected inflation?
-What’s the track record of junk bonds when equities underperform?
-What’s the overall long-term record of junk bonds? From what I’ve read, it’s not great. But I’ve not dived deep into the academic literature nor checked the numbers for myself. Except re: emerging market debt as mentioned. That definitely was an opportunity once, but possibly not one due to repeat.
My final concern is that junk bonds may require market chops and know-how that I personally don’t have. As in, I prefer set-and-forget investments that I can periodically rebalance. My sense is that junk bonds aren’t that. But I could be (a) wrong and (b) you might love a more hands-on approach.
Those are my instant thoughts. If you get a chance, I’d be interested to hear a summary of Howard Marks’ thesis.
The one thing I’d add is that many people have talked about governments inflating away debt since QE became all the rage in 2008. I don’t think ensuing events have supported that view.
The inflationary surge was caused by post-Covid bottlenecks plus an understandably large government stimulus to reboot the economy. Central Banks did then move to tame inflation with interest rate rises. Too late for sure, but as I understand it monetary policy isn’t a precise science. Incumbent administrations around the world copped the blame for that and went down like nine-pins – even though their Central Banks are independent.
It doesn’t seem to me that setting inflation loose is in the interests of government. It also looks like large debt-to-GDP ratios are giving the bond markets the whip hand over elected administrations.
@Al Cam – my guess would be that as a life insurer is an immortal (ideally) then they can afford to take on more risk and stretch for yield, so long as they can pay their ongoing liabilities with a nice feather bed of gilts.
From an individual perspective, anyone who doesn’t really need their portfolio to pay ongoing bills, could do the same thing. We could think of this as the upside portfolio. Anyone who absolutely needs their portfolio to form part of their floor, would likely be better advised to diversify with a healthy chunk of minimal risk assets.
@weenie – Hi! 3-month Treasury bills and MMF are essentially interchangeable. From an asset allocation perspective, they’re fulfilling the same role.
.
@TA (43): On governments inflating away debt…. It’s true that we didn’t see inflation taking off in the 2010s, even with QE. But in the UK at least there was at the same time as QE an offsetting programme of fiscal tightening (Osborne’s austerity). But the current UK administration’s fiscal instincts are not at all the same. For me the verdict is out on how pliant the central bank is to the government’s needs – its record in recent years has not been good.
I don’t think the government is going to want runaway inflation, given the likely political fallout. But it may be happy with enough inflation to be painful for bondholders. Recent inflation has been politically toxic but also a giant fiscal gift to the government, because of the effect of fiscal drag.
We should throw political instability into the mix too, and its potential to impact bond markets. We have an ageing population in a stagnant economy, but no political party is going to propose another bout of austerity if it wants to be elected. We seem destined to cycle through governments that find it impossible to deliver what the electorate wants until – well, I’m not sure what ultimately breaks that cycle this time round. But it might not be great for gilts!
“I don’t think the government is going to want runaway inflation”: Never attribute to design/malice that which can be adequately explained solely by incompetence/neglect.
High inflation will reduce the real value of the overall stock of debt (private and public), but it will simultaneously increase the defecit, as the cost of providing public services rises rapidly under inflation whilst the profit share of the economy and real value of wages (the tax base) shrinks. Can’t see that as an attractive policy choice.
The problem though for narratives about what *could* happen is that we don’t ever know the shape or structure of the ‘possibility landscape’.
We just don’t know how many branches there are, nor what weight to assign each scenario.
So we can’t actually ever assess the likelihood of a soft default – before or after the fact.
It seems plausible before the fact that there could be a soft default, but that alone doesn’t make it probable.
And if it does happen, then it does not by itself tell us if, to what extent, and in what circumstances the previously generally low to negative correlation of bonds to stocks will or might hold in the future (if at all) – i.e. bonds might still provide risk off protection to rebalance into equities (or they might not) whether or not rates/yields rise.
At least bonds do offer some nominal return now for risk.
During ZIRP they offered return free risk.
Maybe they still don’t offer enough return for the risk, or maybe they do; but at least there is now some sort of a non negible return actually available. This isn’t 2009-21 anymore.
And from a sentiment perspective, the loathing towards long bonds (especially the longest linkers) might just perhaps be a contrarian indicator.
@Al Cam (#35)
For those interested, the article at https://www.wtwco.com/en-gb/insights/2017/07/how-does-a-bulk-annuity-provider-invest contains some interesting information about the difference between DB and annuity (and bought out DB) investments. But large DB pensions also invest in a wide range of assets – e.g., USS (https://www.uss.co.uk/how-we-invest) has about 42% of their DB fund invested in liability matching, 62% in growth (commodities, infrastructure, equities), 27% in credit and -30% leverage.
And yes, my comment about collapsing linker ladder with negative yields definitely only applies to individual investors.
@TA (#39)
My understanding (possibly faulty) is that corporate bonds have tended to have higher returns than government bonds since they are riskier and, possibly (anyone know?), large institutions may be first in the queue when a bond defaults.
@TA (#43)
Rob Dix’s book (The Price of Money – it has been in the weekend reading section) appears to suggest that directly held property forms a useful diversifier that is largely inflation linked (rents and capital gains). It would be interesting (but difficult) to compare property against holding linkers over the last 40 years or so.
There are a some ‘high yield’ bond funds that go back to 2007 (e.g., Baillie Gifford High Yield Bond Fund currently has 70% of its portfolio rated at B or below) that could be tested (spoiler alert – the NAV dropped in 2008).
@TA (#43) @ChuckieB (#44)
The ability of the UK Government to inflation away debt has been reduced by the presence of linkers – particularly as the proportion of those is now quite high (in fact, one of the aims of their introduction was to give the gilt markets confidence that inflation wouldn’t be used as a tool to reduce debt value – see https://oro.open.ac.uk/60077/8/60077.pdf for an interesting, YMMV, account of their introduction).
@TA (#43):
I do not know in detail how any Life Insurer (LI) does this; so what follows is largely speculative. Although I understand all [UK] LI’s are subject to the same rules and regs, I suspect they do not all do it the same way. That is, there may well be “tricks of the trade” and proprietary strategies in play. Perhaps this is why it is so hard to establish any one LI’s approach – which may develop over time too.
Having said that, IMO it is perfectly plausible that a LI may time segment or somehow risk* phase its liabilities to reach for yield. However, the BoE pie chart is a [UK] industry-wide view and the segment called “Sovereigns and Quasi Government” contains things other than gilts. And, it can be deduced from Ned C’s report (page 65), some LI’s hold few, if any, gilts (note what Ned labels gilts in his individual LI company pie charts includes non gilts).
Perhaps, one day this may all become clear**. However, based on the rate of progress of my understanding (to date) this could still be some time away!
*AIUI, LI’s have access to many more asset types and risk management tools and techniques than a man in the street – IMO the BoE pie chart gives a flavour of this
**what has been clear for at least a decade now is that LI’s do not only use gilts – although that myth persists widely
@Alan S (#47):
I missed your comment before my latest post (#48).
Looks like an interesting link – which I will follow up.
Perhaps you already know this, but USS is an outlier and one of the few DB schemes who resisted (to an extent) the TPR’s onslaught. They have also been at the forefront of trying to push back against: MTM valuations, proposed employer changes to employee contributions, etc.
Typically, over the last 20 years or so DB schemes have (as cajoled/coerced (take your pick) by the TPR) largely bailed out of growth assets and piled into gilts using leverage to shore up their funding shortfalls. So IMO what they did under the guise of de-risking was actually re-risking!
P.S.
FWIW, Purple Book 2024* shows around 15% of DB scheme assets are now held in equities – ie USS is very much an outlier!
*also known as “The Pensions Universe Risk Profile” by the PPF and now in its 19th edition.
@Alan S,
Interesting WTW* paper – but it is pretty high level and AFAICT dates from 2017**. The most interesting snippet I picked up was that WTW reckon an LI is more likely to use gilts for long end liabilities rather than the short end.
*formerly known as Willis Towers Watson, who I first came across as actuaries but in reality they are much much more than just that, their wikipedia entry seems pretty thorough
**at that time WTW chose to highlight the differing legal requirements facing DB schemes and LI’s; however, the TPR low dependency fundedness regime is actually a higher bar than that the LI’s are legally obliged to operate to
@Al Cam #48 [#warning long post and tangential to MMA vs bonds debate though very relevant to how UK institutional investors and Life Insurers(LI) in particular view bonds.#]
UK LIs have a favourable regulatory regime (compared with pension schemes) thanks to the matching adjustment (MA).
The MA avoids some of the pitfalls of mark to market accounting. The way it works is that higher expected returns on bonds and other fixed income assets result in a higher discount rate for annuity liabilities. Higher liability discount rate means lower value of liabilities. Thus a UK LI applying MA may report a better funding position than a pension fund with equivalent bond investment strategy.
The MA also helps if there is a fall in value of credit risky bonds relative to government bond (where the “spread” or yield in excess of the yield on government bond of similar duration widens) by providing an offset on the liabilities.
The argument made why the MA is acceptable is that a) as long as the insurer holds bonds to maturity, b) makes a suitable allowance for defaults and downgrades and c) matches the profile of interest and redemptions to the expected annuity outgo (thus avoiding being a forced seller), then the insurer is only exposed to the extent actual downgrade and default losses are worse than the allowance in b).
UK Insurers are required to hold additional regulatory capital (in action to assets backing the annuity liabilities) against this risk (of default and downgrades).
There are critics of the MA eg see
https://eumaeus.org/wordp/index.php/2019/11/19/matching-adjustment-on-trial/
The MA allows firms to take credit for investment return that has not yet been earned, and credit is a very “long tail” risk. You have a high probability of making a small return each year and a small probability of a recession leading to a load of default losses. This makes determining b) very subjective. Bond defaults are a bit like London buses, you’ll have several years without seeing any and all of a sudden there will several in short succession.
Perhaps in part to silence some of the critics the UK regulator is for the first time going to publicly publish the results of the LI stress test (LIST).
For the nerdy LIST will be an interesting insight into the makeup of LI investment portfolios backing annuity liabilities and how they respond under stress.
https://www.bankofengland.co.uk/prudential-regulation/publication/2024/july/list-2025
@Prospector (#52):
I had got as far as realising the MA was important; but not much farther. There are clues aplenty – but I could find little useful exposition, as most stuff is rather technical. So a very helpful (if tangential – which is down to me) input -for me at least
Thanks.
@ accumulator. This is probably a (very) stupid question but both these ETFs are total return funds (ie not paying out dividends)? Reason i ask is i couldn’t find the i shares intermediate bond ETF except in distributing form.
@TA
To try and answer your questions:
1) inflation – they’ve not been around that long so probably best to look at the recent bout of inflation we had. See the chart halfway down this page: https://www.allianzgi.com/en/insights/outlook-and-commentary/forget-the-high-yield-myths
A decent hit but full recovery within 2 years.
2) vs equities – will fall in line with equities but not as much….aegon high yield fund was created in 2002 and FT charts have this data so you can see the 2008 crash, while down a lot it came back faster than equities.
3) long term record – see aegon chart or one referenced in link above
4) don’t have a hands on approach but propose to trust fund managers (!) and high diversification in funds…
Howard Marks was basically saying that equities are overvalued and you can get equity like returns from HYBs because the timing was good, very much following the argument in this link : https://www.alliancebernstein.com/apac/en/institutions/insights/investment-insights/high-yield-bonds-an-antidote-to-volatility.html
I don’t have anything further on this in terms of research papers; was my own thoughts and found these links through Google just now. I think where I may be taking a more relaxed view on the volatility to you is because I am thinking of living off the yield of my portfolio which gives me a different perspective. Hope this has been helpful.
@alan s
Thanks for this, will take a look but expect to remain sceptical. Just don’t trust politicians.
@Alan S – All good pointers, thank you. And you make a great point about linker debt as an indicator of government commitment to low inflation.
@Adrian – the results are total return not the funds i.e. the performance figures show the return if dividends/interest are reinvested in the fund rather than spent. That lets me use IGLT’s long track record for the comparison but without worrying about the acc/dis distinction. If you’re looking for a similar acc gilt ETF then VGVA is worth a shufty. Some of the gilt funds on this list may come in acc form, too. I can’t remember off the top of my head:
https://monevator.com/best-bond-funds/
@Chuckie B – Thank you! I’ll follow that up. Intriguingly, I compared the Aegon High Yield Bond Fund GBP B Acc (based on your tip above) against the MSCI World index here:
https://www.morningstar.co.uk/uk/funds/snapshot/snapshot.aspx?id=F0GBR04NHL&tab=13
The high-yield bond fund is slightly behind the MSCI World right now but it’s essentially neck and neck when the differential is averaged across the 23 year timeframe.
Eyeballing the chart, I can see it’s behaving exactly as you say – drawdowns highly correlated with equities but typically a little less severe. Interesting!
Also, I’m comparing it to the World index rather than an actual fee-charging fund. (This is just a quick and dirty comparison so I didn’t take the time to find a World fund that was live in 2002).
Definitely looks worth further investigation. Really interesting. Cheers!
@TA, @Prospector, @Alan S, @TI
Re my comment: “I do not know in detail how any Life Insurer (LI) does this”
My understanding is still very far from complete, but I have made some steps forward over this weekend. This is in no small part due to the input at #52; big H/T @Prospector.
I found this pretty helpful and thought you might too: https://www.bankofengland.co.uk/speech/2022/july/sam-woods-speech-given-at-the-bank-of-england-solvency-ii-striking-the-balance
It is a little dated – and it might take a while to read it through (at c. 10 pages) – but it should be time well spent. It explains what is going on, why it matters, and gives numbers too – some of which IMO are pretty scary/surprising.
Spoiler alert: the incentives seemingly driving LI’s are not necessarily along the lines we thought they could/should be.
I almost hesitate to pen the next few words: but @TI might like to take a look too – as a lot of what is going on here is (in some sense) only possible due to Bxxxxx